Institutional Investors in Global Capital Markets 9781780522425, 1780522428

Table of contents :
001View_ Institutional Investors in Global Capital Markets, as a PDF.Institutional_Investors......Page 1
International finance review......Page 2
003View_ International Finance Review, as a PDF.International_Finance......Page 3
List of contributors......Page 4
bm_......Page 7
Outline placeholder......Page 0
Part I: Introduction to institutional investors in global capital markets......Page 8
Part II: Institutional investors: Their economic and financial impact......Page 10
Part III: Investment preferences of institutional investors......Page 12
Part IV: The benefits of Sovereign wealth fund investments......Page 13
Part V: Sovereign wealth funds: Do political objectives drive their asset allocationquest......Page 14
References......Page 17
Financial liberalization and foreign institutional investors: Literature review......Page 19
Introduction......Page 20
Financial liberalization and international capital flows......Page 21
Liberalization and Market Activity......Page 27
Liberalization and Home Bias in Portfolio Investment......Page 29
Are Foreign Investors Momentum or Contrarian Investors?......Page 32
Do Foreign Investors Destabilize Developing Markets?......Page 35
Target Firm Characteristics......Page 37
Market Characteristics......Page 39
Foreign versus Domestic Institutional Investors......Page 41
The role of foreign institutional investors......Page 43
Conclusion......Page 47
Notes......Page 49
References......Page 50
Institutional investment horizon and firm credit ratings......Page 53
Introduction......Page 54
Related literature and hypothesis......Page 57
Firm Credit Ratings......Page 60
Institutional Investors’ Horizon......Page 61
Research Design......Page 62
Univariate Tests......Page 63
Regression Analysis......Page 64
Robustness Tests......Page 70
Conclusion......Page 78
Notes......Page 79
References......Page 80
Divestment of foreign strategic investment in China’s banking sector: Causes and consequences......Page 85
Introduction......Page 86
Previous Studies......Page 88
Why Do Foreign Financial Institutions Divest in Chinaquest......Page 89
Business cooperation......Page 100
Stock market reactions to the divestment announcement......Page 103
Conclusions......Page 107
Notes......Page 108
References......Page 110
Domestic and foreign institutional investor behavior in China: Financial characteristics and corporate governance......Page 113
Introduction......Page 114
Preference of Foreign Investors......Page 118
QFII: A Formal Classification and Recent Developments......Page 119
Empirical Evidence on QFII......Page 125
Industry Characteristics......Page 126
Financial Characteristics......Page 127
Corporate Governance Characteristics......Page 129
Industrial Distribution......Page 130
Foreign and Domestic Funds: An Empirical Comparison......Page 134
Conclusions......Page 138
Notes......Page 139
References......Page 141
Institutional investors’ participation in foreign firms: Evidence from ADRs......Page 144
Introduction......Page 145
Firm-Level Determinants......Page 148
Country-Level Determinants......Page 150
Firm-Level Variables......Page 151
Country-Level Variables......Page 153
Univariate Analysis Results......Page 157
Multivariate Analysis Results......Page 162
Notes......Page 165
References......Page 166
Do foreign institutional investors (FIIs) exhibit herding and positive feedback trading in Indian stock marketsquest......Page 168
Background......Page 169
Literature review......Page 170
Indian Evidence......Page 174
Analysis of Herding......Page 175
Analysis of Positive Feedback Trading......Page 176
Herding......Page 177
Herding during Bull Phase of the Market......Page 178
Pratio......Page 179
Pratio during Bear Phase......Page 180
Nratio during Bear Phase......Page 181
References......Page 182
Do financial conglomerates have an incentive to prevent managers of other firms from pursuing their own interestquest......Page 185
Introduction......Page 186
Assumptions and Sequence of Events......Page 190
The Performance of the Mutual Fund......Page 193
The Behavior of the Bank......Page 194
Equilibrium Conditions......Page 196
Concluding remarks......Page 198
Notes......Page 199
References......Page 200
The impact of foreign government investments: Sovereign wealth fund investments in the United States......Page 203
Introduction......Page 204
Data description......Page 209
Stated Objectives and Investment Characteristics......Page 211
What Attracts SWF Investmentsquest......Page 213
Univariate Analysis......Page 215
Market reaction around the 13D filing......Page 218
Explaining the cross-sectional differences in market reaction......Page 220
Temporary Price Impact......Page 222
Long-Run Impact of SWF Investments......Page 224
Firm Performance......Page 226
SWFs Investments’ Effect on Tobin’s q......Page 228
Notes......Page 231
Acknowledgments......Page 233
References......Page 234
What do sovereign wealth funds imply for financial stability?......Page 240
Introduction......Page 241
Literature review......Page 244
Data......Page 247
Methodology......Page 249
Empirical results......Page 252
Notes......Page 255
References......Page 256
Africa’s quest for development: Can sovereign wealth funds helpquest......Page 258
African SWFs Motives......Page 262
African SWFs Size......Page 264
African SWFs Governance Structures......Page 265
African SWFs Investments......Page 266
African SWFs Reputation......Page 267
What are foreign SWFs doing in Africaquest......Page 270
What are the benefits of SWFs for African economiesquest......Page 271
What are the challenges facing SWFs INVESTMENTS in Africaquest......Page 276
Concluding remarks......Page 277
Notes......Page 279
References......Page 280
Subsidiaries of the Libyan Investment Authority......Page 282
Linaburg-Maduell Transparency Index......Page 284
Portfolio allocation for sovereign wealth funds in the shadow of commodity-based national wealth......Page 286
Introduction......Page 287
Sovereign wealth funds as investors......Page 289
Sovereign wealth funds as national guardians......Page 291
Methodology and data......Page 294
The results......Page 297
The risk-adjusted return results......Page 298
Conclusion......Page 301
References......Page 302
- List of Total Return Assets......Page 305
Are sovereign wealth funds politically biased? A comparison with other institutional investors......Page 306
Introduction......Page 307
Literature review and stylized facts......Page 309
Defining a Benchmark for SWFs......Page 311
Toward Better Regulation......Page 312
Portfolio Characteristics......Page 314
Geographical Distribution......Page 316
Sector and Industry Distribution......Page 318
Investment in OECD and non-OECD Countries......Page 319
SWF investments: The Political Dimension......Page 322
Internal Governance and Investment Strategy......Page 323
Political Regimes......Page 324
Conclusions......Page 328
Notes......Page 330
References......Page 333
ANNEXES......Page 337
Truman’s Investment Indicators for Commodity and Noncommodity SWFs, 2008 (Index 0-1)......Page 338
Sovereign wealth fund acquisitions: a comparative analysis with mutual funds......Page 347
Introduction......Page 348
Definition and types of sovereign wealth funds......Page 349
The growth of SWFs......Page 352
Firm Size......Page 354
Liquidity......Page 355
Operating Performance......Page 356
Stock Price......Page 357
Dividend Yield......Page 358
The Level of Investor Protection in the Country......Page 359
The Level of the Financial Development of the Host Country......Page 360
Religious and Cultural Factors......Page 361
Firm Visibility......Page 362
Sample and descriptive statistics......Page 363
Descriptive Statistics of the Sample......Page 364
Description of Variables......Page 367
Univariate analysis......Page 370
Conclusion......Page 373
References......Page 374

Citation preview

INSTITUTIONAL INVESTORS IN GLOBAL CAPITAL MARKETS

INTERNATIONAL FINANCE REVIEW Series Editor: J. Jay Choi International Finance Review is an annual book series in the international finance area (broadly defined). The IFR will publish theoretical, empirical, institutional, or policy-oriented articles on multinational financial management and strategies, global corporate governance and risk management, global capital markets and investments, emerging market finance, international financial economics, or related issues. Each volume generally will have a particular theme. Those interested in contributing an article or editing a volume should contact the series editor (J. Jay Choi, Temple University, [email protected]). Volume 1:

Asian Financial Crisis: Financial, Structural and International Dimensions, edited by J. Choi, Elsevier 2000

Volume 2:

European Monetary Union and Capital Markets, edited by J. Choi and J. Wrase, Elsevier 2001

Volume 3:

Global Risk Management: Financial, Operational, and Insurance Strategies, edited by J. Choi and M. Powers, Elsevier 2002

Volume 4:

The Japanese Finance: Corporate Finance and Capital Markets in Changing Japan, edited by J. Choi and T. Hiraki, Elsevier 2003

Volume 5:

Latin American Financial Markets: Developments in Financial Innovations, edited by Harvey Arbela´ez and Reid W. Click, Elsevier 2004

Volume 6:

Emerging European Financial Markets: Independence and Integration PostEnlargement, edited by Jonathan A. Batten and Colm Kearney, Elsevier 2005

Volume 7:

Value Creation in Multinational Enterprise, edited by, J. Choi and Reid W. Click, Elsevier 2006

Volume 8:

Asia-Pacific Financial Markets: Integration, Innovation and Challenges, edited by Suk-Joong Kim and Michael McKenzie, Elsevier 2007

Volume 9:

Institutional Approach to Global Corporate Governance: Business Systems and Beyond, edited by J. Choi and Sandra Dow, Emerald 2008

Volume 10:

Credit, Currency, or Derivatives: Instruments of Global Financial Stability or Crisis? edited by J. Choi and Michael G. Papaioannou, Emerald 2009

Volume 11:

International Banking in the New Era: Post-Crisis Challenges and Opportunities, edited by Suk-Joong Kim and Michael D. McKenzie, Emerald 2010

INTERNATIONAL FINANCE REVIEW VOLUME 12

INSTITUTIONAL INVESTORS IN GLOBAL CAPITAL MARKETS EDITED BY

NARJESS BOUBAKRI School of Business and Management, American University of Sharjah

JEAN-CLAUDE COSSET HEC Montre´al

United Kingdom – North America – Japan India – Malaysia – China

LIST OF CONTRIBUTORS Carlos Alves

Faculdade de Economia and CEF.UP, Universidade do Porto, Porto, Portugal

Najah Attig

Sobey School of Business, Saint Mary’s University, Halifax, Canada

Rolando Avendan˜o

OECD Development Centre, Paris, France and Paris School of Economics, Paris, France

Christopher Balding

HSBC School of Business, Peking University, Beijing, China

Narjess Boubakri

School of Business and Management, American University of Sharjah, Sharjah, United Arab Emirates

Don Bredin

UCD Michael Smurfit Graduate School of Business, Dublin, Ireland

Jean-Claude Cosset

HEC Montre´al, Montre´al, Que´bec, Canada

Sadok El Ghoul

Campus Saint-Jean, University of Alberta, Edmonton, Canada

Issa Faye

Development Research Department, African Development Bank, Tunis, Tunisia

Omrane Guedhami

Moore School of Business, University of South Carolina, Columbia, SC, USA E´cole des sciences de la gestion, Universite´ du Que´bec a` Montre´al, Montre´al, Que´bec, Canada

Olfa Hamza

Heiko Hesse

Monetary and Capital Markets, International Monetary Fund, Washington, DC, USA ix

x

LIST OF CONTRIBUTORS

Maher Kooli

E´cole des sciences de la gestion, Universite´ du Que´bec a` Montre´al, Montre´al, Que´bec, Canada

Yuhua Li

Graduate School of Economics, Kyushu University, Hakozaki, Higashiku, Fukuoka, Japan

Ningyue Liu

UCD Michael Smurfit Graduate School of Business, University College Dublin, Blackrock, Dublin, Ireland

Victor Mendes

CMVM – Comissa˜o do Mercado de Valores Mobilia´rios, Lisboa, Portugal; CEFAGE-UE Universidade de E´vora, E´vora, Portugal

S. V. D. Nageswara Rao

School of Management, IIT Bombay, Mumbai, India

Nabil Samir

HEC Montre´al, Montre´al, Que´bec, Canada

Javier Santiso

ESADE Business School, Madrid Campus, Barcelona, Spain

Elvira Sojli

Rotterdam School of Management, Erasmus University, Rotterdam, the Netherlands

Hyacinthe Y. Some´

HEC Montre´al, Montre´al, Que´bec, Canada

Gohar G. Stepanyan

Faculdade de cieˆncias econo´micas e empresariais, Universidade Cato´lica Portuguesa, Lisbon, Portugal

Tao Sun

Monetary and Capital Markets, International Monetary Fund, Washington, DC, USA

Mangesh Tayde

Bombay Stock Exchange and School of Management, IIT Bombay, Mumbai, India

Wing Wah Tham

Erasmus School of Economics, Erasmus University, Rotterdam, the Netherlands

xi

List of Contributors

Thouraya Triki

Development Research Department, African Development Bank, Tunis, Tunisia

Konari Uchida

Faculty of Economics, Kyushu University, Hakozaki, Higashiku, Fukuoka, Japan

Yao Yao

HSBC School of Business, Peking University, Beijing, China

Emerald Group Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2011 Copyright r 2011 Emerald Group Publishing Limited Reprints and permission service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. No responsibility is accepted for the accuracy of information contained in the text, illustrations or advertisements. The opinions expressed in these chapters are not necessarily those of the Editor or the publisher. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-78052-242-5 ISSN: 1569-3767 (Series)

Emerald Group Publishing Limited, Howard House, Environmental Management System has been certified by ISOQAR to ISO 14001:2004 standards Awarded in recognition of Emerald’s production department’s adherence to quality systems and processes when preparing scholarly journals for print

INTRODUCTION TO INSTITUTIONAL INVESTORS IN GLOBAL CAPITAL MARKETS Narjess Boubakri, Jean-Claude Cosset and Hyacinthe Y. Some´ PART I: INTRODUCTION TO INSTITUTIONAL INVESTORS IN GLOBAL CAPITAL MARKETS Institutional investors have increasingly gained importance since the early 1990s. The assets under management in these funds have increased threefold since 1990 to reach more than US$45 trillion in 2005, including over US$20 trillion in equity (Ferreira & Matos, 2008). Further, the value of institutional investors’ assets represents roughly 162.6% of the OECD gross domestic product in 2005 (Gonnard, Kim, & Ynesta, 2008). Given the magnitude of institutional investors’ holdings relative to the world market capitalization, challenging questions on the economic role of these investors have been raised. One such question concerns their impact on the stability of stock markets. On the one hand, active strategies of buying and selling shares by these investors may contribute to moving stock prices away from their fundamental values. On the other hand, if all institutional investors react to the same information in a timely manner, they are in fact helping to increase market efficiency by speeding up the adjustment of prices to new fundamentals (for competing theories on the role of institutional investors, see, e.g., Lakonishok, Shleifer, & Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 3–13 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012003

3

4

NARJESS BOUBAKRI ET AL.

Vishny, 1992). This view of institutional investors as ‘‘efficiency drivers’’ generated considerable debate for many years (see, e.g., Ferreira & Laux, 2007; French & Roll, 1986). Another important question about institutional investors that has caught the attention of the academic world is their impact on corporate governance practices. Institutional investors are large entities with considerable amounts of money to invest, and are thus more likely to buy sizeable blocks of a target firm’s common stock. In addition, given their informational advantage, these investors are likely to weigh more heavily on target firms while monitoring top management activities (Ferreira & Laux, 2007). Although corporate governance is mostly determined at the country level, institutional investors are considered major drivers of changes in corporate governance systems (Gillan & Starks, 2003). In particular, the effectiveness of institutional investors as a corporate governance mechanism will likely depend on the level of shareholder protection in the country. In this vein, Aggarwal, Erel, Ferreira, and Matos (2011) show that institutional investors play a crucial role in corporate governance practices of local firms, but only in countries with strong investor protection. In countries with weak investor protection, the main drivers of corporate governance improvements are instead foreign institutions that originate from countries with strong investor protection. The recent financial and economic crisis has also raised concerns about the economic and social effect of institutional investment strategies. On the one hand, while short-term investments provide market liquidity and accountability, they may also lead to underinvestment in maintenance, customer loyalty, employment training, research and development owing to their primary focus on labor-market reputation and stock prices. On the other hand, long-term investments have at least two significant impacts on corporations and the society as a whole: first, long-term investors can act as a stabilizing force during economic downturns by buying securities when liquidity dries up; second, long-term investors will lead firms to better align their objectives and activities with long-term economic growth, particularly from long-term environmental, governance, and social perspectives. According to the World Economic Forum report (2011), estimates of global infrastructure needs have reached US$ 3 trillion per annum, a sum which public finances are increasingly unable to meet.1 Although long-term institutional investors represent about half of the world’s professionally managed assets, the report shows that only 25% (about US$ 6.5 trillion as of 2009) of their assets is used for long-term investment. Given such a small percent devoted to long-term investments, the role that institutional

Introduction to Institutional Investors in Global Capital Markets

5

investors might play in the global economy remains limited. The purpose of this book is to shed light on the influence of institutional investors on global markets over the recent decades, and to identify their perspectives for the future. In this book, ‘‘institutional investors’’ refers to investments companies, mutual funds, pension funds, foundations, sovereign wealth funds (SWFs), insurance companies, and investment banks. We shall particularly focus on SWFs defined as ‘‘a government investment vehicle that invests in foreign currency denominated assets and whose management is distinct from that of official reserves’’ (Jen, 2007, p. 1). A SWF is set up for a variety of macroeconomic purposes, which include short-term objectives (economic stabilization) and long-term investment (development funds and savings for future generations). According to the IMF (2008), SWFs probably manage between US$ 2–3 trillion. The increased importance of SWFs in the global financial markets has recently fueled a heated debate on their size, lack of transparency, and investment strategies, assumed by some to be driven by political objectives. Further, SWF investments are vulnerable to host countries’ regulations on capital mobility.2 The present book addresses some of these concerns. Overall, the purpose of this book, titled ‘‘Institutional investors in global capital markets,’’ is to investigate institutional investors’ portfolio preferences, their influence on firm activities and local economies, and their reaction to the recent financial and economic crisis. The book is divided into four parts. Part I is an introduction to the book. Part II covers three chapters which study the economic and financial impact of institutional investors. In Part III, four chapters analyze the investment preferences of institutional investors. Part IV has three chapters which focus on the benefits of SWFs. Finally, three chapters in Part V analyze the drivers of the asset allocation of SWFs.

PART II: INSTITUTIONAL INVESTORS: THEIR ECONOMIC AND FINANCIAL IMPACT This part starts with a chapter titled ‘‘Foreign institutional investors’’ by Gohar Stepanyan who reviews the empirical literature on the process of international financial integration and the growing role of foreign institutional investors. Specifically, Gohar Stepanyan examines how institutional investors accelerate the development of capital markets and

6

NARJESS BOUBAKRI ET AL.

economies abroad. The author also investigates the determinants of their investments, both in the domestic and foreign markets, as well as their role in promoting good corporate governance practices worldwide. In spite of the concern that short-term international capital flows could be harmful in developing market economies, the author reports limited academic evidence of a destabilizing effect of foreign investment activity. The author concludes that the presence of a ‘‘home bias’’ in international portfolio investment is probably due to the systematic preference of institutional investors for common stocks of large and well-known foreign firms. The second chapter, written by Najah Attig, Sadok El Ghoul and Omrane Guedhami, is titled ‘‘Institutional investment horizon and firm credit ratings.’’ This chapter studies the impact of institutional investment horizons on firm credit ratings. The authors find that both the ownership stake and the number of long-term institutional investors contribute to more efficient monitoring, and thus reduced managerial myopism and self-interested behavior, as reflected in higher firm credit ratings. Further, the authors find that the monitoring incentive of institutional investors depends on their heterogeneity, as evidenced by their different investment horizons. From these results, they claim that focusing on institutional shareholdings masks important variations in the governance role of institutional investors, which may help explain mixed evidence in the existing literature on the monitoring role of institutional ownership. The third chapter, ‘‘Divestment of foreign strategic investment in China’s banking sector: Causes and consequences’’ by Yuhua Li and Konari Uchida, examines 10 foreign financial institutions’ divestments in the Chinese banking sector. The results of this study suggest that poor performance of foreign financial institutions, due to the global financial crisis, and the institutions’ regulated low-equity ownership are the primary causes of divestment. In contrast, the authors report that Chinese banks’ poor performance does not affect foreign divestments. The authors also show that business cooperation is usually ended when a foreign financial institution fully divests its equity stakes in a Chinese bank. In addition, the authors show that the Bank of China and China Construction Bank, which experienced large H-share divestments, suffered from economically large declines in the A-share values (A-shares are traded in domestic stock markets while H-shares are traded in the Hong Kong stock exchange). Li and Uchida conclude that these results suggest that banking sector developments that rely on foreign investments are vulnerable to an economic recession in developed countries.

Introduction to Institutional Investors in Global Capital Markets

7

PART III: INVESTMENT PREFERENCES OF INSTITUTIONAL INVESTORS The first chapter of this part, written by Don Bredin and Ningyue Liu, is titled ‘‘Domestic and foreign institutional investor behavior in China: Financial Characteristics and Corporate Governance.’’ The authors compare the financial characteristics and the corporate governance indicators of the companies in which foreign funds operating in China (as Qualified Foreign Institutional Investor (QFII)) and domestic Chinese funds have invested. The results of their analysis suggest that foreign funds prefer to invest in transportation, metals, non-metals and machinery, and generally avoid investing in real estate, construction, media, and culture, all of which require a deep local knowledge. The portfolios of domestic Chinese funds are distributed more evenly across industries than are foreign funds. The comparative analysis also reveals that the firms targeted by foreign funds are significantly different from those targeted by domestic funds, particularly with respect to size, profit, and managerial compensation. The authors conclude that their findings on the differences between QFIIs’ and domestic fund investment preferences should have implications for policy makers who aim to attract foreign investors to emerging markets. The authors of the second chapter of this part, titled ‘‘Institutional investors’ participation in foreign firms: Evidence from ADRs,’’ are Narjess Boubakri, Olfa Hamza, and Maher Kooli. The authors examine the firm- and country-level determinants of US institutional investors’ holdings in American Depositary Receipts (ADRs) from emerging markets. Using a sample of 112 firms from emerging markets that were listed as ADRs between 1990 and 2005, the authors find that institutional investors hold higher stakes in foreign firms that are listed on more restrictive exchanges and in large, privatized, more liquid, and more transparent firms. Mutual investors and other institutional investors also favour firms from countries with weaker institutional environments and with a civil law legal tradition. As noted by the authors, these results have interesting implications for managers of foreign firms which wish to attract capital from foreign institutional investors. In the third chapter, titled ‘‘Do foreign institutional investors exhibit herding and positive feedback trading in Indian markets?’’ Mangesh Tayde and Nageswara Rao study the behavior of foreign institutional investors (FIIs) in the Indian capital market. This chapter constitutes the first empirical study on whether FIIs in India follow herding and positive feedback trading strategies. To identify such behavior, the authors collected

8

NARJESS BOUBAKRI ET AL.

data on the daily purchase and sale trades executed by the FIIs. When they consider the complete sample period (i.e., 2003–2009) and the full sample of firms Tayde and Rao fail to find evidence of a strong herding behavior. Nevertheless, their results suggest that herding is stronger for large companies with better performance, most probably owing to the fact that they have higher liquidity and are more extensively followed by financial analysts. The fourth and last chapter of this part is written by Carlos Alves and Victor Mendes and is titled ‘‘Do financial conglomerates have an incentive to prevent managers of other firms from pursuing their own interest?’’ In this chapter, the authors develop a theoretical model to analyze the role of financial conglomerates in reducing agency costs in target firms. They show that, in certain conditions (e.g., if the fees charged by a bank are within a certain range) conglomerates will not oppose managers pursuing their own interests at the expense of the shareholders. However, if the bank is able to obtain capital inflows that react to mutual fund performance, we should observe more converging interests between the conglomerate shareholders and fund investors.

PART IV: THE BENEFITS OF SOVEREIGN WEALTH FUND INVESTMENTS In the first chapter of Part IV, titled ‘‘The impact of foreign government investments: Sovereign wealth fund investments in the United States,’’ Elvira Sojli and Wing Wah Tham show that foreign and politically connected large investors like SWFs improve firm value. In the short run, the market reacts positively to SWF investments in anticipation of enhanced monitoring and increased benefits from internationalization. In the long run, the target firms’ degree of internationalization and Tobin’s q undergo a marked increase after SWF investments. Interestingly, the authors show that the increase in Tobin’s q is associated with the number of governmentrelated contracts granted by SWF countries. As Sojli and Tham point out, these results suggest that government-related contracts are a mechanism through which government connections can affect firm value. The second chapter of this part, titled ‘‘What do sovereign wealth funds imply for financial stability?’’ and written by Tao Sun and Heiko Hesse examines financial stability issues that arise from the increased presence of SWFs in global financial markets. To do so, the authors use an event-study

Introduction to Institutional Investors in Global Capital Markets

9

approach to determine whether and how stock markets respond to the announcements of investments and divestments by SWFs. Based on 166 publicly traceable events of investments and divestments by major SWFs between 1990 and 2009, the authors evaluate the short-term financial impact of SWFs on host public equity markets. To do so, Sun and Hesse consider different sectors (financial and nonfinancial), actions (buy and sell), market types (developed and emerging), and level of corporate governance (high and low score). Their results suggest that there was no significant destabilizing effect of SWFs on equity markets, which is consistent with available anecdotal evidence. The third chapter of this part is titled ‘‘Africa’s quest for development: Can sovereign wealth funds help?’’ and is written by Thouraya Triki and Issa Faye. This chapter discusses the potential role that SWFs could play in African economies, both as recipient and home countries. The authors use new hand-collected data and therefore a unique database to describe the landscape of African SWFs as well as SWF interventions on the African continent. Triki and Faye observe that African SWFs are relatively small, suffer from poor governance structures and focus on stabilizing local economies. In light of these findings, the authors conclude that the potential role of SWFs as long-term institutional investors aimed at fostering economic growth should be limited unless current practices are changed. However, Triki and Faye also observe that foreign SWFs show a growing interest in Africa and could play a bigger role in supporting the continent’s growth provided African governments use the appropriate strategies to attract their funding.

PART V: SOVEREIGN WEALTH FUNDS: DO POLITICAL OBJECTIVES DRIVE THEIR ASSET ALLOCATION? First chapter of this part is by Christopher Balding and Yao Yao and is titled ‘‘Portfolio allocation for sovereign wealth funds in the shadow of commodity-based national wealth.’’ The key point of this chapter is that SWFs’ investment strategies should not resemble those of other institutional investors. The authors consider a balanced national wealth portfolio that accounts for the implied national wealth of unmonetized natural resources. They then estimate the optimal portfolio for an oil exporting state managing a SWF with a dataset including returns from 19 major assets (encompassing

10

NARJESS BOUBAKRI ET AL.

equity, debt, and commodity holdings). The authors find that when the returns and volatility from oil prices are included in the risk profile of national wealth, SWFs should invest in lower risk equity indices and highquality debt like the S&P 500 and sovereign debt. Further, as oil reserves decrease over time, SWFs should diversify into more balanced portfolios though remaining over weighted in fixed income. Finally, they find that the long-term growth of SWFs depends more on the price of oil and prudent risk management than on financial asset returns. The authors conclude that SWF managers and public policy makers should consider a larger picture than the risk-return adjusted profile of a financial assets portfolio. The second chapter is by Rolando Avendan˜o and Javier Santiso and is titled ‘‘Are Sovereign wealth funds investments politically biased? A comparison with other institutional investors.’’ This chapter studies the objectives of sovereign wealth funds (SWFs), which are often suspected of going beyond risk-return objectives. This study shows that the fear that sovereigns with political motivations use their financial power to secure large stakes in Western companies is unfounded. Indeed, the authors document that SWF investment decisions do not differ greatly from those of other wealth managers, specifically mutual funds. To do so, the authors analyze these investments on a geographical and sector basis. They look at the political regime characteristics of the target countries for both groups of investors and they report that SWFs do not discriminate according to this variable when investing. Both groups invest in democratic and nondemocratic regimes. They also report that there is no significant gap in the corporate governance characteristics of the firms both groups invest in. Finally, they provide a comparison of SWFs and other public funds based on governance features related to investment. Avendan˜o and Santiso conclude that financial strategies rather than political bias drive the asset allocation strategies of SWFs. The third and final chapter of this part is an empirical extension of the previous chapter. It is written by Narjess Boubakri, Jean-Claude Cosset, and Nabil Samir and is titled ‘‘Sovereign wealth fund acquisitions: A comparative analysis with mutual funds.’’ Focusing primarily on firm-level characteristics, the authors show that SWFs have investment tastes that are different from other institutional investors, including mutual funds. Indeed, at the firm level, SWFs, unlike mutual funds, prefer larger, less liquid, less innovative firms, as well as those with a more concentrated ownership. Further, SWFs invest more than mutual funds in firms that have a temporary financial constraint. At the country level, the authors find that the country culture and religion as well as its level of investor protection are not

Introduction to Institutional Investors in Global Capital Markets

11

significant determinants of SWF investment choices, which adds to the evidence of Avendan˜o and Santiso. What is important for SWFs is whether the host country is geographically close to the home country. This book investigates the role of institutional investors in the global economy. Institutional investors are large corporations that invest large amounts of capital. As discussed earlier, institutional investors have increasingly gained in importance since the 1990s, as evidenced by the value of assets under their management (US$45 trillion in 2005 with over US$20 trillion in equity). For the most part, institutional investors manage the equities of large, privatized, more liquid and transparent firms. They also tend to hold assets of ADR-listed foreign firms domiciled in weaker institutional environments (Boubakri, Hamza, and Kooli). Institutional investors’ holdings are evenly distributed across their home country’s industries. However, foreign institutions investing abroad have a preference for industries that need less local knowledge, such as transportation and machinery (Bredin and Liu). In India moreover, it appears that institutional investors follow herding and positive feedback trading strategies (Tayde and Rao). The studies covered in this book have also shown that institutional investors contribute to more efficient capital allocation, more efficient risk sharing, capital market development, and improvement in the structure of external finance (Stepanyan). Specifically, institutional investors with longterm investment horizons contribute to more efficient monitoring and less agency costs, and considerably improve the information environment (Attig, El Ghoul, and Guedhami). Furthermore, under certain conditions, institutional investors like financial conglomerates can be active in monitoring managers of listed companies, thereby reducing agency costs (Alves and Mendes). Finally, a firm’s value is strongly related to foreign institutional holdings in the firm. Thus, in the Chinese banking sector, a significant drop in foreign institutional holdings reduced the market value of domestic holdings (Li and Uchida). The growing number of stock holdings held by institutional investors such as SWFs raises concerns about the motivations behind their investment strategies. These funds have not, however, been shown to have a destabilizing effect on equity markets (Sun and Hesse). Concerns that SWFs have political motivations may therefore be exaggerated. Investment strategies by SWFs do not differ significantly from those of other institutional investors (Avendan˜o and Santiso). Like mutual funds, SWFs invest in large and profitable firms. However, unlike mutual funds, SWFs prefer less liquid, less-innovative firms and those with more concentrated

12

NARJESS BOUBAKRI ET AL.

ownership. They also invest in geographically close host countries (Boubakri, Cosset, and Samir). SWF investments have a positive impact on firm performance. The market reacts positively in the short-run to SWF investment in the expectation of improved monitoring and an increase in internationalisation. In the long-run, firm value increases with the number of government-related contracts granted by SWF countries (Sojli and Tham). In Africa, SWFs should define clear objectives that account for the home as well as the host countries’ interests (Triki and Faye). Conflict of interests between the home and host countries may undermine the return from SWF investments. SWFs that are based on commodities should diversify their portfolio to include low-risk debt and fixed-income assets to balance the higher volatility of commodity prices (Balding and Yao). This will guarantee long-term growth of the funds and savings for future generations. There is a growing need for infrastructure investment as reported by the World Economic Forum (2011). SWFs and other institutional investors should devote a large proportion of their assets to long-term investing, thereby contributing to sustained economic growth and financial stability.

NOTES 1. In March 11, 2011 the Brookings Institution pointed out that sovereign wealth funds (SWFs) may be a solution to fixing the broken infrastructures of the United States with its considerable public finance burden, i.e., a US$1.5 trillion deficit and US$13.5 trillion total debt as of 2010. 2. In 2008, these concerns led the International Working Group of Sovereign Wealth Funds to adopt and implement the Generally Accepted Principles and Practices (GAAP) – the Santiago principles (IWG-SWF, 2008).

REFERENCES Aggarwal, R., Erel, I., Ferreira, M., & Matos, P. (2011). Does governance travel around the world? Evidence from institutional investors. Journal of Financial Economics, 100, 154–181. Ferreira, M. A., & Laux, P. A. (2007). Corporate governance, idiosyncratic risk, and information flow. Journal of Finance, 62, 951–989. Ferreira, M. A., & Matos, P. P. (2008). The color of investors’ money: The role of institutional investors around the world. Journal of Financial Economics, 88, 499–533. French, K. R., & Roll, R. (1986). Stock return variances: The arrival of information and the reaction of traders. Journal of Financial Economics, 17, 5–26.

Introduction to Institutional Investors in Global Capital Markets

13

Gillan, S., & Starks, L. (2003). Corporate governance, corporate ownership, and the role of institutional investors: A global perspective. Journal of Applied Finance, 13, 4–22. Gonnard, E., Kim, E. J., & Ynesta, I. (2008). Recent trends in institutional investors statistics. Financial Market Trends, OECD. International Monetary Fund. (2008). Sovereign Wealth Funds – A work agenda. IWG-SWF. (2008). Sovereign wealth funds general accepted principles and practices, ‘‘Santiago Principles’’. Jen, S. (2007). Sovereign wealth funds: What they are and what’s happening. World Economics, 8(4), 1–7. Lakonishok, J., Shleifer, A., & Vishny, R. W. (1992). The impact of institutional trading on stock prices. Journal of Political Economy, 32, 23–43. World Economic Forum. (2011). The future of long-term investing. World Economic Forum report in collaboration with Olivier Wyman.

FINANCIAL LIBERALIZATION AND FOREIGN INSTITUTIONAL INVESTORS: LITERATURE REVIEW Gohar G. Stepanyan STRUCTURED ABSTRACT Purpose – Examine the role of institutional investors in accelerating the development of capital markets and economies abroad, the determinants of their investment, both in the domestic and foreign markets, and their importance in promoting good corporate governance practices worldwide and facilitating increased financial integration. Methodology/approach – Review and synthesize recent academic literature (1970–2011) on the process of international financial integration and the role of foreign institutional investors in the increasingly global financial markets. Findings – Despite the concern that short-term flow of international capital can be destructive to the emerging and developing market economies, academic evidence on a destabilizing effect of foreign investment activity is limited. Institutional investors’ systematic preference for stocks of large, well-known, globally visible foreign firms can explain the presence of a home bias in international portfolio investment. Research limitations – Given the breadth of the two literature streams, only representative studies (over 45 published works) are summarized. Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 17–50 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012004

17

18

GOHAR G. STEPANYAN

Social implications – Regulators of emerging markets should first improve domestic institutions, governance, and macroeconomic fundamentals, and then deregulate domestic financial and capital markets to avoid economic and financial crises in the initial stages of liberalization reforms. Originality/value of paper – A useful source of information for graduate students, academics, and practitioners on the importance of foreign institutional investors. Keywords: Foreign institutional investors; global financial markets; cross-border capital flows JEL classifications: F21; F32; G11; G15

INTRODUCTION Liberalization efforts by both developed and emerging countries increased access to financial markets around the globe and resulted in unprecedented growth in international capital flows worldwide (see, e.g., Lewis, 1999; Errunza, 2001). Many countries have liberalized their capital markets to complement limited domestic sources of finance with foreign capital that has become an increasingly important source of investment. Both foreign direct investment (FDI) and foreign portfolio investment (FPI) flows have reached peak levels in recent years, as more and more countries have opened their markets to foreign investors. According to the International Monetary Fund, total investment in financial assets by foreign investors exceeded US$53 trillion in 2009, with over US$23 trillion invested in equities. Of total investment in equities, almost US$14 trillion represented cross-border portfolio investment in over 200 countries around the world, whereas the remaining US$9.8 trillion was in the form of direct investment in 66 countries. Moreover, over the past decade, two-thirds of foreign direct investment, on average, has taken the form of cross-border mergers and acquisitions (M&As).1 Such active role of international investors worldwide represents an unprecedented internationalization of the shareholder base of corporations worldwide. A key factor in the internationalization of global capital markets is the growing importance of foreign institutional money managers. Institutional

Financial Liberalization and Foreign Institutional Investors

19

investors are major players not just in developed markets but also in rapidly growing emerging market countries (Ferreira & Matos, 2008; Li, Moshirian, Pham, & Zein, 2006). Table 1 reports the average fraction of stock market capitalization of 26 countries around the globe that is held by domestic and foreign institutions as well cross-border M&As, as a percentage of all deals in each country, over the first 5 years of the twenty-first century. Institutional money investors are most prominent in North America, holding over 70% of the stock market capitalization in the United States and 38% in Canada. However, holdings of foreign institutional investors are more pronounced in European countries such as Finland, Ireland, and the Netherlands, where 32.2%, 29.8%, and 21.2% of the stock market capitalization, respectively, are in the hands of foreign institutions, compared to only 3.3%, 0.6%, and 1.2% of the market held by local money managers.2 Cross-border M&A activity across countries also displays large variation: firms in Japan, for example, are among the least targeted by foreign acquirers, with only 3.6% of all M&A deals being crossborder, whereas 100% of all completed M&As in Ireland and Luxemburg involve foreign acquirer. In this chapter, I review the academic literature on the process of international financial integration and the increasingly important role of foreign institutional investors. Given the breadth of the two literature streams and my own space limitations, the purpose of this synthesis is not to provide a comprehensive survey of all research in the area. Rather, I summarize representative studies, with apologies to the authors of many important and useful papers that were excluded from this review.

FINANCIAL LIBERALIZATION AND INTERNATIONAL CAPITAL FLOWS The concept of increased financial integration is central to the international finance literature. In financially integrated markets, domestic investors are able to invest in foreign assets and foreign investors in domestic assets; hence, assets of identical risk command the same expected return, regardless of trading location, and are priced based on the global price of risk. Liberalization reforms reflect regulatory changes leading to increased market integration (Bekaert, Harvey, & Lumsdaine, 2002). Before 1970s, most countries had restrictions on foreign investments that limited cross-border capital flows. Developed countries started eliminating such

Table 1.

Institutional Ownership and Mergers and Acquisitions by Country. All M&A Ddeals

Sample of Firms

Institutional Ownership (%)

Number of Deals

Number Market Total Domestic Foreign Number Percentage of Firms Capitalization Firms

Australia (AU) Austria (AT) Belgium (BE) Canada (CA) Denmark (DK)) Finland (FI) France (FR) Germany (DE) Greece (GR) Hong Kong (HK) India (IN) Ireland (IE) Italy (IT) Japan (JP) Luxembourg (LU) The Netherlands (NL)

1,753 180 259 1,746 314 223 1,491 1,308 371 1,074 393 127 456 4,070 54 372

584,469 62,072 219,469 888,813 109,511 202,065 1,556,741 1,122,865 108,190 519,263 218,769 89,732 676,377 3,414,759 47,110 748,685

6.4 8.7 10.5 38.4 18.7 35.5 18.3 17.5 5.5 8.7 10.3 30.4 12.2 7.7 16.9 22.4

0.9 0.7 3.3 20.6 7.4 3.3 5.8 7.0 0.3 1.5 1.6 0.6 2.5 1.5 0.7 1.2

5.5 8.0 7.2 17.8 11.3 32.2 12.5 10.5 5.3 7.3 8.7 29.8 9.8 6.2 16.2 21.2

195 6 13 425 17 12 85 73 15 24 39 4 20 251 3 28

11.1 3.3 5.0 24.3 5.4 5.4 5.7 5.6 4.0 2.2 9.9 3.1 4.4 6.2 5.6 7.5

Cross-Border M&A Deals

Value of Deals

Value

77,389 8,821 30,959 188,967 16,930 13,788 125,561 57,110 2,742 45,111 2,861 1,858 19,685 148,564 4,723 38,176

Number of Deals

Percentage Number Percentage Deals Market Capitalization 13.2 14.2 14.1 21.3 15.5 6.8 8.1 5.1 2.5 8.7 1.3 2.1 2.9 4.4 10.0 5.1

35 3 4 115 4 5 31 42 3 6 8 4 6 9 3 20

17.9 50.0 30.8 27.1 23.5 41.7 36.5 57.5 20.0 25.0 20.5 100.0 30.0 3.6 100.0 71.4

Value of Deals

Value

Percentage Deals Value

18,484 8,309 1,027 107,353 2,977 10,390 30,113 28,666 842 6,356 770 1,858 1,241 1,259 4,723 30,864

23.9 94.2 3.3 56.8 17.6 75.4 24.0 50.2 30.7 14.1 26.9 100.0 6.3 0.8 100.0 80.8

Norway (NO) Poland (PL) Portugal (PT) Singapore (SG) South Africa (ZA) Spain (ES) Sweden (SE) Switzerland (CH) The United Kingdom The United States All countries All countries (ex-United States) Other countries All countries (w/other)

330 104 137 617 772 278 550 392 3,592 11,753 32,716 20,963

111,425 40,035 66,648 168,734 220,671 493,337 295,888 781,184 3,047,705 13,992,086 29,786,605 15,794,519

18.2 12.4 9.3 8.8 9.5 15.0 29.2 17.8 18.8 73.3 43.0 16.1

6.6 2.2 1.2 1.0 2.3 1.9 16.3 3.0 7.5 67.9 34.6 5.0

11.6 10.1 8.1 7.7 7.1 13.2 12.8 14.8 11.3 5.4 8.4 11.1

27 14 7 25 34 18 35 17 228 1,714 3,329 1,615

8.2 13.5 5.1 4.1 4.4 6.5 6.4 4.3 6.3 14.6 10.2 7.7

8,829 1,189 828 16,773 9,603 15,070 10,436 9,556 433,782 2,311,874 3,601,183 1,289,310

7.9 3.0 1.2 9.9 4.4 3.1 3.5 1.2 14.2 16.5 12.1 8.2

18 11 5 6 7 6 17 9 82 224 683 459

66.7 78.6 71.4 24.0 20.6 33.3 48.6 52.9 36.0 13.1 20.5 28.4

4,750 1,111 349 3,904 5,999 5,067 4,816 6,572 250,091 314,021 851,910 537,889

53.8 93.4 42.2 23.3 62.5 33.6 46.1 68.8 57.7 13.6 23.7 41.7

7,340 40,056

2,333,791 32,120,396

17.0 41.1

0.1 32.1

16.9 9.0

302 3,631

4.1 9.1

140,430 3,741,613

6.0 11.6

106 789

35.1 21.7

97,973 949,883

69.8 25.4

Notes: This table is extracted from Ferreira et al.(2010) and presents institutional ownership and M&As for the 2000–2005 period by target country: average number of firms and market capitalization (in millions US$); average of total, domestic, and foreign institutional ownership as a percentage of market capitalization; number of completed M&A deals, percentage of listed firms targeted in deals, value of transactions of deals in millions US$ and as a percentage of market capitalization; and number of completed cross-border deals, number of cross-border deals as a percentage of the total number of deals, value of transactions of cross-border deals in millions US dollars and as a percentage of total value of transactions.

22

GOHAR G. STEPANYAN Capital Account

3 Less Liberization 2.5

Emerging Markets

2 1.5 Mature Markets

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1973

More Liberization

1975

1

Domestic Financial Sector Less Liberization

3 2.5 Emerging Markets 2 1.5 1

Mature Markets

1999

2001

2003

2005

1999

2001

2003

2005

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

More Liberization

Stock Market 3 Less Liberization 2.5 Emerging Markets

2 1.5

1997

1995

1993

1991

1989

1987

1985

1983

1981

1977

1975

1979

Mature Markets

1

1973

More Liberization

Fig. 1. Indexes of Financial Liberalization by the Level of Development, 1973–2005. The Three Indexes Display Separately the Liberalization of the Capital Account, the Liberalization of the Domestic Financial Sector, and the Stock Market Liberalization. The Value 3 Means Repression, 2 Means Partial Liberalization, and 1 Means Full Liberalization. The Indexes are a Cross-Country Average. Mature Markets are: Canada, Denmark, Finland, France, Germany, Ireland, Italy, Japan, Norway, Portugal, Spain, Sweden, the United Kingdom, and the United States. Emerging Markets are: Argentina, Brazil, Chile, Colombia, Hong Kong, Indonesia, Korea, Malaysia, Mexico, Peru, Philippines, Taiwan, Thailand, and Venezuela. Source: Kaminsky and Schmukler (2008).

Financial Liberalization and Foreign Institutional Investors

23

restrictions in the late 1970s and the early 1980s, followed by emerging countries’ financial liberalization reforms in the late 1980s and early 1990s. Fig. 1 captures the development of three main aspects of liberalization process – deregulation of the domestic banking industry (financial sector liberalization), removal of controls on international capital flows (capital account liberalization), and the stock market liberalization, across 28 emerging- and mature-market economies. For illustrative purposes, Table 2 provides a comparison of official equity market liberalization dates for 43 emerging countries across several studies. It is important to note that empirical studies examining the impact of the equity market liberalization use multiple indicators, not just the official liberalization dates, given the difficulty in pinpointing the exact date when the market becomes open to foreign investment. For example, investors can avoid capital controls and access foreign markets indirectly through American/Global Depositary Receipts (ADRs/GDRs) or country funds, even though the market itself is technically closed to foreign investors. The capital market liberalization process is a complex process reflecting a decision by a country’s government to allow foreigners invest in domestic equity securities. It may occur in stages and often be a part of more general regulatory change aimed at removing restrictions on both capital inflows and outflows. Many authors emphasize the gradual nature of the shift from closed to open capital markets that usually coincides with other equally important economic reforms such as macroeconomic stabilization, trade liberalization, privatization, and the relaxation of exchange and product market controls (Bekaert & Harvey, 2000; Henry, 2000a, 2000b), and legal reforms such as insider trading (Bekaert et al., 2005). Liberalization reforms may take many different forms, with some policy changes anticipated well in advance while others lacking credibility, and not all reforms taking place at the same time. The liberalization process might have drastic effect on the liberalizing country’s market activity and economy. The arrival of foreign investors may enhance market efficiency and liquidity, thereby reducing the cost of equity capital and increasing the valuation of local companies. The stock market concentration might decrease while trading volume may increase as a result of these new entrants. In addition, individual stocks might become less sensitive to local information and more sensitive to world events, reflecting local market’s integration with the world markets.

Table 2.

Comparison of Official Equity Market Liberalization Dates across Studies.

Argentina Bangladesh Botswana Brazil Chile Colombia Cote d’Ivoire Ecuador Egypt Ghana Greece Iceland India Indonesia Israel Jamaica Japan Jordan Kenya Korea Malaysia Malta Mauritius Mexico Morocco New Zealand Nigeria Oman Pakistan Peru Philippines Portugal Saudi Arabia South Africa Spain Sri Lanka Taiwan Thailand Trinidad and Tobago Tunisia Turkey Venezuela Zimbabwe

Henry (2000a)

Bekaert and Harvey (2000)

Kim and Singal (2000)

Bekaert et al. (2005)

11-1989

11-1989

11-1939

3-1988 5-1987 12-1991

5-1991 1-1992 2-1991

5-1991 9-1989 2-1991

12-1987

8-1986

11-1992 9-1989

11-1992 9-1989

12-1995

1-1978

6 1987 5-1987

1-1992 12-1988

1-1992

5-1989

5-1989

5-1989

1989 1991 1990 1991 1992 1991 1995 1994 1992 1993 1987 1991 1992 1989 1993 1991 1983 1995 1995 1992 1988 1992 1994 1989 1988 1987 1995 1999 1991 1992 1991 1986 1999 1996 1985 1991

6-1986

8-1995 2-1991

2-1991

5-1986

6-1991 7-1986

3-1986 7-1986

5-1986 1-1988

1-1991 9-1987

1-1991 8-1988

1-1990

8-1989 1-1990 6-1993

8-1989 1-1930 7-1993

1987 1997 1998 1989 1990 1993

Notes: This table compares official stock market liberalization dates for 43 emerging countries across authors. The dates are taken from Henry (2000a), Bekaert and Harvey (2000), Kim and Singal (2000), and Bekaert et al. (2005), respectively.

Financial Liberalization and Foreign Institutional Investors

25

Liberalization and Market Activity Numerous studies examine the impacts of increased capital market integration process and international capital flows. Henry (2000a) empirically estimates the influence of stock market liberalization on 12 emerging markets’ equity prices across Latin America and Asia. The author documents a substantial appreciation of aggregate share prices, occurring both in the months leading up to the implementation of a country’s initial stock market liberalization as well as in the implementation month itself. After controlling for co-movements with world stock markets, macroeconomic fundamentals, and major economic policy changes, the average valuation increase still remains large and statistically significant – 3.3% per month (or 26% overall) abnormal return over an 8-month window leading up to the implementation of initial stock market liberalization. Henry (2000a) also addresses the potential endogeneity problem arising from the policymakers’ incentives to liberalize stock markets in response to prolonged run-ups in equity prices. These results indicate an upward revaluation of aggregate share prices in emerging markets in anticipation of future stock market liberalizations, consistent with the prediction of standard international asset pricing models that stock market liberalization may reduce the liberalizing country’s cost of equity capital by allowing for risk sharing between domestic and foreign agents.3 If liberalization of capital markets reduces the liberalizing country’s cost of equity capital, then investment projects that had a negative net present value (NPV) before liberalization will turn into positive NPV projects after liberalization, leading to a surge in private physical investment. Henry (2000b) studies whether stock market liberalizations are indeed associated with increased investment by comparing the growth rate of real private investment on the heels of stock market liberalization in 11 emerging markets with the growth rate during nonliberalization periods. He finds that stock market liberalizations lead to private investment booms: the average growth rate of real private investment is 23%, 27%, and 17% in the first, second, and third year following stock market liberalization, respectively. Moreover, the positive correlation between private investment growth and stock market liberalization persists after controlling for world business cycle effects, contemporaneous economic reforms, and domestic fundamentals. Henry (2000b) finds that the ratio of FDI to private investment also tends to rise following stock market liberalization, suggesting that the increase in private investment does not simply substitute for FDI. One explanation for why FDI increases during

26

GOHAR G. STEPANYAN

these episodes is that stock market liberalization may be positively correlated with other economic reforms that reduce the operating risk of foreign multinationals with subsequent reduction in their cost of equity capital. Although one cannot necessarily conclude that stock market liberalizations cause investment booms – the political decision to liberalize a country’s stock market may be endogenous4 – the evidence presented in Henry (2000b) is relevant for the debate on whether liberalization reforms have any effect on real investment. The fact that stock market liberalizations are followed by a surge in private investment that cannot be explained by world business cycle effects, contemporaneous economic reforms, or domestic aggregate demand conditions suggests that liberalization of emerging capital markets may matter for investment after all. Using a more comprehensive sample of both developed and emerging countries, Bekaert et al. (2005) show that equity market liberalizations increase subsequent average annual real economic growth by about 1%. This growth effect is robust to alternative definitions of the liberalization, is distinct from the effects of capital account liberalization, and does not reflect variation in the world business cycle. Other simultaneous reforms only partially account for the equity market liberalization effect. Interestingly, the growth effect depends positively on development levels – the countries that benefit the most in terms of growth are those that are further along in terms of financial development, have English instead of French or other legal origins, good institutions, favorable conditions for foreign investment, better protection of shareholder rights, and higher accounting standards. It is evident that both firms and economies, as a whole, enjoy better growth opportunities after liberalizations. In a related paper, using a sample of 20 emerging markets across the globe, Bekaert and Harvey (2000) show that liberalization tends to decrease aggregate dividend yields and argue that this price change reflects both a change in the cost of capital and changes in growth opportunities. The cost of capital always decreases after capital market liberalization, with the effect varying between 5 and 75 basis points. In contrast, the effect of the liberalization on the investment activity of the country is positive – an increase of 75 basis points in the investment to gross domestic product (GDP) ratio. At the same time, Bekaert and Harvey (2000) document that the magnitude of the increase in correlation between emerging markets and the world market return is small and, thus, is unlikely to deter foreign investors attracted to developing markets by the gains of international diversification. In a subsequent study, Bekaert et al. (2002) confirm that the

Financial Liberalization and Foreign Institutional Investors

27

emerging markets’ integration with the world market, post liberalization, leads to permanent appreciation in equity prices that decrease dividend yields and expected returns; in addition, integration brings about or is accompanied by increased stock market development – significantly higher market capitalization to GDP ratio, increased U.S. holdings of the local market capitalization, sizable jumps in trading activity and liquidity, and higher correlation of local stock returns with respect to the world market return. Bekaert et al. (2002) also document that integration with the global market is associated with a lower cost of capital, improved credit ratings, appreciation of local currency, and increased real economic growth. Kim and Singal (2000) examine changes in the level and volatility of stock returns, inflation, and exchange rates around removal of restrictions on capital flows in 20 emerging markets. They find that stock returns increase immediately after market openings, indicating greater demand for the domestic securities by foreign investors, yet without accompanying increase in the volatility of stock returns. After this upward adjustment, stock returns fall, reflecting lower expected returns. For the sample as a whole, stock markets become more efficient in impounding information over longer horizons, evidenced by reduction in predictability of stock returns. The improvement in market efficiency and lower predictability in stock prices is consistent with increasing integration with the world market – as the foreign investors take advantage of emerging market inefficiencies, those inefficiencies decrease and the prices react more quickly to new information. Kim and Singal (2000) find no evidence of an increase in inflation or an appreciation of exchange rates. Moreover, volatility of inflation, nominal exchange rates, and real exchange rates all show a decrease following stock market liberalizations. Overall, there is a distinctive pattern of a decrease in volatility after the opening of emerging markets to foreign participation, suggesting that new capital inflows are not disruptive to the economy. Although the authors’ results vary across countries, liberalization of capital markets, on average, appears to have favorable effects on the emerging countries’ stock markets and economies.

Liberalization and Home Bias in Portfolio Investment As previously mentioned, deregulation of capital markets and the relaxation of capital controls lead to a dramatic decline in restrictions to international investment, resulting in substantial cross-border flows. Although not all

28

GOHAR G. STEPANYAN

obstacles have disappeared, investors in most countries can now invest abroad either directly or through financial intermediaries such as mutual funds to diversify their investments internationally. Yet, investors assign only a very small fraction of their wealth to foreign assets despite the well-documented benefits of cross-border diversification of portfolio allocations.5 This evidence constitutes what is generally referred to as a ‘‘home bias’’ – national equity portfolios are concentrated in the domestic equity market of the investor, and foreign ownership of shares is much smaller than one would expect in the presence of international diversification. For example, Kang and Stulz (1997) document that the weight of Japanese equities in portfolios of foreign investors from 1975 to 1991 was disproportionably lower than the weight of Japan in the world market portfolio, thus confirming the existence of a substantial home bias in the Japanese market. Home bias must be caused by some feature of international portfolio investment that offsets gains from diversification. Explanations offered to justify this bias include both explicit and implicit barriers to international capital flows; although, so far, no explanation emerges as generally accepted (Karolyi & Stulz, 2003; Lewis, 1999). Explicit barriers to international investment are those that are directly observable and quantifiable. Among them are the costs associated with cross-border investing such as taxes (withholding taxes on dividends or tax differential on domestic versus foreign interest-rate and equity income) and foreign securities transaction costs, restrictions on foreign exchange transactions, differential access to markets, institutional constraints, inflation risk and foreign exchange rate risk. Cooper and Kaplanis (1994) argue that explicit barriers to international investment are not large enough to explain the observed portfolio allocations of investors; the home-bias puzzle cannot be explained by neither inflation hedging nor direct observable costs of international investment, such as withholding taxes on dividends, unless investors exhibit risk aversion that is below conventional levels. French and Poterba (1991) make a similar point and claim that incomplete diversification is the consequence of investor choices rather than institutional constrains, whereas Tesar and Werner (1995) convincingly refute transaction costs as a likely cause of the observed reluctance of investors to diversify their portfolios internationally by pointing to the high level of international market activity, reflected both in the level of cross-border flows and the rate of turnover on foreign assets. In sum, the composition of the portfolio of foreign securities of a typical investor seems to reflect factors other than diversification of risk.

Financial Liberalization and Foreign Institutional Investors

29

Implicit barriers to international investment, however, are not directly observable or quantifiable and include such barriers as political risk, informational asymmetries between domestic and foreign investors, differences in language and culture, and behavioral biases affecting investor return expectations in foreign markets. For example, Kang and Stulz (1997) present empirical evidence that foreign investment in Japanese equities is concentrated in the largest firms, consistent with foreign investors having relatively less information about small firms than local investors and, thus, preferring equities of well-known firms. The use of informational disadvantage and political risk as explanations for the home bias, though, is difficult to reconcile with the evidence of a dramatic home bias in the U.S. market, where information disclosure standards are high and the risk of expropriation by the government is low. As demonstrated in Fig. 2, although the home bias in the United States decreased substantially over the past three decades of the twentieth century, as of 2001 foreign stocks still constituted only 10% of U.S. investors’ portfolio holdings. This stands in sharp contrast to the dominance of the U.S. market, as a whole, in the world market capitalization (see Table 1).

Fig. 2. U.S. Home Bias, 1973–2000. This Figure is Extracted from Karolyi and Stulz (2003) and Compares Foreign Equities Share in U.S. Investors’ Portfolio Holdings to the Foreign Equities Share of World Market Portfolio. Source: Flow of Funds Accounts of the United States, Flows and Outstandings, Federal Reserve Board.

30

GOHAR G. STEPANYAN

Although neither direct nor indirect barriers to international investment and ownership can explain the home bias per se, they appear to capture some portion of the cross-sectional variation in the home bias. Chan, Covrig, and Ng (2005) examine cross-border investment behavior of more than 20,000 mutual funds from 26 developed and developing countries, using a breakdown of the market value of their equity holdings across 48 countries for 1999 and 2000. Interestingly, all 26 countries exhibit domestic bias – the share of mutual fund holdings in the mutual fund’s domestic market is much larger than the world-market capitalization weight of the country, yet foreign mutual funds invest more in markets that are more developed, less remote from the rest of the world, larger in market capitalization, and have lower transaction costs and similar language. Factors such as economic development, capital controls, withholding taxes and investor protection have a less pervasive impact on the investment decisions of foreign institutional money managers. There is some evidence that institutional investors are more sophisticated in diversifying their investments. Hau and Rey (2008) present data on the degree of home bias at the fund level over the period of 1998–2002, using a detailed dataset on global equity holdings of mutual equity funds incorporated in the most developed 16 financial markets. Mutual funds appear to be home biased, but to a lesser degree than other investors.

FOREIGN INSTITUTIONAL INVESTORS Foreign investor as a category encompasses both institutional and individual investors. However, the cross-border investment activity is likely to be dominated by sophisticated and professionally more skilled institutional investors with relatively low transaction costs and large trades to justify the documented high volume of international capital flows and high foreign investment turnover rates relative to domestic turnover rates (see, e.g., Tesar and Werner (1995).6 In this section, I integrate the theoretical literature with the available empirical evidence on foreign investors’ investment strategies and discuss whether their trades are destabilizing to the developing country’s market and economy. Are Foreign Investors Momentum or Contrarian Investors? Numerous studies document that the trades of foreign investors are affected by past returns, so that they buy when prices have increased and sell when

Financial Liberalization and Foreign Institutional Investors

31

they have fallen. For example, Bohn and Tesar (1996) develop and test an intertemporal, international capital-asset-pricing model, where net purchases are decomposed into (i) transactions required for maintaining a balanced portfolio of securities and (ii) transactions triggered by timevarying investment opportunities, and find that over the 1980–1994 period U.S. investors tend to acquire stocks abroad when returns in a particular foreign market were high, rather than sell off ‘‘winning stocks’’ to maintain balanced portfolio weights. These results suggest that U.S. transactions in foreign equities were primarily driven by return-chasing – that is, moving into markets where returns were expected to be high and retreating from markets when expected returns were low. Interestingly, Bohn and Tesar (1996) find that the return-chasing strategy pursued by U.S. investors resulted, on average, in a return that was lower than what could have been obtained by holding a market-weighted portfolio of foreign equities. Moreover, this loss in return was not compensated by a reduction in risk – the mean return per unit of risk on the market portfolio was higher than the one on the portfolio selected by U.S. investors. Apparently, U.S. investors were chasing returns, but not in the right markets and at the right time. Brennan and Cao (1997) develop a model of international equity portfolio investment flows that is based on informational differences between foreign and domestic investors. The main empirical implication of the model is that purchases of foreign equities will be a linear function of returns on the foreign equity markets since foreign investors are less informed than domestic investors and, thus, are inclined to pursue momentum strategies. Brennan and Cao (1997) test the model by examining equity flows between the United States and 4 developed countries, and net U.S. purchases of equities in 16 emerging markets. The authors find some evidence that U.S. purchases of equities in foreign developed markets tend to be positively associated with the concurrent return in that market, consistent with U.S. investors being less well informed about those markets than local investors. However, there is no indication that investors in these four developed markets are less well informed than U.S. residents about the returns on U.S. equities. When the authors examine U.S. portfolio investment in emerging markets, they find strong evidence that U.S. purchases are positively associated with both the lagged and contemporaneous returns on the local market index. Thus, equity capital flows from the U.S. to emerging markets are consistent with a model in which U.S. investors are at an informational disadvantage relative to locals. In sum, Brennan and Cao (1997) conclude that when domestic investors possess a cumulative information advantage

32

GOHAR G. STEPANYAN

over foreign investors about the payoffs on the domestic market, foreign investors tend to purchase assets in periods when the return on those assets is high and to sell when the return is low. Choe, Kho, and Stulz (1999) use a large sample of Korean stocks to explore how foreign investors trade and how they impact stock prices. Before Korea’s dramatic economic crisis of the late 1997, the authors find strong evidence of positive feedback trading and herding by foreign investors – foreign investors buy (sell) more Korean stocks on days following positive (negative) market return; moreover, they buy (sell) Korean shares that outperformed (underperformed) the market over the previous day. Furthermore, foreign investors trade similarly over a short period of time. During the crisis period, however, herding falls and positive feedback trading by foreign investors mostly disappears. One possible reason for the decrease in herding is that lower liquidity during the crisis may have limited the trading of foreign investors. Grinblatt and Keloharju (2000) use detailed transaction data on stock trades for both individuals and institutional investors from Finland over the two-year period of 1995–1996 and also document that foreign investors pursue momentum strategies, buying past winning stocks and selling past losers. Portfolios of foreign investors outperform those of households, even after controlling for behavior differences. In contrast to Brennan and Cao (1997), Grinblatt and Keloharju (2000) claim that both momentum behavior with respect to near- and intermediate-term past returns and superior investor performance are driven by the level of sophistication of the investor. The most sophisticated players in the Finish financial market are foreign investors – often well capitalized foreign financial institutions with a long history of successful investment in other stock markets such as professionally managed mutual funds, hedge funds, and investment banks. Institutional investors generally take larger positions than individuals, have more resources to expend on research, and in many cases, view investment as a full-time career. Consequently, it is reasonable to view foreign institutional investors in Finland as more sophisticated than all of the other investor categories. Thus, investors perceived as sophisticated follow momentum strategies and exhibit superior performance, whereas investors perceived as naive follow contrarian strategies and exhibit inferior performance, consistent with momentum being a behaviorally driven anomaly in which ‘‘smart’’ investors take advantage of ‘‘naive’’ investors in equilibrium. Froot, O’Connell, and Seasholes (2001) explore daily data from State Street Bank & Trust, one of the world’s largest custodian banks, on international

Financial Liberalization and Foreign Institutional Investors

33

portfolio flows into and out of 16 developed and 28 emerging countries from 1994 through 1998. State Street’s clients are predominantly large institutional investors from developed countries, including pensions, endowments, mutual funds, and governments, and thus can be thought of as a large sample of sophisticated international investors. The authors document several important observations. First, net inflows are strongly influenced by lagged equity and currency returns, once again confirming international investors’ engagement in positive feedback trading. There is also very strong trend following international inflows, but the majority of the co-movement of flows and returns can be attributed to returns predicting future flows. Second, international portfolio inflows are slightly positively correlated across countries, and are more strongly correlated within regions. The correlation of flows in most regions, particularly within Asia, rises strongly during the Asian crisis, but not during the Mexican crisis. Third, international inflows have positive forecasting power for future equity returns in emerging markets, which can occur either due to superior information held by international investors on emerging markets, who reach their desired positions slowly to mitigate transaction costs and price impact of large trades, or due to price pressure of persistent capital flows.

Do Foreign Investors Destabilize Developing Markets? It has been frequently argued by academics, political leaders as well as in the popular press that foreign investors can exert a destabilizing influence on developing countries’ economies. Foreign investors were often blamed for the dramatic difficulties of the Asian and Latin American countries and the collapse of their currencies and stock markets in the late 1990s. As already discussed earlier, academics have pointed out that foreign investors do not follow ‘‘buy and hold’’ strategies; their trades are highly correlated, reflecting positive feedback trading and herding. Investors pursuing such strategies are often viewed as destabilizing since, if they trade as a group, their purchases can overheat the markets that they enter, increasing prices further and leading to bubbles, whereas their sales can create disarray and panic in the markets that they exit, causing prices to fall further and leading to market crashes. If foreign investors can indeed destabilize economies, the benefits from opening markets to investors around the globe are questionable. For one reason, however, positive feedback trading and herding are not necessarily destabilizing – local investors trading on fundamentals may be

34

GOHAR G. STEPANYAN

sufficiently powerful to prevent prices from moving away from fundamental values. For example, Choe et al. (1999) use order and trade data to examine whether large trading imbalances by foreign investors in Korea are followed by price continuations and an increase in volatility. They find no evidence of destabilizing effect of foreign investors’ trades on Korea’s stock market over the period of late 1996 to the end of 1997. In particular, Korean market adjusted quickly and efficiently to large sales by foreign investors, and these sales were not followed by negative abnormal returns. Choe et al. (1999) conclude that the impact of domestic buying on stock returns dominates the impact of foreign selling. Consistent with Choe et al. (1999), Karolyi (2002) finds no evidence of any destabilizing effect of foreign investors’ trading activity on stock and currency returns in Japan around the time of the Asian crisis in 1997. The crisis likely reflected structural and policy distortions in the countries of the Asia-Pacific region, although foreign investors were often blamed for exerting a destabilizing influence on stock prices and foreign currency values. The investment behavior of foreigners in Japan was distinctly different from that of each of the domestic investment groups. Foreign investors were consistently positive-feedback traders before, during, and after the Asian crisis, buying (selling) on Nikkei index increases (decreases) and yen/dollar appreciations (depreciations), whereas Japanese banks, financial institutions, investment trusts and corporations were negativefeedback, or contrarian, investors. The Asian financial crisis did scare foreign investors out of Japan as they became net sellers of Japanese equities during this turbulent period. However, this shift in aggregate foreign portfolio investment activity did not aggravate the effect of the crisis on the market. The sales by foreigners as a group during the crisis period were absorbed by increases in purchases by Japanese corporations, banks and other financial institutions. Interestingly, Karolyi (2002) presents evidence of the success of the foreigners’ positive-feedback trading strategy – they were able to accumulate over 1,250 billion yen (US$12.1 billion) over the entire period, including 210 billion yen (US$1.8 billion) during the crisis period, reflecting foreigners’ good market timing skills. In their examination of daily international portfolio flows worldwide from 1994 through 1998, Froot et al. (2001) point to an interesting fact – international investors did not abandon emerging markets during the Asian crisis. They remained net buyers of emerging market equities over the July 1997–July 1998 period, simply at a reduced rate. The reduced level of net flows may explain why equity prices in most emerging markets declined during the Asian crisis. The persistence that characterizes international

Financial Liberalization and Foreign Institutional Investors

35

portfolio flows suggests that prices in the emerging region had been bid up as a result of high volume of past inflows and in anticipation of further inflows. When these future inflows failed to materialize, prices declined. Although the academic evidence on the adverse effects of foreign investor capital flows on developing economies and markets is weak, Stiglitz (2000) calls for a greater regulation of short-term speculative capital flows, arguing that the volatile nature of these flows can give rise to economic instability and trigger financial and economic crises. At the same time, he states that ineffective regulatory policies and weak financial institutions were at the core of the crises in East Asia, Latin America, and Russia. Before initiating the process of increased financial liberalization and open market economy, emerging market countries should have strong legal and financial framework in place.

DETERMINANTS OF FOREIGN INSTITUTIONAL INVESTMENT Various authors have examined whether the cross-sectional variation in foreign institutional investors’ portfolio allocations can be explained by firm-specific characteristics such as size, leverage, cash positions, dividend policy, growth prospects, and international presence or visibility, or market characteristics such as mandatory disclosure and corporate governance rules. In this section, I summarize the evidence on which firm- and marketspecific attributes are important in explaining investment decisions of foreign investors from studies of (i) foreign holdings in a single country, (ii) worldwide foreign holdings, (iii) foreign investments by investors from a single country, (iv) international stock preferences of certain types of institutional investors from a single country, and (v) stock preferences of different types of institutional investors around the globe. Target Firm Characteristics A number of studies highlight the importance of firm characteristics in the investment decisions of foreign investors, suggesting that institutional money actively selects which foreign firms to invest in, beyond just following country-level allocations for diversification purposes. For example, in their study of stock ownership in Japanese firms by nonJapanese investors from 1975 to 1991, Kang and Stulz (1997) find that

36

GOHAR G. STEPANYAN

foreign investors in Japan have disproportionately high holdings of firms in manufacturing industries (compared with utilities) and in firms with good accounting performance, high market-to-book ratio, low unsystematic risk and low leverage, relative to the weights of the Japanese market portfolio. Although foreign investors prefer primarily large Japanese firms, there is some evidence that small firms that export more have greater foreign ownership, whereas firms with low share turnover, regardless their size, have lower foreign ownership, consistent with the notion that foreigners are reluctant to hold securities of firms they are not familiar with or equities with high transaction costs. Cross listing on a U.S. exchange is also related to foreign ownership since cross-listed firms have more of an international presence and, thus, attract foreign investors due to lower information asymmetries and transaction costs.7 The cost to foreign investors of overinvesting in Japanese firms that they were better informed about was holding a portfolio with much greater volatility than that of the market portfolio, without any gain in expected return. Dahlquist and Robertsson (2001) analyze a detailed dataset of equity ownership of Swedish firms for the period 1991–1997 to identify the common firm-specific determinants of foreign ownership. Their overall results support the findings of Kang and Stulz (1997) that foreign investors tend to prefer firms with certain attributes: firms that are large, have high amount of cash on their balance sheets and low dividend yields. When they further investigate what drives the foreigners’ preference for large firms, Dahlquist and Robertsson (2001) find that market liquidity of a firm’s shares and presence in international goods and services markets, measured through either export sales or listings on other international stock exchanges, seem to characterize foreign holdings in Swedish equities better than firm size alone. Foreigners also tend to underweight firms with highly concentrated ownership. An important contribution of Dahlquist and Robertsson (2001) is that most of the features associated with foreign ownership are driven by the fact that the typical foreign investors in Sweden is a large institution; hence, they conclude that they do not find a foreigner bias per se, but rather that institutional investors in general deviate from holding the market portfolio. Aggarwal, Klapper, and Wysocki (2005) also confirm the importance of firm visibility in foreign investment decisions of institutional investors. They find that firm size and number of analysts following are two of the most important drivers of U.S. actively managed mutual funds’ investment decisions in 30 emerging markets. The U.S. funds tend to significantly overweight their holdings to larger firms, firms with lower leverage and

Financial Liberalization and Foreign Institutional Investors

37

higher price-to-book ratios, and firms that have ADRs. The U.S. funds invest in unlisted ADRs only when these firms adhere to high-quality financial disclosure standards such as consolidated financial statements based on internationally recognized accounting standards or a clean opinion from an internationally recognized auditor. Therefore, emerging market firms, unlisted on the U.S. stock exchanges, can capture similar foreign investment benefits as firms with listed ADRs if they adopt high-quality discretionary disclosure practices. In a comprehensive study of the role of institutional investors from 27 countries, Ferreira and Matos (2008) record substantial diversity in the revealed stock preferences of various groups of institutional investors, depending on their geographic origin. Although all institutional investors, regardless their geographic origin, have a strong preference for the stock of large, widely held firms and firms with strong governance indicators, foreign institutions avoid high dividend-paying stocks and are biased towards firms that are cross-listed in the United States, have high external visibility through high foreign sales and analyst coverage, and are members of the Morgan Stanley Capital International World Index. Interestingly, U.S. institutions diverge from non-U.S. foreign institutions in their preference for value over growth stocks, and a tendency to hold stocks in English-speaking countries and emerging markets.

Market Characteristics As already stated earlier, foreign institutions can play an important role in funding corporations, especially in countries in which domestic sources of outside finance are limited. However, poor corporate governance can often act as a substantial deterrent to foreign investment. Leuz, Lins, and Warnock (2009) study 4,409 firms from 29 countries to assess whether and why concerns about both firm- and country-level corporate governance result in fewer foreign holdings. Building on the notion that foreign investors are often at an informational disadvantage relative to local investors, the authors conjecture that these information asymmetries are particularly pronounced with respect to the evaluation of a firm’s governance and ownership structures and the scope for expropriation by controlling insiders. In many countries around the world, business transactions, financing arrangements, ownership structures, and, ultimately, corporate governance are shaped by relationships among a tightly knit group of economically and politically powerful controlling shareholders,

38

GOHAR G. STEPANYAN

such as families, state, corporations or financial institutions, and managers (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2000; La Porta, Lopez-deSilanes, & Shleifer, 1999). Understanding these relationships and assessing whether they pose a threat to outside investors require intricate social, political, and institutional knowledge, which foreigners lack or find costly to obtain. As a consequence, firms with potentially problematic governance structures are particularly taxing to foreign investors in terms of their information and monitoring costs. Leuz et al. (2009) find that foreigners invest less in firms that reside in countries with weak disclosure requirements, securities regulations and poor shareholder protection, and have ownership structures that are conducive to governance problems and expropriation of outside investors by controlling managers and their families. In contrast, firms with substantial managerial and family control do not experience less foreign investment when they reside in countries with extensive disclosure requirements and strong investor protection. Information problems faced by foreign investors in firms with potentially problematic governance structures play a central role in this result. Stringent disclosure rules make it less costly to evaluate firms’ potential governance problems – they level the playing field among investors, making it less likely that locals have an informational advantage, whereas well-enforced investor protection rules make knowledge about private benefits of control and expropriation less important. In contrast, low disclosure requirements and weak investor protection worsen information problems and their consequences – thus, foreign investors face larger information problems relative to local investors in countries with low disclosure requirements, weakly enforced governance rules and poor investor protection. The findings on the importance of strong shareholder rights and legal framework and high-quality accounting and disclosure practices at the country level to attract foreign investment is in line with the evidence in La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1997, 2000), who emphasize the role of stronger investor protection laws and enforcement in fostering corporate governance climate that attracts outside investors. Interestingly, Aggarwal et al. (2005) present evidence that firm-level discretionary policies such as those related to greater accounting transparency and the issuance of an ADR, are also important in foreign institutional investors’ portfolio allocation choices. Their examination of U.S. mutual funds’ investment decisions in 1,280 firms across 30 emerging markets after the financial market crises of the 1990s highlights the important role of voluntary disclosure choices, including internationally recognized accounting standards, auditor

Financial Liberalization and Foreign Institutional Investors

39

quality, auditor opinion, and the use of consolidated financial statements, in mitigating emerging country’s institutional deficiencies to attract foreign institutional investment. In a related study, Bradshaw, Bushee, and Miller (2004) document that the use of accounting methods conforming to U.S. Generally Accepted Accounting Principles in non-U.S. firms is associated with higher level of investment by U.S. institutions and that this impact is significantly greater among non-U.S. firms already visible to U.S. investors. To summarize, poor firm-level governance deters foreign investment in countries with low transparency and weak property and investor rights that give rise to insider control and opaqueness. However, improvements in disclosure and governance practices at both the country and the firm level are potential levers to create an environment conducive to foreign investment.

Foreign versus Domestic Institutional Investors There appear to be some systematic patterns in the stock preferences of domestic and foreign institutional investors. Falkenstein (1996) and Gompers and Metrick (2001) uncover determinants of domestic institutional investors’ holdings. For example, Falkenstein (1996) finds that investment choices of U.S. open-end mutual funds are not totally driven by conventional proxies for risk alone; instead funds show significant preferences for various security characteristics related to firm visibility and security transaction costs, and are averse to stocks with low idiosyncratic volatility. Relevant to transaction costs, mutual funds show an aversion to stocks with low prices (priced less than $5) and preference for stocks with high liquidity. Funds also tend to avoid stocks with little information, as measured by the number of major newspaper articles or the number of months since the initial listing on the exchange. The breakdown of mutual funds by sector, age, and size suggests that mutual fund preferences are relatively consistent across all sectors except for firm size. The small cap funds subsector shows an affinity towards smaller firms with relatively low liquidity as opposed to the general preference towards larger stocks with higher liquidity. The sensitivity of mutual funds to proxies for transaction costs and information flows has an interesting implication. It suggests that mutual funds behave less like ‘‘insiders’’ and more like the general public. Individual investors, then, must be at a comparative disadvantage in trading illiquid stocks and stocks with low media coverage, unless some other

40

GOHAR G. STEPANYAN

investment group is holding disproportionate amounts of these illiquid, obscure stocks. Gompers and Metrick (2001) document that large institutional investors in the United States, a category encompassing banks, insurance companies, mutual funds, investment advisors, and other institutional money managers with at least $100 million under management, display stable preferences for large, liquid stocks with relatively low past returns. Institutional investors’ preference for stocks with large market capitalizations and liquid markets is explained by higher frequency of their trading and turnover of their portfolios that makes then more sensitive to the transactions costs caused by large-percentage bid-ask spreads for illiquid or low-priced stocks. An interesting finding of Gompers and Metrick (2001) is that large institutions in the United States are not momentum investors. The authors further find that the level of institutional ownership in a stock can help to forecast its future return, and provide evidence that this predictive power is due to demand shocks resulting from the compositional shift in ownership toward institutions. Large institutions nearly doubled their share of the U.S. common stock market from 1980 to 1996; by 1996 they controlled over half of the equity market. This compositional shift by itself accounts for a nearly 50% increase in the price of large-company stock relative to small-company stock and can explain part of the disappearance of the historical small-stock premium, originally documented by Banz (1981). The stock preferences of domestic institutional investors have been compared and contrasted with those of foreign institutional investors. For example, Dahlquist and Robertsson (2001) employ a firm-specific dataset of the holdings of foreign investors in Sweden, primarily foreign institutions, and compare it with the holdings of domestic mutual funds, other institutional investors, and individual investors to understand equity holdings in general, and the home bias puzzle in particular. This comparison makes it possible to discover the presence of a foreigner bias in stock picking while controlling for other effects, such as the possibility that institutional investors systematically deviate from holding the market portfolio. Studies of foreign institutional holdings reveal foreign investors’ preference for large firms, firms with the most liquid stocks, firms with large exports, firms with low ownership concentration, and firms that are listed abroad, consistent with the conjecture that asymmetric information may be an important factor explaining why foreign investors deviate from holding the market portfolio. However, Dahlquist and Robertsson (2001) document similar deviations from the market portfolio for domestic institutional investors. Domestic institutional investors also have a disproportionately higher ownership in large firms. The

Financial Liberalization and Foreign Institutional Investors

41

main differences between foreign and domestic institutions with respect to their preference for certain firm attributes is that foreigners prefer firms with high current ratios, low dividend yields and firms that are listed abroad, whereas domestic institutions invest relatively less in firms with high turnover and high idiosyncratic risk. These results suggest that (i) domestic investors seem to avoid the most risky firms, as compared to foreign investors, (ii) foreign investors might prefer firms paying low dividends due tax advantages in foreign jurisdictions for capital gains relative to dividend income, and (iii) foreigners appreciate firms with a high objective measure of financial strength such as large cash balances. Dahlquist and Robertsson (2001) conclude that the preference for large firms is not a pure foreigner bias and is common to all institutional investors. Covrig, Lau, and Ng (2006) investigate whether there are any differences in the investment behavior of foreign and domestic fund managers, as revealed by their portfolio holdings in 11 developed markets, and find strong evidence of both similar and differential stock preferences. Both groups of managers prefer stocks with high return on equity, large turnover, and low return variability. However, domestic fund managers favor firms that pay large dividends, have low financial distress and high growth potential, whereas foreign fund managers prefer to invest in corporations that are globally well known – firms with large market capitalization, large foreign sales, extensive analyst coverage, and whose stocks have foreign listings and index memberships. The demand for globally visible stocks by foreign managers is especially strong when their fund mandate is to achieve global or regional diversification, and becomes weaker when their stock holdings are concentrated mainly in a specific local market. The results also show no difference in the stock preferences of American-, European- and Asianbased funds. Overall, the evidence suggests that the differential mandates of fund managers influence their stock preferences, but not the geographic location of the fund itself (i.e., whether the fund is local or foreign).

THE ROLE OF FOREIGN INSTITUTIONAL INVESTORS The emergence of institutional investors as large equity owners in corporations worldwide offers the potential for increased monitoring of firm management. Institutions’ involvement can range from indirect influence such as threatening the sale of their shares to the active use of their ownership rights through voting or meetings with management. The

42

GOHAR G. STEPANYAN

involvement of large shareholders in monitoring or control activities has the potential to limit agency problems of managerial discretion, caused by the separation of ownership and control. Another important role served by large institutional investors is to act as a credible mechanism for transmitting information to other market participants. Foreign and more independent institutional investors are often credited with taking a more active role in prompting changes in many corporate governance practices, as other, less independent and pressure-sensitive institutions with business ties with corporations may feel compelled to be loyal to the management (Davis & Kim, 2007; Gillan & Starks, 2003). For example, Fidelity is reported to be more outspoken on governance issues in Europe, but is relatively passive in the United States, where it holds a substantial stake in at least 100 of the nation’s largest companies and controls over US$1.3 trillion in combined assets.8 Due to the increased globalization of their investments during the past decades and significance of their equity ownerships worldwide, foreign investors’ role in influencing firm governance and performance has become a subject of increased attention. Ferreira and Matos (2008) are the first to explore the monitoring role of institutional investors across 27 countries, using a comprehensive database of over 5,300 institutional investors’ holdings of stock of non-U.S. firms for the period of 2000–2005. They conjecture that the presence of foreign and independent institutions with large stakes enhances firm value through monitoring such as direct intervention to voice the interest of shareholders to corporate management (e.g., in proxy contests) or indirect monitoring through divestment of their investment in a company, thereby pushing up its cost of capital. These institutions are able to exert pressure because they have fewer business ties with firms, unlike domestic or ‘‘dependent’’ institutions. Overall, the authors document that ownership by foreign and independent institutions worldwide has a significantly positive impact on firm value and operating performance, confirming their involvement in monitoring corporations worldwide. Furthermore, the presence of foreign and independent institutions also reduces capital expenditures, suggesting that through their monitoring efforts, these groups of institutions seem to reduce managers’ incentives to overinvest. Ferreira, Massa, and Matos (2010) highlight the importance of institutional investors in cross-border M&As over the years 2000 through 2005. They find that cross-border M&As are more likely to occur in countries where foreign institutions hold a higher fraction of the local stock market, even after controlling for such important determinants of cross-

Financial Liberalization and Foreign Institutional Investors

43

border M&A activity as legal environment and economic development and growth. Foreign institutional ownership increases the likelihood that a merger deal is cross-border, successfully completed, and that the acquisition will result in full control of the target firm. This relation is more pronounced in countries with weaker legal institutions, lower shareholder protection, and less developed equity markets, where investors face higher transaction costs and information asymmetry, suggesting some substitutability between country-level governance and foreign investors in facilitating cross-border M&A transactions. The effect of foreign institutional ownership is also stronger in firms with higher information asymmetries (small and growth firms), less liquid shares, and large controlling shareholders, consistent with foreigners being better able to reduce the information gap in international takeovers and resist target firm management efforts to block deals. These results are direct evidence of the facilitation role played by foreign institutions in the international market for corporate control. Foreign institutions build bridges between firms from different countries to facilitate deals that involve negotiation processes between bidder and target with different regulatory and cultural environments, thus reducing bargaining and transaction costs associated with cross-border deals. Ferreira et al. (2010) conclude that cross-border portfolio investments of institutional money managers and cross-border M&As are complementary mechanisms in promoting financial integration worldwide. Increased foreign institutional investment has become an important influence in many economies, particularly emerging economies, as the demand for capital in these countries has increased. On the one hand, firms and countries may be motivated to improve their corporate governance in order to attract foreign capital. On the other hand, increased investment by foreign institutions may act as a catalyst to improve the governance climate of emerging market firms and their growing economies. Aggarwal, Erel, Ferreira, and Matos (2011) study the role of international institutional investors as a channel for promoting better governance and convergence in corporate governance practices around the world. Their main finding is that firm-level governance is positively associated with institutional ownership and foreign institutional ownership, in particular. Panels A, B, and C of Fig. 3, extracted from their study, clearly demonstrate the relationship between the overall governance climate in a country and international institutional investment across 23 developed and developing economies. Further, Aggarwal et al. (2011) find that changes in international institutional ownership over time drive subsequent changes in firm-level governance, but the opposite is not true. This evidence establishes a direct

Fig. 3. International Institutional Investment and Governance. Panel A Shows the Average of the Firm-Level Governance Index by Country and Year in 2004–2008. Governance Index is the Percentage of the 41 Governance Attributes Related to Board, Audit, Anti-Takeover and Compensation/Ownership Provisions that a Firm Meets. An Index of 100% Means that a Firm has Adopted All 41 Governance Provisions. Panel B Shows the Average Total Institutional Ownership by Country and Year in 2003–2007. Institutional Ownership is the Sum of the Holdings of All Institutions in a Firm’s Stock, as a Fraction of Its Year-End Market Capitalization. Panel C Shows the Average Institutional Ownership by Foreign and Domestic Institutions at the End of 2007. Domestic (Foreign) Institutional Ownership is the Sum of the Holdings of All Institutions Domiciled in the Same Country (in a Different Country) in Which the Stock is Listed, as a Fraction of Its Year-End Market Capitalization. Source: Aggarwal et al. (2011).

Financial Liberalization and Foreign Institutional Investors

45

link between international portfolio investment and the adoption of better corporate governance practices that promote corporate accountability and empower shareholders worldwide, supporting the view that institutions are not simply attracted to firms with stronger governance but also play an important monitoring role. Foreign institutions and, more generally, institutions from countries with strong shareholder protection are the main promoters of good governance practices outside of the United States. The extent of shareholder protection in the country where the portfolio firm is located also matters. Domestic institutions are the main drivers of governance improvements in common-law countries, but in civil-law countries the main drivers are foreign institutions. In countries with a weaker legal environment, domestic institutional money managers are more likely to have business ties to local corporations, share the benefits of control, and be more sympathetic to incumbent management. In contrast, foreign institutions seem to be able to exert pressure over local management. Aggarwal et al. (2011) conclude that there is a positive association between ‘‘governance at home’’ of institutional investors and firm-level governance of their stock holdings abroad, indicating that institutions export good corporate governance practices across countries. Institutional investors affect both which corporate governance mechanisms are in place and the outcomes. Firms with higher institutional ownership are more likely to terminate poorly performing CEOs, whereas firms with higher foreign institutional ownership also exhibit improvements in valuation over time. Foreign institutional ownership is also associated with a more shareholder-friendly board structure by making it more likely that the board has a majority of independent directors and an appropriate number of directors, and that it does not adopt a staggered board provision. Thus, foreign institutions take a lead role in improving governance and shareholder activism that local investors seem unable to take outside of the U.S. Monitoring and activism by independent institutions travel beyond country borders and lead to better firm performance. Aggarwal et al. (2011) highlight the importance of market forces, institutional investors in particular, in promoting good corporate governance practices around the world beyond the effect of government regulations.

CONCLUSION Elimination of capital controls and restrictions on foreign investment in many countries around the globe resulted in a surge in cross-border capital

46

GOHAR G. STEPANYAN

flows. Opening domestic stock markets to foreign investors is associated with several potential benefits for the emerging and developing market economies: (i) reduction in financing constraints, (ii) attraction of new capital to finance economic growth, and (iii) advancement of wellfunctioning financial and equity markets. Foreign equity flows also promote global diversification of risk, reducing the cost of equity finance and increasing firm valuations. In addition, foreign investors demand greater transparency and improved disclosure practices that are crucial for efficient allocation of capital. They insist on accountability of management and better protection of shareholder rights to reduce the risk of expropriation by controlling investors. The policy makers of those economies, however, are cautious in contemplating the benefits of increased liberalization against various uncertainties. One area of common concern is the regulation of transient, so-called ‘‘hot’’ money – international funds that are very sensitive to differences in interest rates, expectations of future economic growth, and returns from holding securities. Even a small shock to the emerging market economies can lead to volatile movements in these investment flows, magnifying the impact of the shock and destabilizing domestic equity markets and currencies. For instance, floods of ‘‘hot’’ money may put upward pressure on the domestic currency, reducing export competitiveness and threatening the country’s competitive position in the global marketplace.9 Even academics question the desirability of free flows of capital and full liberalization of capital markets, citing China’s success with restraints on short-term capital movements (Stiglitz, 2000). The emerging consensus, however, is that the overall benefit of foreign investment is likely to outweigh the perceived risks associated with free flow of capital. Even though there may be some truth to the claim that recent crises were in part a consequence of irresponsible behavior of foreign portfolio investors, the academic evidence to date does not seem to support such claim. If anything, there is more evidence in support of foreign investors contributing towards more efficient risk sharing and resource allocation, development of domestic capital markets, and improvements in the structure of external finance. Errunza (2001) blames the imprudent and unsustainable policy rules in many emerging countries for their economic and financial difficulties. Kim and Singal (2000) also argue that the act of market liberalization itself was not responsible for the emerging market financial crises of the late 1990s. Economic mismanagement and value destruction in the corporate sector were among likely causes.

Financial Liberalization and Foreign Institutional Investors

47

Kaminsky and Schmukler (2008) reconcile the two apparently conflictive views on financial liberalization by pointing out to the time-varying effects of liberalization. Financial liberalization in emerging markets fuels financial instability only in the short run due to a variety of political and business distortions that pervade these markets; in the long run, institutions improve and financial markets stabilize. At the core of this time-varying relation is the interaction between financial deregulation and the reform of government policies and institutions. Since deregulation takes place in many economies with weak institutions, agency problems might fuel risky behavior and crises in the initial stages of liberalization process. As liberalization persists, it triggers reforms that improve institutions and stabilize capital markets. Stulz (2005) also highlights that international financial integration can improve the functioning of domestic markets and institutions by alleviating the ‘‘twin agency problems’’ prevalent in many countries around the globe – the incentives of rulers of sovereign states and corporate insiders to expropriate outside investors. This explains why mature markets with better government institutions did not experience large crashes in the aftermath of liberalization reforms. Thus, regulators of emerging markets should first improve domestic institutions, governance, and macroeconomic fundamentals, and then deregulate domestic financial and capital markets to avoid the short-run pain of rushed liberalization reforms. One interesting avenue for further inquiry is what role foreign institutional investors played in shaping newly liberalized emerging market economies’ policy reforms directed at strengthening financial and regulatory systems.

NOTES 1. Organization for Economic Co-Operation and Development, Investment News, November 2010. 2. Patterns of domestic and foreign institutional ownership may be explained by regulatory constraints on investment behavior of certain types of institutional investors, country-level corporate governance environment, market characteristics (bank- versus market-centered economy, the degree of financial market development), or institutional investment industry characteristics such as size and concentration (Li et al., 2006, see also Tesar & Werner, 1995). 3. If stock market liberalization reduces the aggregate cost of equity capital then, holding expected future cash flows constant, we should observe an increase in an

48

GOHAR G. STEPANYAN

emerging country’s equity price index when the market learns that stock market liberalization is going to occur. See Stulz (1999a, 1999b) for an extensive survey of this literature. 4. Liberalizations may be strategically timed by policy makers to coincide with (i) high points in the world business cycle, (ii) the contemporaneous implementation of other economic reforms (macroeconomic stabilization programs, trade reforms, privatization programs, and easing of exchange controls), and (iii) positive shocks to aggregate investment demand or to marginal productivity of capital. 5. See, for example, Levy and Sarnat (1970), Grauer and Hakansson (1987), and DeSantis and Gerard (1997), among others, on the gains from international diversification of investment portfolios. 6. According to Dahlquist and Robertsson (2001), the representative foreign investor in Sweden is a large institution. Similarly, Grinblatt and Keloharju (2000) report that foreign investors in the Finish stock market are often well capitalized financial institutions with a long history of successful investment in other stock markets such as professionally managed mutual funds, hedge funds, and investment banks. 7. Cross-listing in the U.S. necessitates many substantive changes related to information disclosure in accordance with SEC regulations and the rules/regulations of the stock exchange, thus, lowering the direct and indirect barriers to international investment for many foreign investors. 8. Fidelity’s Divided Loyalties, Business Week, October 16, 2006. 9. G-20 to Address Hot Money, Wall Street Journal, February 12, 2011.

REFERENCES Aggarwal, R., Erel, I., Ferreira, M., & Matos, P. (2011). Does governance travel around the world? Evidence from institutional investors. Journal of Financial Economics, 100, 154–181. Aggarwal, R., Klapper, L., & Wysocki, P. (2005). Portfolio preferences of foreign institutional investors. Journal of Banking and Finance, 29, 2919–2946. Banz, R. (1981). The relationship between return and market value of common stocks. Journal of Financial Economics, 9, 3–18. Bekaert, G., & Harvey, C. (2000). Foreign speculators and emerging equity markets. Journal of Finance, 55, 565–613. Bekaert, G., Harvey, C., & Lumsdaine, R. (2002). Dating the integration of world equity markets. Journal of Financial Economics, 65, 203–247. Bekaert, G., Harvey, C., & Lundblad, C. (2005). Does financial liberalization spur growth? Journal of Financial Economics, 77, 3–55. Bohn, H., & Tesar, L. (1996). U.S. equity investment in foreign markets: Portfolio rebalancing or return chasing? American Economic Review, 86, 77–81. Bradshaw, M., Bushee, B., & Miller, G. (2004). Accounting choice, home bias, and U.S. investment in non-U.S. firms. Journal of Accounting Research, 42, 795–841. Brennan, M., & Cao, H. (1997). International portfolio investment flows. Journal of Finance, 52, 1851–1880.

Financial Liberalization and Foreign Institutional Investors

49

Chan, K., Covrig, V., & Ng, L. (2005). What determines the domestic bias and foreign bias? Evidence from mutual fund equity allocations worldwide. Journal of Finance, 60, 1495–1534. Choe, H., Kho, B., & Stulz, R. (1999). Do foreign investors destabilize stock markets? The Korean experience in 1997. Journal of Financial Economics, 54, 227–264. Cooper, I., & Kaplanis, E. (1994). Home bias in equity portfolios, inflation hedging, and international capital market equilibrium. Review of Financial Studies, 7, 45–60. Covrig, V., Lau, S., & Ng, L. (2006). Do domestic and foreign fund managers have similar preferences for stock characteristics? A cross-country analysis. Journal of International Business Studies, 37, 407–429. Dahlquist, M., & Robertsson, G. (2001). Direct foreign ownership, institutional investors, and firm characteristics. Journal of Financial Economics, 59, 413–440. Davis, G., & Kim, E. H. (2007). Business ties and proxy voting by mutual funds. Journal of Financial Economics, 85, 552–570. DeSantis, G., & Gerard, B. (1997). International asset pricing and portfolio diversification with time-varying risk. Journal of Finance, 52, 1881–1912. Errunza, V. (2001). Foreign portfolio equity investments, financial liberalization, and economic development. Review of International Economics, 9, 703–726. Falkenstein, E. (1996). Preferences for stock characteristics as revealed by mutual fund portfolio holdings. Journal of Finance, 51, 111–135. Ferreira, M., Massa, M., & Matos, P. (2010). Shareholders at the gate? Institutional investors and cross-border mergers and acquisitions. Review of Financial Studies, 23, 601–644. Ferreira, M., & Matos, P. (2008). The colors of investors’ money: The role of institutional investors around the world. Journal of Financial Economics, 88, 499–533. French, K., & Poterba, J. (1991). Investor diversification and international equity markets. American Economic Review, 81, 222–226. Froot, K., O’Connell, P., & Seasholes, M. (2001). The portfolio flows of international investors. Journal of Financial Economics, 59, 151–193. Gillan, S., & Starks, L. (2003). Corporate governance, corporate ownership, and the role of institutional investors: A global perspective. Journal of Applied Finance, 13, 4–22. Gompers, P., & Metrick, A. (2001). Institutional investors and equity prices. Quarterly Journal of Economics, 116, 229–259. Grauer, R., & Hakansson, N. (1987). Gains from international diversification: 1968–85 Returns on portfolios of stocks and bonds. Journal of Finance, 42, 721–739. Grinblatt, M., & Keloharju, M. (2000). The investment behavior and performance of various investor types: A study of Finland’s unique data set. Journal of Financial Economics, 55, 43–67. Hau, H., & Rey, H. (2008). Home bias at the fund level. American Economic Review, 98, 333–338. Henry, P. (2000a). Stock market liberalization, economic reform, and emerging market equity prices. Journal of Finance, 55, 529–564. Henry, P. (2000b). Do stock market liberalizations cause investment booms? Journal of Financial Economics, 58, 301–334. Kaminsky, G., & Schmukler, S. (2008). Short-run pain, long-run gain: Financial liberalization and stock market cycles. Review of Finance, 12, 253–292. Kang, J., & Stulz, R. (1997). Why is there a home bias? An analysis of foreign portfolio equity ownership in Japan. Journal of Financial Economics, 46, 3–28.

50

GOHAR G. STEPANYAN

Karolyi, A. (2002). Did the Asian financial crisis scare foreign investors out of Japan? Pacific Basin Finance Journal, 10, 411–442. Karolyi, A., & Stulz, R. (2003). Are financial assets priced locally or globally? In: G. Constantinides, M. Harris, & R. Stulz (Eds), Handbook of the economics of finance (Vol. 1, pp. 975–1020). Amsterdam, The Netherlands: North Holland. Kim, E. H., & Singal, V. (2000). Stock market openings: Experience of emerging economies. Journal of Business, 73, 25–66. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1997). Legal determinants of external finance. Journal of Finance, 52, 1131–1150. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (2000). Investor protection and corporate governance. Journal of Financial Economics, 58, 3–27. La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. (1999). Corporate ownership around the world. Journal of Finance, 54, 471–517. Leuz, C., Lins, K., & Warnock, F. (2009). Do foreigners invest less in poorly governed firms? Review of Financial Studies, 22, 3245–3285. Levy, H., & Sarnat, M. (1970). International diversification of investment portfolios. American Economic Review, 60, 668–675. Lewis, K. (1999). Trying to explain home bias in equities and consumption. Journal of Economic Literature, 37, 571–608. Li, D., Moshirian, F., Pham, P., & Zein, J. (2006). When financial institutions are large shareholders: The role of macro corporate governance environments. Journal of Finance, 61, 2975–3007. Stiglitz, J. (2000). Capital market liberalization, economic growth, and instability. World Development, 28, 1075–1086. Stulz, R. (1999a). Globalization, corporate finance and the cost of capital. Journal of Applied Corporate Finance, 12, 8–25. Stulz, R. (1999b). International portfolio flows and security markets. In: M. Feldstein (Ed.), International capital flows (pp. 257–293). Chicago, London: University of Chicago Press. Stulz, R. (2005). The limits of financial globalization. Journal of Finance, 60, 1595–1638. Tesar, L., & Werner, I. (1995). Home bias and high turnover. Journal of International Money and Finance, 14, 467–492.

INSTITUTIONAL INVESTMENT HORIZON AND FIRM CREDIT RATINGS Najah Attig, Sadok El Ghoul and Omrane Guedhami STRUCTURED ABSTRACT Purpose – Study the impact of the heterogeneity of institutional investors, evident in their investment horizon, on firm credit ratings. Methodology/approach – Use a large sample of U.S. firms over the period from 1985 to 2006 (20,670 U.S. firm-year observations) to empirically investigate the relationship between institutional investment horizon and firm credit ratings. Test whether institutional investors with long-term investment horizon are associated with important monitoring and informational roles and thus higher credit ratings. Findings – Stable shareholdings and relationship investing of institutional investors contribute to their monitoring and informational roles and result in higher firm credit ratings. Namely, ownership stakes of long-term institutional investors are associated with higher firm credit ratings than those of short-term institutional investors. In addition, the predominance and number of institutional investors with a long-term investment horizon affect firm’s agency costs and information quality. Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 51–82 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012005

51

52

NAJAH ATTIG ET AL.

Social implications – Institutional monitoring incentives seem to be susceptible to the heterogeneity of institutional investors. The results point to the benefits of the long-term investment horizon of institutional investors (beyond their shareholdings) that seem to be associated with more efficient monitoring and thus reduced managerial myopia and opportunism. Originality/value of the chapter – This is the first work to provide evidence on the extent to which the heterogeneity of institutional investors, evident in their investment horizon, alters firm’s credit ratings. Keywords: Institutional investors; investment horizon; credit ratings; corporate governance Jel classifications: G34; G23

INTRODUCTION Research on institutional investor heterogeneity has gained momentum in the recent governance literature. An important dimension on which incentives for institutional activism can be distinguished is investment horizon. Prior literature suggests two competing views on the extent to which the investment horizon of institutional investors plays a governance role by affecting monitoring and information asymmetry. With respect to the monitoring role, one can advance the argument that a long-term investment horizon is likely to be associated with ‘‘relationship investing’’ (Chidambaran & John, 2000). Such investing, combined with large institutional holdings, mitigates the free-rider problem and thus gives rise to more efficient monitoring, which leads managers to display less myopic and self-interested behavior (e.g., overpay and perquisites) and results in positive externalities to firm stakeholders. With respect to the informational role, one can conjecture that, ceteris paribus, long-term institutional investors have an informational advantage that stems from their ability to invest more resources in the investee firms, engage in quality research, and collect and process corporate information more efficiently than short-term institutional investors. Alternatively, one may maintain that institutions with short-term investment horizons have superior ability to gather and process information to generate short-term benefits (or avoid future losses) rather than engaging in value-enhancing but costly monitoring. As a result,

Institutional Investment Horizon and Firm Credit Ratings

53

the presence of institutions with short-term investment horizon (i.e., frequent trading by institutional investors) may reflect their tendency to ‘‘vote with their feet’’ (e.g., Parrino, Sias, & Starks, 2003), which may exert some disciplinary pressure on managers. Extant evidence on the governance role of institutional investors’ investment horizon (IIIH) is mixed. For instance, Gaspar, Massa, and Matos (2005) show that institutions with short horizons are associated with weak monitoring and a weak bargaining position in acquisitions because they tend to focus on short-term performance. Chen, Harford, and Li (2007) show that (independent) institutions with long-term investments exert more effective monitoring, as reflected in better post-merger performance. However, Yan and Zhang (2009) find that, in contrast to long-term institutional trading, short-term institutional trading is positively related to future stock returns and future earnings surprises and hence conclude that short-term institutions are better informed than long-term institutions. In this chapter, we contribute to this emerging line of research by examining the extent to which IIIH affects firm credit ratings. Our interest in examining firm credit ratings is prompted by the role of credit ratings in mirroring firms’ disclosure and information quality (e.g., Sengupta, 1998) and the role of firm credit ratings in corporate financing and investment decisions (e.g., Blume, Lim, & Mackinlay, 1998). Furthermore, we focus on firm-level credit ratings rather than specific debt issue ratings because the former reflect the overall default risk of the company, as opposed to the default risk associated with a single bond issue, and thus are more likely to be affected by corporate governance (Weber, 2006). Our findings are consistent with long-term institutional investors playing important monitoring and informational roles as evidenced by higher firm credit ratings. More specifically, in univariate and multivariate analyses, we first find that institutional investors with a long-term investment horizon are associated with higher firm credit ratings. This result implies that institutional investors’ monitoring incentive is susceptible to their heterogeneity, as evidenced in their different investment horizons. In additional tests, we find that the predominance and number of institutional investors with a long-term investment horizon affects a firm’s agency costs and information quality. This result is consistent with Foster and Viswanathan (1996) and Back, Cao, and Willard (2000), who find that information about the firm is revealed more rapidly when the number of institutional investors is large, particularly the number of long-term institutional investors. Our results are robust to a wide array of controls and a number of alternative specifications.

54

NAJAH ATTIG ET AL.

Our examination of the impact of IIIH on firm credit ratings makes a contribution to several strands of the literature. First, we contribute to the large and longstanding literature on the governance role of institutional shareholdings. Our finding that institutional investors’ monitoring incentive varies with their investment horizon implies that a focus on institutional shareholdings in aggregate is likely to mask variation in institutions’ governance role and thus provides a potential explanation for the mixed evidence on the governance role of institutional ownership. Second, we contribute to the growing literature that examines the extent to which corporate governance affects firm credit and bond ratings. Bhojraj and Sengupta (2003) and Mansi, Maxwell, and Miller (2004) are the first to provide direct evidence on the impact of corporate governance on corporate debt ratings. More recently, Ashbaugh-Skaife, Collins, and LaFond (2006) investigate the role of different governance mechanisms on firm credit ratings. This chapter adds to these studies by presenting the first evidence to our knowledge on the impact of institutional investment horizon on firm credit ratings. It also extends the literature that examines the effect of shareholder heterogeneity as discussed in previous studies (e.g., Bagwell, 1991; Bushee, 2001; Chen et al., 2007; Gaspar et al., 2005). Equally important, and more broadly, we provide evidence that, through their stable shareholdings and relationship investing, institutional investors play a significant role in today’s context of global capital markets, not only by holding and managing financial assets but also by financing corporate growth opportunities and disciplining management. Our evidence on the impact of IIIH on firm credit ratings has a number of implications for practitioners. First, the chapter’s insights on the potential impact of IIIH on firm creditworthiness and financing costs should be informative to managers about the benefits of attracting and retaining institutional investors with a long-term investment horizon. Second, because investors use firm credit ratings in making investment decisions, our evidence should improve investors’ decision-making by helping them develop a better understanding of the monitoring role and informational advantage of institutional investors and their heterogeneity. Third, our findings point to ways regulators can improve corporate governance and thereby enhance investor protection and confidence in capital markets. In this context, and in light of recent developments in the global financial system, our findings suggest that long-term horizon of institutional shareholdings is likely to provide systemic safeguards to corporations and, thus, highlight the need for securities regulators to provide incentives to institutional investors to increase the stability of their corporate ownership.

Institutional Investment Horizon and Firm Credit Ratings

55

The rest of the chapter proceeds as follows. In the next section, we review related studies and describe our main hypothesis. Then we describe our data construction and explain our research design. Results are then discussed and conclusions drawn.

RELATED LITERATURE AND HYPOTHESIS Although a voluminous body of literature has contributed greatly to our understanding of the monitoring role of financial institutions, empirical evidence on the role of institutional activism is mixed.1 On a prima facie basis, institutional investors’ large shareholdings and informational advantage (compared to individual investors) might be expected to lead to efficient monitoring (e.g., Shleifer & Vishny, 1986). A competing view, however, suggests that myopic behavior of institutional investors can reduce their monitoring incentive and lead to passivism in corporate governance (e.g., Del Guercio & Hawkins, 1999; McConnell & Servaes, 1990; Smith, 1996). In this chapter, we argue that the lack of unequivocal evidence on institutional activism is likely due to heterogeneity among financial institutions differentially shaping their governance roles. Institutional investors differ in a number of respects. For example, institutions differ in their investment objectives and styles, legal restrictions, and competitive pressures. Such differences are likely to affect their informational and monitoring roles. Brickley, Lease, and Smith (1988), for instance, find that pressure-resistant institutions are more likely to oppose firm managers than pressure-sensitive institutions when voting on antitakeover amendments. A new yet burgeoning line of research suggests that the informational role of institutional investors also depends on their investment horizon. Yan and Zhang (2009) find that, in contrast to long-term institutions’ trading, short-term institutions’ trading is positively related to future stock returns and future earnings surprises, suggesting that short-term institutions are better informed than long-term ones. However, Gaspar et al. (2005) show that institutions with short horizons are associated with weak monitoring and a weak bargaining position in acquisitions due to a focus on short-term performance, and Chen et al. (2007) show that concentrated holdings by independent long-term institutions are associated with more effective monitoring as reflected in better post-merger performance.2 Similarly, Burns, Kedia, and Lipson (2010) show that the presence of short-horizon institutional investors degrades the quality of financial information, as reflected in higher levels of discretionary accruals.

56

NAJAH ATTIG ET AL.

Two viewpoints guide us in our investigation of the governance role of IIIH, as evidenced by its impact on firm credit ratings. The first relates to the monitoring role of IIIH. Long-term horizon investors (LTHI) are likely to be more actively involved in managerial monitoring than short-term horizon investors (STHI), for a number of reasons. First, long-term institutional investors engage in ‘‘relationship investing’’ (Chidambaran & John, 2000), which, in turn, increases their incentive to engage in monitoring rather than selling shares. Indeed, the longevity of their investments provides additional incentive to LTHI to engage in costly monitoring in an effort to safeguard their investments.3 Second, compared to STHI, LTHI are likely to have larger shareholdings,4 which can mitigate the free-rider problem and provide LTHI with more incentive to exert monitoring on managers. Third, the combined effect of the longevity and the size of their shareholdings is likely to worsen the adverse impact of the threat of LTHI exit, which will make the managers more receptive to LTHI monitoring to prevent the price impact of LTHI exit (e.g., Admati & Pfleiderer, 2009). Fourth, given higher ‘‘exit’’ costs (e.g., large blockholdings, high transaction costs, tax timing, and rebalancing costs) for LTHI, short-term trading is less attractive and monitoring becomes more desirable (Hirschman, 1970). Fifth, as a result of their indexing strategies, some LTHI tend to have stable and diversified stakes in nearly all the largest publicly traded firms (Hawley, 1995), which we expect to decrease incentive to trade frequently and increase governance commitment. A sixth argument for the monitoring role of LTHI stems from their ability to influence the outcome of M&A deals. Gaspar et al. (2005), for instance, argue that shareholders with a long-term investment horizon have greater bargaining power in M&A deals and can afford to stay in the firm until all benefits of the acquisition are realized. They find that target firms with STHI are more likely to receive acquisition bids with lower acquisition premiums and experience lower abnormal returns around the merger announcement than target firms with LTHI. Notwithstanding the various arguments above in support of the efficient monitoring role of LTHI, their governance role may be limited to the extent that over time they exhibit loyalty to managers. In line with Pound (1988), who argues that institutional investors have a tendency to vote in favor of entrenched management, relationship investing may evolve into a mutually beneficial business relationship, in which case LTHI are likely to become pressure-sensitive (Brickley et al., 1988) and go along with management. Institutional monitoring by long-term investors can also be limited by their sensitivity to liquidity shocks (Admati & Pfleiderer, 2009) and with the costs associated with exit making a threat to exit less credible. Finally, if liquidity

Institutional Investment Horizon and Firm Credit Ratings

57

costs are low, the temptation to ‘‘vote with their feet’’ may ultimately trump LTHI incentives to monitor managers (Bhide, 1993).5 The second viewpoint guiding our investigation relates to the informational role of IIIH. Prior work shows that institutional investors are likely to be more informed than individual investors. For instance, Sengupta (1998) and Healy, Hutton, and Palepu (1999) document a positive association between institutional shareholdings and financial analysts’ ratings of overall corporate disclosure practices. Extending this line of research, we conjecture that the investment horizon of institutional investors shapes their informational role. First, as we discussed earlier, LTHI are likely to have an information advantage due to relationship investing. For instance, as a result of their relationship investing, LTHI may have more precise information and will revise their beliefs less frequently (Chen et al., 2007). By virtue of their ability to convey private information (e.g., through infrequent but informed trading), LTHI – through their threat to exit – can pressure management to improve information quality (e.g., through better disclosure).6 Similarly, the presence of LTHI is likely to convey information about management’s commitment to long-term prospects and thus be associated with less uncertainty. An alternative view, however, suggests that STHI may trade frequently to exploit their informational advantage (e.g., Wermers, 2000). This is particularly true in a liquid market such as the United States, where STHI can maintain the liquidity of their holdings and offload ownership blocks without depressing stock prices.7 On the contrary, the informational role of STHI might be limited to the extent that they trade on the basis of noise (Yan & Zhang, 2009) or imperfect short-term informational signals (Gaspar et al., 2005). In summary, the discussion earlier suggests that institutional investment horizon is likely to have a governance role through its effects on monitoring and information asymmetry. However, whether a long-term investment horizon plays a more pronounced governance role than a short-term investment horizon appears to be an empirical question. The focus of this chapter is thus to discriminate the effects of IIIH on governance as evidenced by firm credit ratings. To this end, and for expository convenience, we conjecture that institutional investors with a long-term horizon are more likely to enhance governance quality as reflected by better firm credit ratings than institutional investors with a short-term horizon. This hypothesis is based on Bhojraj and Sengupta (2003), Mansi et al. (2004), and Ashbaugh-Skaife et al. (2006), who show that corporate governance can reduce default risk and thus result in better firm credit and bond ratings, by alleviating agency costs and information asymmetry problems.8 We focus here on firm credit ratings rather than bond ratings, as the former reflect a firm’s overall default risk and

58

NAJAH ATTIG ET AL.

creditworthiness and thus are more likely to be affected by corporate governance (Weber, 2006). Taken together, we posit that through their monitoring and informational roles, LTHI are more influential in reducing agency risk and information asymmetry, and thus produce better firm credit ratings, than STHI. More formally, H1. Institutional investors with a long-term investment horizon have a more influential monitoring and informational role than institutional investors with a short-term investment horizon, as evidenced by firm credit ratings.

DATA AND RESEARCH DESIGN Sample Selection To empirically investigate the relationship between institutional investment horizon and firm credit ratings, we combine data from three sources: CDA/ Spectrum (Thomson 13-F data), which provides data on institutional ownership; COMPUSTAT, which provides credit ratings and financial statement data; and the Center for Research in Security Prices (CRSP), which provides stock return data. We restrict our analysis to firms with total institutional ownership in the [0, 100%] interval. To construct firm-specific controls for our regressions, we require firms to have total assets and long-term debt in COMPUSTAT and at least 24 monthly stock returns during the previous five years in CRSP. Our sample period starts in 1985 because credit ratings become available in COMPUSTAT as of that year. These screens yield an unbalanced panel of 20,670 U.S. firm-year observations over the period from 1985 to 2006. Firm Credit Ratings To capture the governance role of IIIH, we examine its impact on firm credit ratings. Following related studies (Ashbaugh-Skaife et al., 2006; Bhojraj & Sengupta, 2003; Blume et al., 1998; Mansi et al., 2004, among others), we construct our measure of firm credit ratings (RATING) by converting the long-term issuer credit ratings compiled by Standard & Poor’s (S&P) and reported in COMPUSTAT to an ordinal scale. More specifically, we assign a value of 8 if the firm has an S&P rating of AAA, 7 if AA, 6 if A, 5 if BBB, 4 if BB, 3 if B, 2 if CCC, and 1 if CC. Table 1 reports the sample breakdown

59

Institutional Investment Horizon and Firm Credit Ratings

Table 1.

Sample Breakdown by S&P Credit Rating and Year.

Year

1

2

3

4

5

6

7

8

Total

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

1 0 0 0 0 3 6 7 1 0 0 1 0 1 0 1 4 3 0 1 1 1

0 33 28 18 14 15 15 11 6 11 11 14 14 13 20 22 25 30 23 18 17 14

26 182 198 170 145 108 101 110 131 151 168 198 219 235 238 251 221 204 213 226 222 214

18 159 163 145 141 125 128 149 191 201 217 238 262 290 293 292 292 317 333 339 329 317

37 167 167 163 171 174 179 203 218 242 254 274 305 323 331 328 326 317 316 326 318 303

47 208 206 209 202 189 204 202 219 217 236 251 254 255 244 215 198 189 183 174 176 157

28 94 92 83 83 85 83 79 75 75 70 65 54 54 44 39 34 24 21 19 18 20

6 19 18 18 16 15 14 17 14 13 12 15 14 11 13 12 11 9 9 9 8 8

163 862 872 806 772 714 730 778 855 910 968 1,056 1,122 1,182 1,183 1,160 1,111 1,093 1,098 1,112 1,089 1,034

Total

31

372

3,931

4,939

5,442

4,435

1,239

281

20,670

Notes: This table presents the number of issuer ratings by year. S&P’s long-term issuer credit ratings are converted to an ordinal scale according to the following schedule: 8 (AAA), 7 (AA), 6 (A), 5 (BBB), 4 (BB), 3 (B), 2 (CCC), and 1 (CC). The sample consists of 20,670 firm-year observations between 1985 and 2006.

by S&P credit rating and year. We can see from the table that the observations are fairly evenly dispersed over the sample period, but less evenly dispersed across credit rating scores; for instance, firms rated A, BBB, and BB represent about two-thirds of the sample. Interestingly, the BBB and BB credit ratings display a fairly pronounced increasing pattern over time, in contrast to the credit ratings of AAA and AA. Institutional Investors’ Horizon To construct our proxies for IIIH, we begin by following Gaspar et al. (2005) and calculate institutional investors’ churn rate. Institutional investors with more than $100 million in equity securities are required to

60

NAJAH ATTIG ET AL.

report their quarter-end holdings in 13F filings with the Securities and Exchange Commission (SEC). We use the CDA/Spectrum database, which compiles 13F filings, to compute the quarterly churn rate of institutional investor k in quarter t as follows: PN k;t jSk;i;t Pi;t  Sk;i;t1 Pi;t1  S k;i;t1 DPi;t j (1) CRk;t ¼ i¼1 PN k;t S P S P k;i;t

i¼1

i;t

k;i;t1

i;t1

2

where Nk,t is the number of firms included in institutional investor k’s portfolio in quarter t; Sk,i,t is the number of firm i’s shares held by institutional investor k in quarter t; and Pi,t is firm i’s share price in quarter t. The churn rate measures the frequency with which an institutional investor alters its position in the firm’s stock. A higher churn rate therefore indicates a shorter investment horizon. We next calculate institutional investor k’s average churn rate over the past four quarters: AVG_CRk;t ¼

4 1X CRk;trþ1 4 r¼1

(2)

Our first proxy for IIIH is the weighted average of the churn rates of a firm’s institutional investors: WACRj;t ¼

M j;t X

wk;i;t AVG_CRk;t

(3)

k¼1

where wi,j,t is the percentage ownership of institutional investor k in firm j, and Mj,t is the number of institutional investors in firm j. Our second and third proxies for IIIH are the firm’s percentage ownership held by long-term and short-term institutional investors, that is, long-term institutional ownership (LTIO) and short-term institutional ownership (STIO), where we define long-term (short-term) investors as investors whose average churn rates are in the bottom (top) tercile. Finally, we capture the preponderance of long-term institutional investors using an indicator variable for whether long-term institutional ownership is higher than short-term institutional ownership (D(LTIOWSTIO)). Research Design To analyze the impact of IIIH on firm credit ratings, we run a model that represents credit ratings as a function of firm characteristics and corporate governance attributes. In particular, we estimate several specifications of the

Institutional Investment Horizon and Firm Credit Ratings

61

following ordered probit because the eight credit rating categories provide ordinal risk assessments (subscripts suppressed for notational convenience): RATING ¼ a þ b1 IIIH þ b2 Z þ Industry & year effects þ e

(4)

where RATING is our measure of a firm’s credit rating, IIIH is one of our four proxies for institutional investment horizon, and Z is a vector of control variables routinely used in firm rating analysis (Ashbaugh-Skaife et al., 2006; Bhojraj & Sengupta, 2003; Blume et al., 1998; Mansi et al. 2004, among others): SIZE, the natural logarithm of total assets in millions of U.S. dollars; COVERAGE, the ratio of earnings before interest and taxes plus interest expense divided by interest expense; MARGIN, the ratio of operating income to sales; LEVERAGE, the ratio of long-term debt to total assets; CAPINT, the ratio of property, plant, and equipment to total assets; BETA, the stock return beta measured over the fiscal year, estimated using Dimson’s (1979) model with one lag and one lead of the CRSP valueweighted index; LOSS, an indicator variable set to 1 if net income before extraordinary items is negative in the current and previous years, and 0 otherwise. More detailed variable definitions are provided in the appendix. Table 2 reports descriptive statistics for the key variables used in this study. The mean (median) RATING is 4.70 (5.00). The mean (median) institutional ownership is 55% (56%). Long-term institutional shareholders own on average 14% of their portfolio firms’ shares, compared to 17% for short-term institutional investors. Furthermore, long-term institutional ownership is higher than short-term institutional ownership in only 43% of the firm-year observations. Table 3 provides the pairwise correlation coefficients between all the key variables used in this study. Generally, the pairwise correlation coefficients among the control variables are low, especially those between our test variables. As expected, institutional ownership (IO) displays substantial correlations with both long-term institutional ownership (LTIO) and shortterm institutional ownership (STIO). However, two highly correlated test variables never enter the same regression.

EMPIRICAL EVIDENCE Univariate Tests To provide preliminary insights into the relationship between firm credit ratings and IIIH, we conduct univariate tests. We first split the sample

62

NAJAH ATTIG ET AL.

Table 2. Descriptive Statistics for Regression Variables. Variable RATING IO WACR D(LTIOWSTIO) LTIO STIO SIZE COVERAGE MARGIN LEVERAGE CAPINT BETA LOSS UTILITY FINANCIAL INDUSTRIAL

Mean

Q1

Median

Q3

Standard Deviation

4.70 0.55 0.19 0.43 0.14 0.17 7.60 6.73 0.17 0.31 0.64 0.97 0.11 0.10 0.07 0.77

4.00 0.37 0.16 0.00 0.07 0.09 6.56 2.43 0.09 0.18 0.31 0.55 0.00 0.00 0.00 1.00

5.00 0.56 0.18 0.00 0.13 0.16 7.51 4.02 0.15 0.29 0.59 0.90 0.00 0.00 0.00 1.00

6.00 0.72 0.21 1.00 0.19 0.24 8.54 7.02 0.24 0.41 0.94 1.30 0.00 0.00 0.00 1.00

1.29 0.25 0.04 0.50 0.10 0.12 1.52 9.46 0.16 0.20 0.41 0.65 0.32 0.30 0.25 0.42

Notes: This table provides the mean, first quartile, median, third quartile, and standard deviation of the regression variables. The sample consists of 20,670 firm-year observations between 1985 and 2006. The appendix provides definitions and data sources for all variables.

according to whether the weighted average churn rate (WACR) is above or below the sample median and perform mean and median comparison tests to assess differences in firm credit ratings across the two subsamples. The results are reported in Table 4. In panel A, mean and median comparison tests indicate that firm years with above-median churn rates display a significantly lower RATING than firms with below-median churn rates. This finding suggests that firms benefit more from the presence of institutional investors with a long-term investment horizon than from institutional investors with a short-term investment horizon. This result finds further support from our second test (panel B), which indicates that firms in which long-term institutional investors dominate display better credit ratings than firms in which short-term institutional investors dominate.

Regression Analysis To determine whether institutional activism is associated with a longer institutional investment horizon, we run a more thorough empirical analysis

Table 3.

RATING IO WACR D(LTIOWSTIO) LTIO STIO SIZE COVERAGE MARGIN LEVERAGE CAPINT BETA LOSS UTILITY FINANCIAL INDUSTRIAL

RATING

IO

1.00 0.09 0.27 0.21 0.26 0.07 0.55 0.34 0.25 0.50 0.12 0.18 0.39 0.21 0.09 0.20

1.00 0.14 0.14 0.55 0.68 0.30 0.19 0.05 0.20 0.14 0.21 0.12 0.24 0.04 0.12

Correlation Matrix.

WACR D(LTIOWSTIO) LTIO STIO SIZE COVERAGE MARGIN LEVERAGE CAPINT BETA LOSS UTILITY FINANCIAL INDUSTRIAL

1.00 0.61 0.40 0.53 0.10 0.05 0.01 0.14 0.07 0.17 0.06 0.09 0.02 0.07

1.00 0.42 0.53 0.11 0.06 0.01 0.10 0.06 0.17 0.02 0.10 0.01 0.07

1.00 0.05 0.32 0.18 0.05 0.25 0.04 0.03 0.12 0.10 0.05 0.02

1.00 0.08 0.08 0.03 0.06 0.11 0.26 0.07 0.18 0.01 0.10

1.00 0.17 0.22 0.32 0.01 0.03 0.18 0.07 0.19 0.18

1.00 0.13 0.42 0.14 0.06 0.21 0.11 0.08 0.04

1.00 0.02 0.22 0.11 0.27 0.19 0.15 0.20

1.00 0.14 0.04 0.31 0.03 0.18 0.09

1.00 0.19 0.00 0.34 0.36 0.08

1.00 0.11 1.00 0.24 0.10 0.02 0.05 0.15 0.10

1.00 0.09 0.62

1.00 0.50

1.00

Notes: This table presents pairwise correlation coefficients among the regression variables. Coefficients reported in bold are significant at the 1% level. The sample consists of 20,670 firm-year observations between 1985 and 2006. The appendix provides definitions and data sources for all variables.

64

NAJAH ATTIG ET AL.

Table 4.

Univariate Tests.

Panel A: WACR

WACR p median WACR W Median Difference t-Stat/z-stat

(1) (2) (2)-(1)

N

Mean RATING

Median RATING

10,335 10,335

5.02 4.37 0.65 37.53

5.00 4.00 1.00 36.47

Panel B: D(LTIOWSTIO) D(LTIOWSTIO) ¼ 0 D(LTIOWSTIO) ¼ 1 Difference t-Stat/z-stat

(3) (4) (4)-(3)

11,817 8,853

4.46 5.01 0.55 30.53

4.00 5.00 1.00 29.54

Notes: This table reports mean and median comparison tests of firms’ credit ratings (RATING) across subsamples. WACR is the value-weighted average of institutional investors’ churn rates. D(LTIOWSTIO) is a dummy variable set equal to 1 if institutional ownership by long-term investors is higher than institutional ownership by short-term investors. The sample consists of 20,670 firm-year observations between 1985 and 2006. Statistical significance at the 1% level.

of the impact of IIIH on firm credit ratings. In all model specifications, we use binary variables to identify firms that operate in utility (UTILITY), financial (FINANCIAL), and industrial (INDUSTRIAL) sectors because default risk varies across industries (e.g., firms operating in regulated industries are expected to have lower default risk and thus higher credit ratings). The estimated coefficients of our baseline model, reported in model 1 of Table 5, are consistent with prior research (e.g., Ashbaugh-Skaife et al., 2006; Bhojraj & Sengupta, 2003; Blume et al., 1998). For instance, the estimated coefficient on SIZE loads positively and significantly on firm credit ratings, indicating that large firms are less likely to default, possibly because they tend to be older, with more established product lines and more varied sources of revenue. In line with the findings of Blume et al. (1998) and Bhojraj and Sengupta (2003), BETA displays a negative and significant coefficient, reflecting an adverse effect of increased equity risk on firm creditworthiness. Similarly, the estimated coefficients on LEVERAGE and LOSS are negative and significant, suggesting that nonmarket risk reduces firm credit ratings, plausibly because it provides information about the competency of management (Blume et al., 1998) and thus about the firm’s

Ordered Probit Results of the Effect of Institutional Investors’ Investment Horizon on Firm Credit Ratings. (1)

IO

(2)

(3)

0.062 (0.770)

0.073 (0.879) 5.408 (12.766)

WACR D(LTIOWSTIO)

(4)

(5)

0.345 (12.211)

1.450 (5.865)

STIO

COVERAGE MARGIN LEVERAGE CAPINT BETA

0.463 (27.143) 0.026 (10.683) 0.672 (5.263) 2.305 (19.985) 0.281 (4.989) 0.359 (16.317)

0.465 (26.836) 0.026 (10.662) 0.675 (5.289) 2.311 (19.812) 0.280 (4.971) 0.355 (16.270)

0.455 (26.388) 0.026 (10.935) 0.740 (5.783) 2.213 (19.282) 0.267 (4.790) 0.294 (13.616)

(7)

0.973 (7.280) 0.469 (27.731) 0.026 (10.738) 0.719 (5.596) 2.312 (19.925) 0.268 (4.791) 0.315 (14.738)

1.487 (5.848) 1.000 (7.492) 0.449 (25.867) 0.025 (10.810) 0.740 (5.731) 2.246 (19.577) 0.259 (4.644) 0.305 (14.358)

0.074 (0.918)

LTIO

SIZE

(6)

0.455 (26.513) 0.025 (10.685) 0.719 (5.657) 2.263 (19.673) 0.267 (4.818) 0.315 (14.775)

0.442 (25.277) 0.025 (10.740) 0.691 (5.378) 2.241 (19.624) 0.272 (4.852) 0.350 (15.981)

Institutional Investment Horizon and Firm Credit Ratings

Table 5.

65

66

Table 5. (Continued )

LOSS UTILITY FINANCIAL INDUSTRIAL Year effects N Pseudo-R2

(1)

(2)

(3)

(5)

(6)

(7)

0.933

0.935

0.958

(4) 0.952

0.926

0.958

(22.465) 0.617 (5.560) 0.090 (0.838) 0.082 (1.091) Yes 20,670 0.259

(22.392) 0.606 (5.432) 0.095 (0.883) 0.081 (1.077) Yes 20,670 0.259

(22.800) 0.612 (5.541) 0.108 (0.998) 0.071 (0.950) Yes 20,670 0.270

(22.719) 0.601 (5.465) 0.098 (0.913) 0.073 (0.966) Yes 20,670 0.265

(22.433) 0.686 (6.156) 0.072 (0.669) 0.087 (1.169) Yes 20,670 0.263

(22.863) 0.559 (5.077) 0.125 (1.157) 0.070 (0.936) Yes 20,670 0.262

0.951 (22.864) 0.628 (5.689) 0.107 (0.987) 0.075 (1.010) Yes 20,670 0.266

NAJAH ATTIG ET AL.

Notes: This table presents results of ordered probit regressions of firms’ credit ratings (RATING) on proxies for institutional investors’ investment horizon and a number of controls. S&P’s long-term issuer credit ratings are converted to an ordinal scale according to the following schedule: 8 (AAA), 7 (AA), 6 (A), 5 (BBB), 4 (BB), 3 (B), 2 (CCC), and 1 (CC). t-Statistics based on robust standard errors adjusted for clustering by firm are reported in parentheses. The sample consists of 20,670 firm-year observations between 1985 and 2006. The appendix provides definitions and data sources for all variables. Statistical significance at the 1% level. Statistical significance at the 5% level. Statistical significance at the 10% level.

Institutional Investment Horizon and Firm Credit Ratings

67

ability to make scheduled payments. In contrast, COVERAGE, MARGIN, and CAPINT load positively and significantly on firm credit ratings, suggesting that greater interest coverage, operating margin, and capital intensity lower default risk and thus enhance firm credit ratings. In model 2 of Table 5, we augment the baseline model with the fraction of the firm’s shares held by institutional investors (IO). Interestingly, the estimated coefficient on IO is negative but not significant. We conjecture that the insignificance of IO reflects the trade-off between institutional activism and passivism, as evidenced by the heterogeneity of investment horizon. Turning to our main research question, in models 3–7, we introduce our proxies for IIIH. The estimated coefficient on WACR in model 3 is negative and significant, suggesting that the propensity of institutional investors to alter the position of their shareholdings adversely affects firm credit ratings. This inference is further supported by the positive and statistically significant (at the 1% level) coefficient on D(LTIOWSTIO), which indicates that firms in which institutions with a long-term horizon are dominant have higher credit ratings. The sign and significance of both proxies (WACR and D(LTIOWSTIO)) lend support to the hypothesis that the presence of institutional investors with a long-term investment horizon is associated with reduced agency costs and information asymmetry, which translate into a better assessment of a firm’s creditworthiness. Overall, improved credit ratings appear to be due to more efficient monitoring and reduced informational asymmetries by long-term institutional investors. To shed further light on the role of institutions’ heterogeneity in affecting institutional activism and corporate governance, we replace the aggregate institutional ownership variable with the long-term institutional ownership (LTIO) in model 5 and short-term institutional ownership (STIO) in model 6. The estimated coefficients on LTIO and STIO are consistent with the view that long-term investment horizon is associated with more valuable monitoring than short-term investment horizon. Indeed, the estimated coefficient on LTIO is positive and significant, whereas that on STIO is negative and significant coefficient. This evidence is suggestive of more ‘‘relationship investing’’ by institutions with long-term investment horizons. Moreover, rating agencies appear to perceive a weak monitoring role and possibly an adverse informational role associated with short-term institutional ownership, which may exacerbate managerial myopia and entrenchment and thus translate into lower firm credit ratings. We obtain similar results when we simultaneously include long-term and short-term institutional ownership stakes in model 7.

68

NAJAH ATTIG ET AL.

Our evidence that institutional investors with a long-term investment horizon improve governance, as reflected in higher firm credit ratings, is in line with Gaspar et al. (2005), who conclude that institutions with short horizons are associated with weaker monitoring incentives. Our evidence also lends support to Chen et al. (2007), who find that investors with longer investment horizons have greater ability to collect and process information. Similarly, our findings are consistent with Burns et al. (2010), who show that institutions with short investment horizons (transient institutions) are associated with the use of discretionary accruals and the likelihood and severity of financial restatements. In sum, the evidence reported in Table 5 suggests that it is important to control for institutions’ heterogeneity when testing for institutional activism. Our findings indicate that institutional ownership and institutional investment horizon are mutually reinforcing, working together to affect corporate governance. Furthermore, our findings suggest that institutional investors’ monitoring incentive is susceptible to their heterogeneity as evidenced by their different investment horizons and that focusing on institutional shareholdings masks important variation in the governance role of institutions. This implication may explain the insignificance of the estimated coefficient on IO in model 2 and, more broadly, the inconclusive evidence in extant literature on the monitoring role of institutional ownership.

Robustness Tests In this section, we conduct additional analyses to assess the robustness of our results to using additional controls (Tables 6 and 7) and alternative measures of IIIH (Table 8). Our first set of robustness tests augments our baseline model with additional control variables to control for potential omitted correlated variables. First, we control for the ownership stakes of public pension funds (PUBO) that are members of the Council of Institutional Investors, as these funds are expected to be independent (Chen et al., 2007) and hence to exert more efficient monitoring. Second, we control for the Herfindahl index of institutional investors’ ownership to account for institutional ownership concentration (Bhojraj & Sengupta, 2003). Third, as an alternative proxy for institutional ownership concentration, we control for the value-weighted average of institutional investors’ Herfindahl index of ownership stakes. Fourth, to control for institutional investors’ industry specialization, we

Table 6.

Additional Controls for Institutional Investors’ Characteristics.

ADD_VAR ¼ PUBO

IO WACR

(1)

(2)

0.050 (0.590) 5.151 (11.993)

0.070 (0.849)

D(LTIOWSTIO)

0.330 (11.515)

LTIO STIO SIZE COVERAGE MARGIN LEVERAGE CAPINT BETA

0.448 (25.925) 0.026 (11.003) 0.790 (6.171) 2.171 (18.850) 0.254 (4.556) 0.299 (13.822)

0.446 (25.930) 0.026 (10.781) 0.779 (6.132) 2.210 (19.133) 0.251 (4.535) 0.318 (14.950)

ADD_VAR ¼ HERF1 (3)

1.289 (5.131) 1.101 (8.429) 0.439 (25.264) 0.026 (10.919) 0.794 (6.148) 2.192 (19.042) 0.246 (4.421) 0.310 (14.598)

(4)

(5)

0.300 (3.198) 5.896 (13.812)

0.249 (2.714)

0.447 (25.874) 0.026 (10.896) 0.735 (5.747) 2.178 (18.961) 0.260 (4.651) 0.303 (13.917)

0.367 (13.090)

0.448 (26.124) 0.025 (10.625) 0.713 (5.606) 2.239 (19.437) 0.261 (4.713) 0.323 (15.126)

ADD_VAR ¼ HERF2 (6)

2.289 (10.220) 0.828 (6.006) 0.436 (25.107) 0.025 (10.711) 0.739 (5.726) 2.201 (19.191) 0.242 (4.341) 0.311 (14.560)

(7)

(8)

0.073 (0.878) 5.407 (12.765)

0.074 (0.918)

0.455 (26.388) 0.026 (10.936) 0.740 (5.783) 2.213 (19.279) 0.267 (4.790) 0.294 (13.620)

0.345 (12.209)

0.455 (26.514) 0.025 (10.685) 0.719 (5.657) 2.262 (19.671) 0.267 (4.819) 0.315 (14.779)

ADD_VAR ¼ IOIW (9)

1.488 (5.848) 1.000 (7.489) 0.449 (25.868) 0.025 (10.811) 0.740 (5.731) 2.246 (19.575) 0.259 (4.645) 0.305 (14.364)

(10)

(11)

0.081 (0.979) 5.448 (13.121)

0.076 (0.951)

0.456 (26.382) 0.026 (10.970) 0.753 (5.847) 2.208 (19.179) 0.266 (4.780) 0.293 (13.527)

0.345 (12.275)

0.455 (26.437) 0.025 (10.700) 0.723 (5.679) 2.261 (19.616) 0.267 (4.813) 0.314 (14.740)

(12)

1.562 (6.799) 0.999 (7.485) 0.450 (25.859) 0.025 (10.853) 0.761 (5.844) 2.237 (19.432) 0.257 (4.620) 0.302 (14.230)

Table 6. (Continued ) ADD_VAR ¼ PUBO

LOSS UTILITY FINANCIAL INDUSTRIAL ADD_VAR Year effects N Pseudo-R2

ADD_VAR ¼ HERF1

ADD_VAR ¼ HERF2

ADD_VAR ¼ IOIW

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

0.953

0.946

0.945

0.942

0.939

0.933

0.958

0.952

0.951

0.951

0.950

(22.687) 0.608 (5.468) 0.098 (0.904) 0.065 (0.865) 4.042 (5.087)

(22.561) 0.597 (5.384) 0.086 (0.805) 0.065 (0.865) 4.777 (6.174) Yes 20,670 0.267

(22.704) 0.632 (5.675) 0.091 (0.845) 0.068 (0.913) 4.454 (5.583) Yes 20,670 0.268

(22.463) 0.624 (5.634) 0.070 (0.653) 0.067 (0.896) 2.028 (4.227) Yes 20,670 0.272

(22.433) 0.610 (5.538) 0.068 (0.633) 0.070 (0.928) 1.553 (3.551) Yes 20,670 0.267

(22.442) 0.639 (5.802) 0.066 (0.619) 0.071 (0.960) 2.716 (6.039) Yes 20,670 0.269

(22.797) 0.612 (5.540) 0.108 (0.997) 0.071 (0.949) 0.000 (1.493) Yes 20,670 0.270

(22.716) 0.601 (5.464) 0.098 (0.913) 0.073 (0.965) 0.000 (1.183) Yes 20,670 0.265

(22.861) 0.628 (5.689) 0.107 (0.986) 0.075 (1.009) 0.000 (1.535) Yes 20,670 0.266

(22.535) 0.627 (5.597) 0.091 (0.840) 0.075 (0.996) 0.369 (0.969) Yes 20,670 0.270

(22.651) 0.606 (5.434) 0.093 (0.850) 0.074 (0.982) 0.126 (0.351) Yes 20,670 0.265

0.941 (22.438) 0.654 (5.850) 0.080 (0.742) 0.081 (1.086) 0.585 (1.490) Yes 20,670 0.266

20,670 0.271

Notes: This table presents results of ordered probit regressions of firms’ credit ratings (RATING) on proxies for institutional investors’ investment horizon, additional institutional investors’ characteristics, and a number of controls. S&P’s long-term issuer credit ratings are converted to an ordinal scale according to the following schedule: 8 (AAA), 7 (AA), 6 (A), 5 (BBB), 4 (BB), 3 (B), 2 (CCC), and 1 (CC). t-Statistics based on robust standard errors adjusted for clustering by firm are reported in parentheses. The sample consists of 20,670 firm-year observations between 1985 and 2006. The appendix provides definitions and data sources for all variables. Statistical significance at the 1% level. Statistical significance at the 5% level. Statistical significance at the 10% level.

Additional Controls for Corporate Governance, Transparency, and Information Environment.

Table 7.

ADD_VAR ¼ GINDEX

IO WACR

(1)

(2)

0.054 (0.388) 9.038 (11.747)

0.180 (1.371)

D(LTIOWSTIO)

0.412 (11.711)

LTIO STIO SIZE COVERAGE MARGIN LEVERAGE CAPINT BETA LOSS UTILITY

0.413 (18.402) 0.023 (8.697) 1.175 (5.839) 2.275 (12.780) 0.171 (2.377) 0.420 (13.667) 1.062 (16.670) 0.435 (3.316)

0.413 (18.590) 0.022 (8.244) 1.139 (5.697) 2.350 (13.263) 0.160 (2.218) 0.458 (15.131) 1.046 (16.440) 0.371 (2.887)

ADD_VAR ¼ STKMIX (3)

1.654 (4.616) 1.601 (8.027) 0.406 (18.041) 0.022 (8.453) 1.154 (5.686) 2.317 (13.134) 0.161 (2.230) 0.448 (14.610) 1.047 (16.604) 0.432 (3.319)

(4)

(5)

0.242 (1.432) 10.589 (11.507)

0.236 (1.471)

0.469 (18.065) 0.025 (7.715) 1.226 (5.215) 2.644 (11.830) 0.096 (1.160) 0.482 (12.660) 0.935 (12.140) 0.507 (3.169)

0.524 (12.084)

0.483 (18.860) 0.025 (7.553) 1.165 (5.043) 2.691 (12.063) 0.091 (1.104) 0.520 (13.575) 0.906 (11.930) 0.473 (3.008)

ADD_VAR ¼ AQ (6)

2.742 (4.784) 1.622 (7.546) 0.477 (18.242) 0.025 (7.689) 1.246 (5.201) 2.641 (11.780) 0.067 (0.809) 0.495 (12.751) 0.929 (12.123) 0.502 (3.118)

(7)

(8)

0.371 (3.661) 5.949 (11.592)

0.424 (4.234)

0.490 (23.526) 0.030 (10.835) 0.843 (6.048) 2.028 (15.809) 0.204 (3.360) 0.331 (13.290) 0.919 (19.262) 0.739 (5.470)

0.447 (13.801)

0.493 (23.738) 0.030 (10.752) 0.826 (5.974) 2.062 (16.088) 0.199 (3.291) 0.344 (13.970) 0.913 (19.087) 0.732 (5.418)

ADD_VAR ¼ ACOV (9)

2.211 (6.510) 0.868 (5.939) 0.488 (23.267) 0.030 (10.769) 0.862 (6.075) 2.057 (15.942) 0.183 (2.991) 0.327 (13.280) 0.917 (19.267) 0.736 (5.443)

(10)

(11)

0.080 (0.930) 6.568 (14.570)

0.085 (1.010)

0.343 (17.493) 0.021 (8.479) 0.742 (5.328) 2.262 (18.221) 0.118 (1.939) 0.379 (16.787) 0.966 (21.101) 0.534 (4.671)

0.402 (13.862)

0.339 (17.259) 0.020 (8.053) 0.721 (5.206) 2.313 (18.563) 0.113 (1.851) 0.406 (18.217) 0.958 (20.756) 0.521 (4.566)

(12)

1.667 (6.833) 1.408 (10.231) 0.327 (16.421) 0.020 (8.178) 0.750 (5.335) 2.276 (18.388) 0.099 (1.615) 0.394 (17.596) 0.962 (21.064) 0.567 (4.942)

Table 7. (Continued ) ADD_VAR ¼ GINDEX

FINANCIAL INDUSTRIAL ADD_VAR Year effects N Pseudo-R2

ADD_VAR ¼ STKMIX

ADD_VAR ¼ AQ

ADD_VAR ¼ ACOV

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

0.066 (0.459) 0.150 (1.568) 0.031 (3.322)

0.083 (0.599) 0.134 (1.420) 0.031 (3.387) Yes 12,203 0.246

0.081 (0.571) 0.148 (1.558) 0.030 (3.223) Yes 12,203 0.247

0.081 (0.475) 0.234 (2.001) 0.016 (0.917) Yes 8,377 0.283

0.073 (0.443) 0.231 (1.993) 0.017 (0.901) Yes 8,377 0.275

0.108 (0.614) 0.248 (2.106) 0.015 (0.899) Yes 8,377 0.280

0.081 (0.613) 0.120 (1.227) 0.156 (2.401) Yes 15,660 0.293

0.064 (0.490) 0.124 (1.258) 0.161 (2.481) Yes 15,660 0.292

0.094 (0.706) 0.124 (1.267) 0.159 (2.447) Yes 15,660 0.292

0.042 (0.373) 0.005 (0.059) 0.033 (9.764) Yes 19,422 0.279

0.030 (0.274) 0.005 (0.061) 0.035 (10.302) Yes 19,422 0.274

0.040 (0.352) 0.003 (0.042) 0.035 (10.438) Yes 19,422 0.275

12,203 0.255

Notes: This table presents results of ordered probit regressions of firms’ credit ratings (RATING) on proxies for institutional investors’ investment horizon, proxies for firms’ corporate governance, transparency, and information environment, and a number of controls. S&P’s long-term issuer credit ratings are converted to an ordinal scale according to the following schedule: 8 (AAA), 7 (AA), 6 (A), 5 (BBB), 4 (BB), 3 (B), 2 (CCC), and 1 (CC). t-Statistics based on robust standard errors adjusted for clustering by firm are reported in parentheses. The sample consists of 20,670 firm-year observations between 1985 and 2006. The appendix provides definitions and data sources for all variables. Statistical significance at the 1% level. Statistical significance at the 5% level. Statistical significance at the 10% level.

Alternative Proxies for Institutional Investors’ Investment Horizon.

Alternative Threshold (1) IO

(2)

0.055 (0.686)

Number of Shareholders (3)

(4)

0.002

0.002

(10.349)

(7.756)

Alternative Churn Rate (5)

WACR D(LTIOWSTIO)

0.345 (12.619)

LTIO STIO SIZE COVERAGE MARGIN LEVERAGE CAPINT

0.449 (26.039) 0.025 (10.507) 0.709 (5.571) 2.258 (19.656) 0.274 (4.960)

1.728 (4.659) 1.283 (7.738) 0.444 (25.435) 0.025 (10.602) 0.727 (5.621) 2.254 (19.636) 0.271 (4.874)

0.406 (11.392)

0.282 (12.399) 0.017 (6.558) 0.611 (4.892) 2.220 (19.589) 0.276 (4.939)

0.314 (13.864) 0.018 (7.002) 0.648 (5.233) 2.244 (19.649) 0.278 (5.018)

0.009 (10.806) 0.004 (4.560) 0.338 (14.533) 0.019 (7.207) 0.680 (5.476) 2.294 (19.907) 0.272 (4.941)

(6)

(7)

0.116 (1.371) 13.109 (9.736)

0.091 (1.117)

0.459 (26.642) 0.026 (10.898) 0.746 (5.814) 2.232 (19.411) 0.265 (4.753)

0.375 (9.486)

0.468 (27.320) 0.025 (10.817) 0.687 (5.383) 2.284 (19.765) 0.280 (5.028)

(8)

1.712 (5.165) 0.878 (7.569) 0.454 (26.017) 0.026 (10.762) 0.741 (5.714) 2.253 (19.506) 0.258 (4.622)

Institutional Investment Horizon and Firm Credit Ratings

Table 8.

73

74

Table 8. (Continued ) Alternative Threshold

BETA LOSS UTILITY FINANCIAL INDUSTRIAL Year effects N Pseudo-R2

Number of Shareholders

Alternative Churn Rate

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

0.318 (14.876) 0.954 (22.766) 0.583 (5.308) 0.093 (0.864) 0.069 (0.914) Yes 20,670 0.265

0.308 (14.474) 0.948 (22.675) 0.614 (5.584) 0.107 (0.981) 0.072 (0.963) Yes 20,661 0.266

0.352 (16.163) 0.915 (22.080) 0.701 (6.407) 0.096 (0.899) 0.001 (0.014) Yes 20,670 0.271

0.301 (14.148) 0.941 (22.485) 0.656 (6.000) 0.076 (0.697) 0.008 (0.104) Yes 20,670 0.277

0.287 (13.694) 0.944 (22.907) 0.679 (6.181) 0.055 (0.510) 0.020 (0.267) Yes 20,670 0.276

0.289 (13.226) 0.962 (22.776) 0.595 (5.422) 0.121 (1.110) 0.070 (0.941) Yes 20,670 0.269

0.328 (15.271) 0.956 (22.714) 0.590 (5.380) 0.105 (0.970) 0.078 (1.037) Yes 20,670 0.264

0.301 (14.167) 0.955 (22.806) 0.609 (5.515) 0.130 (1.184) 0.073 (0.987) Yes 20,666 0.267

NAJAH ATTIG ET AL.

Notes: This table presents results of ordered probit regressions of firms’ credit ratings (RATING) on alternative proxies for institutional investors’ investment horizon, additional institutional investors’ characteristics, and a number of controls. S&P’s long-term issuer credit ratings are converted to an ordinal scale according to the following schedule: 8 (AAA), 7 (AA), 6 (A), 5 (BBB), 4 (BB), 3 (B), 2 (CCC), and 1 (CC). In models 1 and 2, an institutional investor is classified as a long-term (short-term) investor if its churn rate is in the bottom (top) quartile. In models 3 through 5, we define IO as the number of institutional investors, D(LTIOWSTIO) as a dummy variable set equal to 1 if the number of long-term institutional investors is higher than the number of short-term institutional investors, and LTIO (STIO) as the number of long-term (short-term) institutional investors. In models 6 through 8, we use the churn rate definition proposed by Yan and Zhang (2009). t-Statistics based on robust standard errors adjusted for clustering by firm are reported in parentheses. The sample consists of 20,670 firm-year observations between 1985 and 2006. The appendix provides definitions and data sources for all variables. Statistical significance at the 1% level. Statistical significance at the 5% level. Statistical significance at the 10% level.

Institutional Investment Horizon and Firm Credit Ratings

75

include the value-weighted average of the fraction of institutional investors’ portfolios that is invested in the firm’s industry. Table 6 reports the results. We find that our results on the impact of institutional investment horizon remain qualitatively unchanged after controlling for these additional explanatory variables, which further supports the conclusion that LTHI are associated with more efficient monitoring and in turn better firm credit ratings. Our second set of robustness checks addresses the impact of additional proxies for corporate governance, transparency, and information environment on our main findings. First, we control for GINDEX, Gompers, Ishii, and Metrick’s (2003) index of 24 antitakeover provisions, which measures the strength of shareholder rights. Second, we control for the ratio of a CEO’s stock-based compensation to total compensation, STKMIX, to account for the extent to which managers’ interests are aligned with those of shareholders. We next control for firm information quality, as measured by the standard deviation of discretionary accruals over the past five years (AQ)9 and the number of analysts providing one-year-ahead forecasts of earnings per share (ACOV). The results are reported in Table 7. The stability of our results to these additional controls supports our previous inference that, in an effort to safeguard their investments, institutions with a long investment horizon display greater institutional activism and more efficient monitoring, which results in a better assessment of firm creditworthiness and thus better firm credit ratings. Next, we test the robustness of our inferences to the use of different cutoffs for the classification of long- versus short-term institutional investors, the use of the number of institutional investors instead of ownership stakes, and the use of alternative definition of the churn rate. First, in models 1 and 2 of Table 8, we identify long-term (short-term) investors using the bottom (top) quartile of the churn rate. Our results remain unchanged, as can be seen by the positive (negative) coefficients on D(LTIOWSTIO) and LTIO (STIO). Second, we build on a growing body of literature that captures institutional investors’ monitoring role by their number, as doing so may reveal relevant information about a firm more efficiently than institutional shareholdings (Back et al., 2000; Foster & Viswanathan, 1996).10 Specifically, in models 3 and 5 of Table 8, we define IO as the number of institutional investors, LTIO as the number of longterm institutional investors, and STIO as the number of short-term institutional investors, respectively; and the variable D(LTIOWSTIO) in model 4 takes the value 1 if the number of long-term investors is greater than the number of short-term investors. The estimated coefficients on the

76

NAJAH ATTIG ET AL.

variables of interest (IO, D(LTIOWSTIO), LTIO, and STIO) suggest that when the number of long-term institutional investors is larger than the number of short-term institutional investors, firms benefit from a reduction in agency costs and information risk and thus enjoy higher credit ratings. This result lends support to our previous conclusion that firms in which long-term institutional investors dominate (in terms of ownership stakes) are associated with lower agency costs and less pronounced information problems. In addition, this evidence extends the argument of Foster and Viswanathan (1996) and Back et al. (2000) that information about the firm is revealed more rapidly when the number of institutional investors is large. In our last set of robustness tests, we test the stability of our results to an alternative proxy for firms’ churn rate. Yan and Zhang (2009) propose a proxy for the churn rate that minimizes the impact of investor cash flows on portfolio turnover. Specifically, they calculate firms’ churn rate as follows: CRk;t ¼

MinðAGG_PURCHk;t ; AGG_SALEk;t Þ PN k;t Sk;i;t Pi;t Sk;i;t1 Pi;t1 i¼1

(5)

2

where AGG_PURCHk;t :

Nk P

jSk;i;t Pi;t  Sk;i;t1 Pi;t1  S k;i;t1 DPi;t j

N k;i;t 4N k;i;t1

AGG_SALEk;t :

Nk P

jS k;i;t Pi;t  S k;i;t1 Pi;t1  S k;i;t1 DPi;t j

N k;i;t 4N k;i;t1

In the last three models of Table 8, we recompute all of our investor horizon variables using Yan and Zhang’s (2009) measure of churn rate. The nearequivalency between the results reported in Table 5 and the evidence shown in models 6, 7, and 8 of Table 8 suggests that our inference of more efficient monitoring by long-term institutional investors, as evidenced in higher firm credit ratings, is not sensitive to the measure of churn rate employed.

CONCLUSION The effect of institutional activism, as evidenced by institutions’ shareholdings, has received widespread attention. This literature finds mixed evidence on the monitoring and informational advantage of institutional investors. However, a review of the literature suggests that existing studies

Institutional Investment Horizon and Firm Credit Ratings

77

generally assume that institutional investors represent a homogeneous group, with comparable investment objectives and governance styles. In this chapter, we examine the impact of institutions’ heterogeneity, captured by their different investment horizons, on firm credit ratings. We consistently find support for a more efficient monitoring role by institutions with a longterm investment horizon, as indicated by better firm credit ratings. Our evidence indicates that the monitoring incentive and informational advantage of institutional investors are susceptible to heterogeneity in their investment horizon and hence that focusing on institutional shareholdings masks important variation in the governance role of institutional investors. This result may help explain the mixed evidence in existing research on institutional monitoring. Our evidence also suggests that institutional ownership and institutional investment horizon are mutually reinforcing, working together to affect corporate governance and highlights the importance of the number of long-term institutional investors in both reducing agency costs and enhancing information quality. Taken together, these findings underscore the need for further research to enrich our understanding of the impact of institutions’ heterogeneity on their governance role.

NOTES 1. For instance, McConnell and Servaes (1990), Smith (1996), and Del Guercio and Hawkins (1999) find that institutional ownership positively and significantly affects firm performance. Similarly, Hartzell and Starks (2003) and Chung, Firth, and Kim (2002) show that institutional ownership reduces agency problems, as reflected in opportunistic managerial compensation and earnings management, respectively. Yet, several other studies express skepticism about the efficiency of institutional investors’ governance role, pointing to evidence that institutional ownership has a neutral effect on firm performance (e.g., Agrawal & Knoeber, 1996; Demsetz & Lehn, 1985; Faccio & Lasfer, 2000). 2. In related work, Bushee (2001) shows that transient investors prefer short returns. Earlier studies such as Nofsinger and Sias (1999) and Wermers (1999) attribute the positive relationship between changes in institutional ownership and returns to the feedback from institutional trading. 3. Arguably, the effects of improved corporate governance become evident over the middle to long term. Chen et al. (2007) argue that institutions weigh the costs and the benefits of trading versus monitoring, and only long-term investors specialize in monitoring rather than trading. 4. This is a plausible assumption because LTHI trade less frequently than STHI (e.g., to avoid transaction costs and price impact associated with trading of large blocks).

78

NAJAH ATTIG ET AL.

5. However, Maug (1998) concludes that the impact on monitoring would be unambiguously positive, as liquidity provides incentives for other shareholders to accumulate large stakes and exert corporate control. 6. Such an informational role can also apply to other institutional investors who tend to ‘‘act as a group’’ (Cornett, Marcus, Saunders, & Tehranian, 2007). 7. Edmans (2009) shows that blockholders can improve firm value through their trading. Other recent studies point out that activism through ‘‘exit’’ can be a viable governance option, especially in the presence of asymmetric information (e.g., Harris & Raviv, 2008). Gopalan (2008) argues that exit by an informed large shareholder creates value by encouraging another bidder to acquire information and implement improvements. 8. Bhojraj and Sengupta (2003) show that firms with greater institutional ownership enjoy higher bond ratings. Mansi et al. (2004) find that auditor quality and tenure are negatively related to the cost of debt financing. Ashbaugh-Skaife et al. (2006) document that better-governed firms are associated with higher credit ratings. 9. We estimate discretionary accruals using McNichols’s (2002) accruals generating process: TACi,t ¼ F0 þ F1CFOi,t1 þ F2CFOi,t þ F3CFOi,tþ1 þ F4DREVi,t þ F5 PPEi,t þ ei,t, where i indexes firms; t indexes years; CFO ¼ cash flow from operations; TAC ¼ total accruals defined as income before extraordinary items minus cash from operations; DREV ¼ change in sales; PPE ¼ gross property, plant, and equipment; and e ¼ error term. All variables are scaled by average total assets. The model is estimated using the COMPUSTAT universe every year for each Fama and French (1997) industry group with at least 20 firms. Discretionary accruals are the firm-level residuals (winsorized at the 1% of each tail to reduce the influence of extreme values). 10. Cornett et al. (2007) find a significant relation between a firm’s operating cash flow and the number of institutional stockholders. O’Neill and Swisher (2003) show that an increase in the number of institutional investors is associated with a decrease in the extent of informed trading. Chen, Hong, and Stein (2002) show that the number of mutual funds holding a stock predicts stock returns.

ACKNOWLEDGMENTS We appreciate the generous financial support from Canada’s Social Sciences and Humanities Research Council.

REFERENCES Admati, A. R., & Pfleiderer, P. C. (2009). The ‘Wall Street Walk’ and shareholder activism: Exit as a form of voice. Review of Financial Studies, 22, 2445–2485. Agrawal, A., & Knoeber, C. R. (1996). Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, 31, 377–397. Ashbaugh-Skaife, H., Collins, D. W., & LaFond, T. (2006). The effects of corporate governance on firms’ credit ratings. Journal of Accounting and Economics, 42, 203–243.

Institutional Investment Horizon and Firm Credit Ratings

79

Back, K. H., Cao, H. H., & Willard, G. A. (2000). Imperfect competition among informed traders. Journal of Finance, 55, 2117–2155. Bagwell, L. (1991). Shareholder heterogeneity: Evidence and implications. American Economic Review, 81, 218–221. Bhide, A. (1993). The hidden costs of stock market liquidity. Journal of Financial Economics, 34, 31–51. Bhojraj, S., & Sengupta, P. (2003). Effect of corporate governance on bond ratings and yields: The role of institutional investors and outside directors. Journal of Business, 76, 455–475. Blume, M. E., Lim, F., & Mackinlay, A. C. (1998). The declining credit quality of U.S. corporate debt: Myth or reality. Journal of Finance, 53, 1389–1413. Brickley, J., Lease, R., & Smith, C. (1988). Ownership structure and voting on antitakeover amendments. Journal of Financial Economics, 20, 267–291. Burns, N., Kedia, S., & Lipson, M. (2010). Institutional ownership and monitoring: Evidence from financial misreporting. Journal of Corporate Finance, 16, 443–455. Bushee, B. (2001). Do institutional investors prefer near-term earnings over long-run value. Contemporary Accounting Research, 18, 207–246. Chen, X., Harford, J., & Li, K. (2007). Monitoring: Which institutions matter? Journal of Financial Economics, 86, 279–305. Chen, J., Hong, H., & Stein, J. C. (2002). Breadth of ownership and stock returns. Journal of Financial Economics, 66, 171–205. Chidambaran, N. K., & John, K. (2000). Managerial compensation, voluntary disclosure, and large shareholder monitoring. Unpublished paper. New York University, New York, NY. Chung, R. M., Firth, M., & Kim, J. B. (2002). Institutional monitoring and opportunistic earnings management. Journal of Corporate Finance, 8, 29–48. Cornett, M. M., Marcus, A. J., Saunders, A., & Tehranian, H. (2007). The impact of institutional ownership on corporate operating performance. Journal of Banking & Finance, 31, 1771–1794. Del Guercio, D., & Hawkins, J. (1999). The motivation and impact of pension fund activism. Journal of Financial Economics, 52, 293–340. Demsetz, H., & Lehn, K. (1985). The structure of corporate ownership. Journal of Political Economy, 93, 1155–1177. Dimson, E. (1979). Risk measurement when shares are subject to infrequent trading. Journal of Financial Economics, 7, 197–226. Edmans, A. (2009). Blockholder trading, market efficiency, and managerial myopia. Journal of Finance, 64, 2481–2513. Faccio, M., & Lasfer, M. A. (2000). Do occupational pension funds monitor companies in which they hold large stakes? Journal of Corporate Finance, 6, 71–110. Fama, E. F., & French, K. R. (1997). Industry costs of equity. Journal of Financial Economics, 43, 153–193. Foster, D. F., & Viswanathan, S. (1996). Strategic trading when agents forecast the forecasts of others. Journal of Finance, 51, 1437–1478. Gaspar, J., Massa, M., & Matos, P. (2005). Shareholder investment horizons and the market for corporate control. Journal of Financial Economics, 76, 135–165. Gompers, P. A., Ishii, J., & Metrick, A. (2003). Corporate governance and equity prices. Quarterly Journal of Economics, 118, 107–155.

80

NAJAH ATTIG ET AL.

Gopalan, R. (2008). Institutional stock sales and takeovers: The disciplinary role of voting with your feet. Working Paper. Washington University, St. Louis, MO. Harris, M., & Raviv, A. (2008). A theory of board control and size. Review of Financial Studies, 21, 1797–1832. Hartzell, J. C., & Starks, L. T. (2003). Institutional investors and executive compensation. Journal of Finance, 58, 2351–2374. Hawley, P. J. (1995). Political voice, fiduciary activism, and the institutional ownership of U.S. corporations: The role of public and noncorporate pension funds. Sociological Perspectives, 38, 415–435. Healy, P., Hutton, A., & Palepu, K. (1999). Stock performance and intermediation changes surrounding sustained increases in disclosure. Contemporary Accounting Research, 16, 485–520. Hirschman, A. O. (1970). Exit, voice, and loyalty: Responses to decline in firms, organizations and states. Cambridge, MA: Harvard University Press. Mansi, A. A., Maxwell, W. F., & Miller, D. P. (2004). Does auditor quality and tenure matter to investors? Evidence from the bond market. Journal of Accounting Research, 42, 755–793. Maug, E. (1998). Large shareholder as monitors: Is there a tradeoff between liquidity and control? Journal of Finance, 53, 65–98. McConnell, J., & Servaes, H. (1990). Additional evidence on equity ownership and corporate value. Journal of Financial Economics, 27, 595–613. McNichols, M. (2002). Discussion of the quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review, 77, 61–69. Nofsinger, J. R., & Sias, R. W. (1999). Herding and feedback trading by institutional and individual investors. Journal of Finance, 54, 2263–2295. O’Neill, M., & Swisher, J. J. (2003). Institutional investors and information asymmetry: An event study of self-tender offers. Financial Review, 38, 197–211. Parrino, R., Sias, R. W., & Starks, L. T. (2003). Voting with their feet: Institutional ownership changes around forced CEO turnover. Journal of Financial Economics, 68, 3–46. Pound, J. (1988). Proxy contests and the efficiency of shareholder oversight. Journal of Financial Economics, 20, 237–265. Sengupta, P. (1998). Corporate disclosure quality and the cost of debt. The Accounting Review, 73, 459–474. Shleifer, A., & Vishny, R. (1986). Large shareholders and corporate control. Journal of Political Economy, 94, 461–488. Smith, M. (1996). Shareholder activism by institutional investors: Evidence from CalPERS. Journal of Finance, 51, 227–252. Weber, J. (2006). Discussion of the effects of corporate governance on firms’ credit ratings. Journal of Accounting and Economics, 42, 245–254. Wermers, R. (1999). Mutual fund herding and the impact on stock prices. Journal of Finance, 54, 581–622. Wermers, R. (2000). Mutual fund performance: An empirical decomposition into stock-picking talent, style, transactions costs, and expenses. Journal of Finance, 55, 1655–1695. Yan, X., & Zhang, Z. (2009). Institutional investors and equity returns: Are short-term institutions better informed? Review of Financial Studies, 22(1), 893–924.

Variable Panel A: Dependent variable RATING

Definition

Source

S&P’s long-term issuer credit ratings converted to an ordinal scale according to the following schedule: 8 (AAA), 7 (AA), 6 (A), 5 (BBB), 4 (BB), 3 (B), 2 (CCC), and 1 (CC)

Authors’ calculations based on COMPUSTAT data

Panel B: Institutional ownership variables IO Fraction of firm’s shares held by institutional investors WACR D(LTIOWSTIO)

LTIO STIO PUBO HERF1 HERF2 IOIW Panel C: Control variables SIZE COVERAGE

Value-weighted average of institutional investors’ churn rates Dummy variable set equal to 1 if institutional ownership by long-term investors is higher than institutional ownership by short-term investors, and 0 otherwise Long-term institutional ownership Short-term institutional ownership Ownership by public pension funds that are members of the Council of Institutional Investors Herfindahl Index of institutional investors’ stakes in the firm Value-weighted average of institutional investors’ Herfindahl Index of ownership stakes Value-weighted average of the fraction of institutional investors’ portfolios invested in the firm’s industry Natural logarithm of total assets in US$ million

As above As above As above As above As above As above

Authors’ calculations based on COMPUSTAT data As above

81

Earnings before interest and taxes plus interest expense divided by interest expense

Authors’ calculations based on Thomson 13-F data As above As above

Institutional Investment Horizon and Firm Credit Ratings

APPENDIX: VARIABLE DEFINITIONS AND DATA SOURCES

82

Appendix. (Continued ) Variable

Definition

Source

Ratio of operating income to sales Ratio of long-term debt to total assets Ratio of property, plant, and equipment to total assets Market beta estimated over the fiscal year using Dimson’s (1979) model with one lag and one lead of the CRSP value-weighted index

As above As above As above Authors’ calculations based on CRSP data

LOSS GINDEX

Indicator variable set to 1 if net income before extraordinary items is negative in the current and previous years, and 0 otherwise Gompers et al. (2003) index of 24 antitakeover provisions

STKMIX

Ratio of a CEO’s stock-based compensation to total compensation

AQ

Standard deviation of discretionary accruals over the past five years. We estimate discretionary accruals using McNichols’s (2002) accruals generating process: TACi,t ¼ F0 þ F1CFOi,t1 þ F2CFOi,t þ F3CFOi,tþ1 þ F4DREVi,t þ F5PPEi,t þ ei,t, where i indexes firms; t indexes years; CFO ¼ cash flow from operations; TAC ¼ total accruals defined as income before extraordinary items minus cash from operations; DREV ¼ change in sales; PPE ¼ gross property, plant, and equipment; and e ¼ error term. All variables are scaled by average total assets. The model is estimated using the COMPUSTAT universe every year for each Fama and French (1997) industry group with at least 20 firms. Discretionary accruals are the firm-level residuals (winsorized at the 1% of each tail to reduce the influence of extreme values) Number of analysts providing one-year-ahead forecasts of earnings per share

Authors’ calculations based on COMPUSTAT data Authors’ calculations based on RiskMetrics data Authors’ calculations based on ExecuComp data Authors’ calculations based on COMPUSTAT data

ACOV

I/B/E/S

NAJAH ATTIG ET AL.

MARGIN LEVERAGE CAPINT BETA

DIVESTMENT OF FOREIGN STRATEGIC INVESTMENT IN CHINA’S BANKING SECTOR: CAUSES AND CONSEQUENCES Yuhua Li and Konari Uchida STRUCTURED ABSTRACT Purpose – Investigate the causes and consequences of foreign financial institutions’ divestments in China’s banking sector which is an example of cross-border transactions by institutional investors. Methodology – Use a sample of 26 foreign financial institutions’ strategic investments in Chinese banks. Ten of those investments are divested after the global financial crisis. We investigate determinants of the divestment, business cooperation after the divestment, and Chinese banks’ stock price reactions to the divestment announcement. Findings – The poor performance of foreign financial institutions, which is attributable to the global financial crisis, and the institutions’ regulated low equity ownership are important causes of divestment (or whole divestment). In contrast, Chinese banks’ poor performance does not cause foreign divestments. Foreign financial institutions that fully divest their equity stakes usually terminate their cooperative business, which was required by the strategic investment agreement. The Bank of China and Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 83–110 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012006

83

84

YUHUA LI AND KONARI UCHIDA

the China Construction Bank, which experienced large H-share divestments, experienced large economic declines in A-share values. Social implications – Foreign financial institutions’ strategic investments created substantial shareholder value before the divestment. Banking sector developments that rely on foreign investments are vulnerable to economic downturns in developed countries. Originality/value of paper – To the best of our knowledge, this is the first trial to analyze the impact of divestments on divested bank performance. Keywords: Foreign strategic investment; divestment; Chinese banks; business cooperation; event study JEL classifications: G15; G21; G24; M16; M21

INTRODUCTION Foreign institutional investors are likely to play a key role in emerging markets by improving the efficiency and governance of local companies. Indeed, previous studies show that in emerging countries foreign financial institutions are more efficient than domestic banks and their entry and investments play a key role in the banking sector (Berger, Hasan, & Zhou, 2009; Bhattacharya, Lovell, & Sahay, 1997; Bonin, Hassan, & Wachtel, 2005a, 2005b; Claessens, Demirgu¨c- -Kunt, & Huizinga, 2001; Denizer, 2000; Goldberg, Dages, & Kinney, 2000; Goldberg & Saunders, 1981; Levine, 1996; Walter & Gray, 1983; Weill, 2003). However, previous studies also show that foreign banks that performed poorly in their home country tend to exit from the domestic market (Hryckiewicz & Kowalewski, 2011; Hryckiewicz, Kowalewski, & Niepodleglosci, 2010; Peek & Rosengren, 1997, 2000; Tschoegl, 2004; Williams, 1996). These facts naturally suggest banking sector developments that rely on foreign investments are vulnerable to exogenous shocks that occur in the home country. It is important to analyze the impact of financial institutions’ cross-border transactions on emerging markets in examining the role of institutional investors in the globalized environment. This chapter is principally intended to investigate the causes and consequences of foreign financial institutions’ divestments in China’s banking sector. As liberalization and the opening up of financial markets have been

Divestment of Foreign Strategic Investment in China’s Banking Sector

85

developed, foreign strategic investments have become an important foreign entry mode in the Chinese banking sector (Leigh & Podpiera, 2006). Indeed, 21 Chinese banks received foreign strategic investments from 29 foreign banks as of the end of 2006; the total investment amounted to about US$19.0 billion (China Banking Regulatory Commission (CBRC), ‘‘Report on the Opening-up of the Chinese Banking Sector,’’ January 25, 2007). In foreign strategic investment, foreign financial institutions that acquire certain equity stakes of Chinese banks are required to cooperate with (or assist) the Chinese bank on some businesses (retail banking, private banking, corporate governance, risk management, etc.). The cooperative business helps Chinese banks to improve their performance, product development, governance, risk management, reputation, and so on (Hope & Hu, 2006). Since the global financial crisis, however, some foreign financial institutions divested by cutting part or whole of their equity stakes in Chinese banks. Importantly, those Chinese banks are still listed even after divestment, and stock price data is available. In addition, different types of shares of a single Chinese bank are traded in segmented stock markets (A-shares in domestic stock markets and H-shares in the Hong Kong Stock Exchange), but most divestments are conducted in the Hong Kong Stock Exchange. Previous studies suggest that Chinese A- and H-share prices are not strongly interrelated (Hatemi-J & Roca, 2004; Wang & Jiang, 2004; Yeh & Lee, 2000). This environment allows us to investigate stock returns that are isolated, at least to a certain extent, from the effect of foreign financial institutions’ selling pressure. In other words, divestments in China’s banking sector are advantageous in measuring relatively pure effects of foreign financial institutions’ divestments on stock returns of divested firms. This characteristic motivates us to examine how emerging countries’ banking sectors are sensitive to economic downturn in developed countries by using Chinese data. To the best of our knowledge, this is the first trial to analyze the impact of divestments on divested bank performance.1 Our main results are summarized as follows. First, we find that poor financial performance and regulated low-equity ownership of divesting financial institutions are significantly related to the likelihood that Chinese banks experience foreign divestments (or whole divestments). In contrast, we do not find evidence that Chinese bank performance is related to the probability of being divested. Those results are consistent with previous works (Hryckiewicz & Kowalewski, 2011; Hryckiewicz et al., 2010; Peek & Rosengren, 1997, 2000; Tschoegl, 2004; Williams, 1996), and suggest that in the Chinese banking sector, foreign divestments are an exogenous shock,

86

YUHUA LI AND KONARI UCHIDA

which is associated with the global financial crisis. Although foreign financial institutions that partially divested their equity stakes maintain business cooperation with Chinese banks, those that divested their whole equity stake tend to end business cooperation. The Bank of China (BOC) and the China Construction Bank (CCB), which experienced large divestments in the Hong Kong Stock Exchange, show negative excess returns in A-shares for a few days surrounding divestments; in aggregate, A-shareholders of BOC (CCB) lost approximately 10 percent (8.4 percent) of their value during five days surrounding divestments. These results suggest that economic downturns in the home country have a serious impact on an emerging country’s banking sector; banking sector developments that rely on foreign financial institutions’ investments are exposed to economic downturns of the home country. The reduced shareholder value also implies that foreign strategic investments created substantial shareholder value before divestment. This provides additional evidence that foreign investments play an important role in an emerging market’s banking sector. In sum, financial institutions’ cross-border transactions have a substantial impact on emerging markets. Our analyses contribute to the literature in that we successfully avoid endogeneity problems and measure relatively pure effects of foreign divestments on stock returns of divested banks. The remainder of this chapter is as follows. The second section examines causes of divestments for China’s banking sector. The third section presents a case study of the impact of divestments on business cooperation between foreign financial institutions and Chinese banks. Fourth section examines stock market reactions to divestment announcements. A brief summary is presented in the conclusion.

CAUSES OF DIVESTMENTS IN THE CHINESE BANKING SECTOR Previous Studies Numerous previous studies have examined the causes of foreign divestment (Boddewyn, 1983; Buckley & Casson, 1976; Casson 1982; Dunning, 1980, 1988, 2000; Rugman, 1980). Among those studies, eclectic theory focuses on home country-side factors rather than on host country-side factors (Dunning, 1980, 1988, 2000).2 If a company has firm-specific advantages that generate excess profits (e.g., financial and managerial skills, product innovation, know-how, human capital resources, and so on), it will have an

Divestment of Foreign Strategic Investment in China’s Banking Sector

87

incentive to enter foreign markets and transfer the advantages to the foreign subsidiaries. Conversely, companies will exit from a foreign market if they face financial difficulties in their home country that damage their firmspecific advantages (Dunning, 2000). Several empirical studies that investigate international banks support this view.3 Williams (1996) finds that the market share of Japanese banks declined concurrently with the collapse of asset bubbles in Japan. Tschoegl (2004) argues that economic stagnation in a home country serves as a major parent bank-side factor associated with its exit from international markets. Peek and Rosengren (2000) show evidence that the position of Japanese banks in the US banking sector declined after the financial crisis during the early 1990s in Japan. Furthermore, Hryckiewicz and Kowalewski (2011) and Hryckiewicz et al. (2010) find that the closure of subsidiaries in developed countries is associated mainly with a decline in the financial performance of the parent bank in the home country rather than with problems of the foreign subsidiary in the host country. In contrast, some previous studies suggest that poor economic conditions in the host country give foreign banks an incentive to divest their equity stakes. Tschoegl (2005) argues that parent banks may sell the subsidiary when host country markets are depressed and the foreign owners find little benefit from staying in the country. For example, the Spanish Banco Bilbao Vizcaya Argentaria (BBVA) sold its Brazilian operations to Bradesco in 2003 because it was too costly for BBVA to keep the minimum asset size that generates profits. The Argentina Government announced on February 3, 2002 that all bank deposits denominated in dollars would be ‘‘pesified,’’ or converted to pesos at the government-imposed rate of 1.4 pesos to one dollar (the market exchange rate was approximately two pesos to one dollar at that point). In addition, all bank loans or other dollar debts would be converted into pesos at the rate of one peso to one dollar (Jacobs, 2003). Calomiris, Klingebiel, and Laeven (2005) suggest that several foreign banks exited from Argentina in 2002 because the ‘‘pesofication’’ policy substantially depreciated assets denominated in foreign currency and placed a heavy burden on foreign banks.4 It would be important to examine whether foreign divestments in China are attributable to the Chinese economic downturn during the global financial crisis. Why Do Foreign Financial Institutions Divest in China? As mentioned, there are two potential views on the causes of foreign divestments. If the divesting institutions’ deteriorating performance is a

88

YUHUA LI AND KONARI UCHIDA

more important cause than Chinese banks’ poor performance, divestments can be viewed as an exogenous shock for divested Chinese banks; this is consistent with our idea that emerging countries’ banking sectors are vulnerable to economic downturns in developed countries. In contrast, our notion is not supported if foreign investors divest their equity stakes in Chinese banks mainly due to the Chinese banks’ poor performance; in that case, it is difficult to accurately investigate the impact of foreign divestments on divested banks (endogeneity problem). To test this issue, we collect cases of foreign strategic investments and divestments in Chinese listed banks. We had access to information by Chinese banks, foreign financial institutions, and CBRC. We also had access to information from financial press websites as well as from Garcı´ a-Herrero and Santaba´rbara (2008) and Zhu, Zeng, Li, and He (2008). As a result, 26 foreign strategic investments in 14 Chinese listed banks were adopted as our strategic investment sample (Table 1). This indicates that reputable international banks, including Goldman Sachs (GS) and UBS, undertook strategic investments in Chinese banks during the period 2003–2006. In many cases, the strategic investment agreements comprise a lock-up period of three years or longer, and therefore foreign financial institutions commit to long-term relationships with a Chinese bank. Of the 26 strategic investments, divestments are conducted in 10 investments which involve five Chinese banks (divestment sample) (Table 2). In most cases, the news of their share selling was released without any preannouncements.5 Thus, we define the event day as the date on which foreign financial institutions cut their equity stakes. For the case of Newbridge’s (NB) divestment in Shenzhen Development Bank (SDB), NB concluded a share sale agreement with Ping An Insurance (Group) Company of China (Ping An) in which NB sold all SDB’s shares to Ping An by December 31, 2010. We do not find any news release that indicates NB actually sold SDB’s shares to Ping An, and therefore use the share sale agreement day as the event day. Table 2 indicates that the divestments are concentrated in the sevenmonth period from December 2008 to June 2009. The concentrated period suggests that the global financial crisis that originated in the US financial market made foreign financial institutions exit from strategic investments. It is noteworthy that divestments (excepting SDB divestments) took place on the Hong Kong Stock Exchange. Of the five divested banks, SDB experienced the largest divestment in terms of the percentage of divested shares over the total outstanding shares; NB held 17.89 percent of SDB’s outstanding shares before they sold all the stock. CCB represented the

89

Divestment of Foreign Strategic Investment in China’s Banking Sector

Table 1.

Foreign Strategic Investments in Chinese Listed Banks.

Chinese Divested Bank (Symbol)

Foreign Divesting Financial Institution (Symbol)

Strategic Investment Agreement Date

Bank of China (BOC)

UBS AG (UBS) Royal Bank of Scotland (RBS) Li Ka Shing Foundation (LKS) Asia Development Bank (ADB) Temasek Holdings Pte.Ltd. (TH) Bank of America (BofA)

2005/9/27 2005/8/18

500 1,600

3 years 3 years

2005/8/18

800

3 years

2005/10/1

75

3 years

2005/8/31

3,100

3 years

2005/6/17

2,500

TH Goldman Sachs (GS) Allianz American Express (AE) Newbridge Asia AIV III, L.P. (NB) TH Hang Seng Bank International Financial Corporation (IFC) Deutsche Bank Sal Oppenheim Citigroup

2005/7/1 2006/1/27 2006/1/27 2006/1/27 2004/5/29

146.6 2,582.22 1,000 200 150

3 years after IPO NA 3 years 3 years 3 years 5 years

2004/10/16 June 2005 2004/7/2

110 NA 23.5

NA NA NA

2005/10/17 2005/10/17 2003/1/1

233 96 73

5 years 5 years 5 years

174.7

NA

2006/11/22 2003/12/17

648 65

3 years NA

2003/12/17 2003/12/17 2005/10/12 2005/3/25 2005/5/25 2006/1/10

52 208 87 215 54 70

NA NA NA NA NA 10 years

China Construction Bank (CCB)

Industrial & Commercial Bank of China (ICBC) Shenzhen Development Bank (SDB) China Mingseng Bank (CMB)

Huaxia Bank (HXB) Shanghai Pudong Development Bank (SPDB) Bank of Communication (BofC) China CITIC Bank (CITIC) China Industrial Bank (CIB)

Bank of Nanjing (BON) Bank of Beijing (BOB) Bank of Ningbo (NBCB)

HSBC BBVA Government of Singapore Investment IFC Hang Seng Bank BNP Paribas ING IFC OCBC Bank

2004/8/6

Investment Amount (Million Dollars)

Lock-Up Period

Source: Annual reports, public news, Zhu et al. (2008) and Garcı´ a-Herrero and Santaba´rbara (2008). Note: This table shows a list of foreign strategic investments in Chinese listed banks.

90

Table 2.

Foreign Financial Institutions’ Divestments in the Chinese Banking Sector.

Chinese Divested Bank (Symbol)

Bank of China (BOC)

China Construction Bank (CCB) Industrial & Commercial Bank of China (ICBC)

UBS AG (UBS) Royal Bank of Scotland (RBS) Li Ka Shing Foundation (LKS) Bank of America (BofA) Bank of America (BofA) Goldman Sachs (GS) Allianz American Express (AE) Newbridge Asia AIV III, L.P. (NB) Temasek Holdings Pte. Ltd. (TH)

Typea

Percentage Ownership at the Strategic Investment (%)

Divested Shares Over Total Outstanding Shares (%)

31-12-2008 14-01-2009 7-01-2009

Whole Whole Partial

1.55 4.26 2.40

1.33b 4.26 0.8

7-01-2009 12-05-2009 1-06-2009 28-04-2009 28-04-2009 12-06-2009

Partial Partial Partial Partial Partial Whole

9.1 9.1 4.90 1.90 0.40 17.89

2.5 5.6 0.98 0.93 0.2 17.89

October 2008c

Whole

NA

NA

Divestment Event Date

Source: Annual reports, public news, Zhu et al. (2008) and Garcı´ a-Herrero and Santaba´rbara (2008). Note: This table shows cases of foreign financial institutions’ divestments in Chinese listed banks, NA, Not available. a Whole is divestments in which a foreign financial institution sells all of its shares of a Chinese bank. Partial is divestments in which a foreign financial institution sells a part of its shares of a Chinese bank. b UBS acquired 1.55 percent of BOC’ shares as a strategic investment in 2005. However, the percentage ownership was diluted to 1.33 percent by BOC’s initial public offerings in 2006. c Temasek Holdings Pte. Ltd. is a government holding company, and the exact event date is not available.

YUHUA LI AND KONARI UCHIDA

Shenzhen Development Bank (SDB) China Mingseng Bank (CMB)

Foreign Divesting Financial Institution (Symbol)

Divestment of Foreign Strategic Investment in China’s Banking Sector

91

largest total equity stake of foreign financial institutions before divestment (19.13 percent of total A- and H-shares). Although CCB experienced partial divestments, a large portion of shares (8.1 percent of total H-shares) was sold at divestment. Three foreign financial institutions had 8.21 percent of BOC’s shares before divesting; approximately 78 percent of those shares (6.39 percent of outstanding H-shares) were sold in three divestments that included two whole divestments. In contrast, Industrial Commercial Bank of China (ICBC) experienced relatively small divestments; only 2.1 percent of shares were sold in three divestments. Panel A of Table 3 presents performance measures of foreign divesting financial institutions over 2007–2009. As has been often cited, international banks suffered from a severe performance decline during the post-financial crisis period; UBS AG (UBS), Royal Bank of Scotland (RBS), and Allianz fell into the red for the year 2007–2008. UBS operates in many mortgage businesses, which were severely damaged by the collapse of the subprime market. RBS paid about US$140 billion for the acquisition of ABN Amro., and thereby reported US$25,455 million losses in 2008; this is the biggest loss in Britain’s history.6 Net income, ROE (return on equity), and EPS (earnings per share) of the Bank of America (BofA), GS, and American Express (AE) substantially declined in 2008, although they were still positive. It is noteworthy that BofA, GS, and AE received bailouts from the US Government’s Troubled Asset Relief Program (TRAP) fund at the end of 2008. This fact suggests that those three banks also experienced a critical situation during the post-financial crisis period. Importantly, foreign financial institutions that cut their whole equity stakes (UBS, RBS, and Temasek Holdings Pte. Ltd. (TH)) tended to perform more poorly than those that conducted partial divestments (BofA, GS, and AE). Especially in 2008, UBS and RBS reported ROE lower than 40 percent, which sharply contrasts with the AE’s ROE (22.3 percent). TH also reported relatively poor ROE in 2009, although its ROE was relatively high in 2008. In contrast, we do not find evidence that the performance of Chinese divested banks deteriorated during the event period. Panel B of Table 3 suggests that Chinese divested banks did not experience a serious decline in ROA (return on assets), ROE, and EPS in 2008. Importantly, we do not find a large difference in performance measures between divested and nondivested Chinese banks. For example, Huaxia Bank (HXB) and Guangdong Development Bank (GDB), which do not experience divestments, reported a ROA in 2007 that was lower than that of any divested banks (Panel C of Table 3). The results in Panel B suggest that Chinese banks that experience whole divestments tend to perform more poorly than

92

Table 3.

Performance of Foreign Divesting Financial Institutions and Chinese Banks. Panel A: Foreign divesting financial institutions

Net income (million US$) ROA (%)

ROE (%)

EPS (US$)

UBS (Whole)

RBS (Whole)

BofA (Partial)

GS (Partial)

Allianz (Partial)

AE (Partial)

2007 2008 2009 2007 2008 2009 2007 2008 2009 2007 2008 2009

2,997 25,455 2,365 0.23 0.99 0.16 10.5 58.7 7.8 2.00 7.05 0.69

15,428 63,311 3,621 0.38 1.56 0.17 13.78 50.57 5.98 0.13 2.68 0.99

14,982 4,008 6,246 0.94 0.22 0.26 11.08 1.8 4.18 3.29 0.54 0.29

11,599 2,322 13,385 1.18 0.23 1.54 32.7 4.9 22.5 24.73 4.47 22.13

11,927 3,200 6,037 0.80 0.22 0.56 15.0 9.7 12.7 24.24 8.00 13.20

4,012 2,699 2,130 2.89 1.96 1.70 37.3 22.3 14.6 3.36 2.32 1.54

TH (Whole) 12,976 22,474 9,089 3.99 6.77 2.28 8.9 14.1 4.7 NA NA NA

Panel B: Chinese divested banks

Net income (million RMB) ROA (%)

ROE (%)

Year

BOC (Whole/Partial)

CCB (Partial)

ICBC (Partial)

CMB (Whole)

SDB (Whole)

2007 2008 2009 2007 2008 2009 2007 2008

62,017 65,073 85,349 1.10 1.02 1.09 14.00 14.55

69,142 92,642 106,836 1.15 1.31 1.24 19.50 20.68

81,990 111,151 129,350 1.02 1.21 1.20 16.23 19.43

6,335 7,885 12,104 0.77 0.8 0.98 18.23 15.23

2,650 614 5,031 0.75 0.13 0.86 33.41 4.32

YUHUA LI AND KONARI UCHIDA

Year

2009 2007 2008 2009

16.62 0.22 0.25 0.32

20.87 0.30 0.40 0.46

20.15 0.24 0.33 0.39

20.19 0.36 0.42 0.63

26.59 0.97 0.20 1.62

Panel C: Chinese nondivested banks

Net income (million RMB)

ROA (%)

ROE (%)

EPS (RMB)

Year

HXB

SPDB

BofC

CITIC

CIB

BON

BOB

NBCB

2007 2008 2009 2007 2008 2009 2007 2008 2009 2007 2008 2009

2,000 3,060 3,698 0.41 0.46 0.48 17.12 18.23 13.04 0.50 0.70 0.75

5,499 12,516 13,217 0.69 1.13 0.90 19.43 30.03 19.45 0.694 1.579 1.621

20,323 28,524 30,118 1.07 1.20 1.01 18.85 20.87 19.49 0.42 0.58 0.61

8,290 13,262 14,320 0.97 1.08 0.94 14.30 13.27 12.91 0.24 0.40 0.37

8,586 11,385 13,282 1.17 1.22 1.13 25.34 26.06 24.54 1.75 2.28 2.66

909 1,456 1,544 1.16 1.94 1.43 15.95 13.71 13.23 0.62 0.79 0.84

3,348 5,417 5,634 1.07 1.40 1.19 23 18 16 0.63 0.86 0.90

951 1,332 1,457 1.44 1.49 1.09 18.15 15.91 15.79 0.43 0.53 0.58

Source: Annual reports. Note: This table shows performance measures of foreign divesting financial institutions (Panel A), divested Chinese banks (Panel B), and nondivested Chinese banks (Panel C). ROA is computed by net income divided by assets. ROE is computed by net income divided by book equity. EPS is earnings per share. NA: Not available. TH is a sovereignty company and EPS is not available.

Divestment of Foreign Strategic Investment in China’s Banking Sector

EPS (RMB)

93

94

YUHUA LI AND KONARI UCHIDA

those that experience partial divestments. However, even those banks show performance measures that are not substantially different from those of nondivested banks. Those results support the eclectic theory that divesting financial institutions’ factors are an important driver of foreign divestments in the Chinese banking sector (Peek & Rosengren, 1997, 2000; Tschoegl, 2004; Williams, 1996; Hryckiewicz & Kowalewski, 2011; Hryckiewicz et al., 2010). It is also noteworthy that divestments provide troubled foreign financial institutions with large gains. Table 4 shows that foreign financial institutions initially bought Chinese banks’ shares at low prices and thereby achieved high capital gains through divestments. Foreign financial institutions that suffered from deteriorating performance are likely to divest equity stakes in Chinese banks to achieve high gains.7 Indeed, it was reported that GS sold ICBC’s shares to increase its capital ratio.8 This fact reinforces the view that divestment is an exogenous shock for Chinese divested banks.

Table 4. Chinese Banks

BOC

CCB ICBC

SDB CMB

Divestment Returns.

Foreign Financial Institution

Initial Investment Value

Divestment Value

Type

Total Return (%)a

Annual Return (%)b

UBS RBS LKS BofAc BofAc GS Allianz AE NB TH

US$0.4916 bn US$1.573 bn US$0.8 bn US$2.5 bn US$2.5 bn HK$1.21/share HK$1.21/share HK$1.21/share US$150 mil. US$0.11 bn

US$0.808 bn US$2.383 bn US$0.524 bn US$2.8 bn US$7.3 bn HK$4.88/share HK$3.86/share HK$3.86/share US$1.5 bn NA

Whole Whole Partial Partial Partial Partial Partial Partial Whole Whole

64.4 51.5 96.3 307.7 374.5 303.3 219 219 464 NA

21.5 17.2 32.1 102.6 124.8 101.1 73 73 92.8 NA

Source: Annual reports and public news (Business Week, The Economist, The Wall Street Journal, Investors Journal, etc.). Note: This table shows realized returns of strategic investments for divesting financial institutions. NA: Not available, mil.: million, bn: billion. Bank names are denoted by their symbol. For formal names, see Table 1. a Total return ¼ (divestment value – initial investment value)/initial investment value. For return computation in partial divestments (LKS, BofA, GS, Allianz, and AE), we replace the initial investment value with the initial investment value  the number of divested shares/the number of initial invested shares. b Annual return is the total return over the lock-up year. See Table 1 for lock-up years.

Divestment of Foreign Strategic Investment in China’s Banking Sector

95

To further test the idea that poorly performing foreign financial institutions divest, we conduct probit regression analyses that adopt a dummy variable as a dependent variable which takes a value of one for divested strategic investments and zero for nondivested ones. The key independent variables are foreign financial institutions’ ROE (F_ROE) and Chinese banks’ ROE (C_ROE). Of the foreign strategic investments shown in Table 1, we adopt 19 investments for which F_ROE is available as a sample of this analysis. Results are shown in Table 5. Model (1) shows that F_ROE is negatively related to the likelihood that foreign strategic investments are divested, although the significance level is marginal. In contrast, C_ROE is not significantly associated with the probability of Chinese banks being divested. We can include only a small number of independent variables in the regression analyses due to the limited number of observations. We conduct several probit regression analyses that add one additional variable in the independent variable to F_ROE and C_ROE (models (2)–(4)). Model (2) adds the natural logarithm of foreign financial institution’s size (F_Ln(ASSET)) as an additional independent variable; the result suggests that a foreign financial institution’s size does not have a significant impact on the divestment probability. On the contrary, model (3) that includes the natural logarithm of a Chinese divested bank’s assets (C_Ln(ASSET)) engenders a positive and significant coefficient on C_Ln(ASSET); foreign financial institutions are more likely to divest large Chinese banks. In unreported analyses, we include the foreign financial institution’s type dummy (one when the foreign financial institution is a nonprofit organization (Asia Development Bank, International Financial Corporation, and Li Ka Shing Foundation) or a sovereign fund (TH and Government of Singapore Investment)). We find an insignificant coefficient on this variable. It is likely that foreign financial institutions divest from Chinese banks because they are not allowed to hold a substantial portion of Chinese bank’s shares. Eclectic theory suggests that investing firms can benefit from internalizing cross-border activities (Dunning, 1980, 1988, 2000). In other words, strategic investments become less attractive if investing firms have no opportunity to internalize benefits from cross-border activities. This situation is probably evident for investments in China’s banking sector because CBRC allows foreign financial institutions to hold only up to 25 percent of outstanding shares; in addition, no single foreign financial institution is allowed to have 20 percent or more equity stakes (Hope & Hu, 2005, 2006).9 There is anecdotal evidence that supports this view. Both UBS and BOC wanted to cooperate on private banking business when UBS

(1)

(2)

96

Table 5.

Regression Results.

(3)

(4)

(5)

(6)

(7)

Dependent Variable

Divestment ¼ 1 Nondivestment ¼ 0

Whole Divestment ¼ 2 Partial Divestment ¼ 1 Nondivestment ¼ 0

Estimation Method

Probit

Ordered Probit

(8)

F_ROE

–1.933 (–1.72)

–1.786 (–1.32)

–2.097 (–1.74)

–1.647 (–1.45)

–2.867 (–2.41)

–3.053 (–2.28)

–3.041 (–2.36)

–3.664 (–2.82)

C_ROE

–3.856 (–0.62)

–3.934 (–0.63) 0.038 (0.15)

–10.107 (–0.99)

–4.101 (–0.50)

–7.929 (–1.28)

–7.778 (–1.27) –0.049 (–0.19)

–17.706 (–1.48)

–11.034 (–1.32)

F_Ln(Asset)

FOWN Constant Pseudo-R2 N

0.400 (0.35) 0.0891 19

–0.109 (–0.03) 0.0899 19

–10.311 (–2.05) 0.4696 19

–26.339 (–2.26) 1.939 (1.18) 0.3973 18

0.732 (2.34)

0.1627 19

0.1637 19

0.3853 19

–31.615 (–2.34)

0.4506 18

Note: This table shows the regression results. The dependent variable is a dummy variable that takes a value of one for divested strategic investments and zero for others (models (1)–(4)), and a variable that takes a value of two for wholly divested strategic investments, one for partially divested strategic investments, and zero for nondivested strategic investments. Models (1)–(4) use probit regression, and models (5)–(8) adopt ordered probit regression for estimation. F_ROE is the foreign financial institution’s ROE for 2007. C_ROE is the Chinese bank’s ROE for 2007. F_Ln(ASSET) (C_Ln(ASSET)) is the natural logarithm of assets of the foreign financial institution (Chinese bank) for 2008. FOWN is the percentage ownership by the foreign financial institution for 2007. Significant at the 1% level; significant at the 5% level; and significant at the 10% level.

YUHUA LI AND KONARI UCHIDA

0.924 (2.62)

C_Ln(Asset)

Divestment of Foreign Strategic Investment in China’s Banking Sector

97

initially invested in BOC in 2005. However, UBS could not hold equity stakes in the private banking sector of BOC’s subsidiary (BOC International Limited China) because of regulations relating to joint ventures of private banking.10,11 In addition, the liberalization policy that originated in China’s entry into WTO potentially gives an incentive to conduct divestments to foreign financial institutions that were not permitted to hold substantial equity stakes in Chinese banks. In the WTO accession agreement, China was required to open the banking market to foreign financial institutions on December 11, 2006 (Leigh & Podpiera, 2006). This allows foreign financial institutions to establish incorporated banks in China, which can provide RMB denominated services to local Chinese customers without geographical restrictions (CBRC, ‘‘Report on the Opening-up of the Chinese Banking Sector,’’ January 25, 2007). This fact suggests that foreign financial institutions can internalize cross-border activities by establishing their own banks in China.12 To test the view that regulated low-equity stakes motivate foreign financial institutions to divest from Chinese banks, model (4) of Table 5 includes the percentage ownership by foreign financial institutions for 2007 (FOWN) as an additional independent variable. FOWN has a negative and significant coefficient; the result provides support for the notion that regulated low-equity ownership, which makes it difficult for foreign financial institutions to internalize the benefits of cross-border activities, is an important factor associated with foreign divestments. In this model, the F_ROE’s coefficient becomes insignificant. We do not simultaneously include C_Ln(ASSET) and FOWN in the independent variable because those two variables are highly correlated (correlation coefficient is 0.498). Foreign financial institutions are more likely to conduct whole divestments as they perform more poorly, given that divestments in China’s banking sector achieve large gains. To distinguish whole divestments from partial divestments, models (5)–(8) conduct ordered probit regressions in which the dependent variable takes a value of two for wholly divested strategic investments, one for partially divested strategic investments, and zero for nondivested ones. Models (5)–(8) engender a negative and significant coefficient on F_ROE. The result clearly suggests that the deteriorating performance of foreign financial institutions affects their decision to divest from Chinese banks; most poorly performing financial institutions conduct whole divestments. In contrast, C_ROE has a not significant coefficient in models (5)–(8). Those results that are consistent with previous studies (Hryckiewicz & Kowalewski, 2011; Hryckiewicz et al., 2010; Peek & Rosengren, 1997, 2000; Tschoegl, 2004; Williams, 1996)

98

YUHUA LI AND KONARI UCHIDA

provide support for the eclectic theory that argues that loss of firm-specific advantages of foreign firms causes foreign divestments. As with model (3), model (7) suggests that large Chinese banks are more likely to experience divestments (or whole divestments). Model (8) indicates that foreign financial institutions are more likely to conduct whole divestments when their ownership is low; again, difficulties in internalizing benefits from crossborder activities serve as a cause of divestments by foreign financial institutions. Overall, our results suggest that divestment is an exogenous shock for Chinese divested banks. It is significantly associated with the deteriorating performance of foreign financial institutions and their regulated low-equity ownership. We stress that divestments in the Chinese banking sector provide good material to measure the impact of foreign divestments on divested firms by avoiding endogeneity problems.

BUSINESS COOPERATION In strategic investments, foreign financial institutions are required to help Chinese banks in various dimensions. In exchange, Chinese banks would sell their equity stakes to foreign financial institutions at a low price. Indeed, Qingming Zhao, a senior analyst at CCB, said, ‘‘Maintaining a stake was the premise for mutual cooperation.’’13 Equity stakes are also likely to give foreign financial institutions an incentive to improve Chinese bank performance through cooperative works. These facts suggest the importance of tracing the change in business cooperation in accurately examining the impact of foreign divestments on the Chinese banking sector. We collect news on our sample banks from major business journals, such as The Wall Street Journal, The New York Times, Business Week, Bloomberg, the Financial Times, and Reuters. We also access major Chinese journals such as Securities Times and China Finance Information. The results are summarized in Table 6. In most of the whole divestments, business cooperation has been completely terminated (RBS and BOC; TH and CMB; NB and SDB). For example, RBS and BOC have stopped their cooperation in the area of personal banking, corporate finance, and the credit card business. NB gave advice on SDB’s management, nonperforming loans disposal, and human resource management as a result of strategic investments, but this cooperative relationship reportedly ended after the divestment. In the UBS’s divestment in BOC, there were conflicting reports about their business cooperation after divestment. However, we do not find

Chinese Divested Bank

BOC

Foreign Divesting Financial Institution

Cooperative Relationship between Foreign Divesting Financial Institutions and Chinese Divested Banks. Type

Cooperation on the Agreement

Post-divestment Cooperation (News Release Date)

UBS

Whole

UBS assists in investment banking

UBS said it remained committed to its business relationship with BOC and to its businesses in China as a whole. Bank of China’s Mr. Wang said the stake sale would not harm cooperation with UBSa (January 1, 2009). UBS and BOC basically do not have any cooperation projectb (January 2, 2009) BOC states that RBS and BOC have stopped their cooperation in the area of personal banking, corporate finance, and credit card businessd (March 5, 2009)

RBS CHINAc

Whole (RBS)/ Partial (LKS)

The consortium (RBS CHINA) assists BOC in credit cards, wealth management, corporate banking, and personal life insurance

CCB

BofA

Partial

Strategic assistance in areas of corporate governance, risk management, IT, financial management, human resource management, retail banking (including credit cards), and global treasury services. BofA will provide approximately 50 people to advise CCB

Both sides restate that the divestment would not affect their strategic cooperatione (January 7, 2009). Bank of America and CCB signed an agreement on January 15, 2009 to jointly set up 10 assistance projects and 53 experience-sharing projects in 2009. CCB will send 40 employees in 4 batches for training in the United Statesf (February 18, 2009)

ICBC

GS

Partial

ICBC states that GS is still a strategic investorg (September 30, 2009). GS agrees to extend the lockup-period for other nondivesting stakes by one more yearh (March 25, 2009)

Allianz

Partial

AE

Partial

GS assists ICBC to enhance corporate governance, risk management, and internal controls. It also provides expertise in treasury, asset management, corporate and investment banking, nonperforming loans disposal, and product innovation Allianz group cooperates with ICBC to provide insurance products and services to ICBC’s clients AE and ICBC develop their existing strategic cooperation in the credit card business

99

They restate their strategic cooperative relationshipi (March 25, 2009) They still plan to expand their strategic partnership, which includes credit card products development, marketing, risk management, customer service, and staff training. ICBC has been issuing American Express-branded credit cards in China since 2004j (April 28, 2009)

Divestment of Foreign Strategic Investment in China’s Banking Sector

Table 6.

100

Table 6. (Continued ) Chinese Divested Bank

Foreign Divesting Financial Institution

Type

Cooperation on the Agreement

Post-divestment Cooperation (News Release Date)

After NB sells its stake to Ping An, an insurance company in China, the cooperation between New Bridge Capital and SDB comes to an endk (June 18, 2009) The division of Small Medium Enterprises are not set up; basically no business cooperationl (March 12, 2009)

SDB

NB

Whole

Give comments and suggestions on risk management, financial management, nonperforming loan disposal, and human resource management

CMB

TH

Whole

Both want to cooperate on loans to Small Medium Enterprises

YUHUA LI AND KONARI UCHIDA

Source: Annual reports and public news. Note: Bank names are denoted by their symbol. For formal name, see Table 1. a The New York Times website: http://www.nytimes.com/2009/01/01/business/worldbusiness/01ubs.html b New Express Daily website, January 2, 2009: http://press.idoican.com.cn/detail/articles/20090102150A175/ c RBS and five other foreign financial institutions, which include LKS, co-established the RBS CHINA in 2005. As a strategic investor, they invested US$3.048 billion in BOC’s H-shares, which is about 8.48 percent of BOC’s total stock capital before its IPO. d Spokesman of Bank of China, Zhaowen Wang, http://www.gtja.com/f10.do?method¼xwzxDetail&type¼gsbd&guid¼{7F6DB1B2-D1364A28-9AD8-BB80F1845334} e Sina website: http://finance.sina.com.cn/stock/hkstock/redchipsnews/20090107/12025727712.shtml f Asia Pulse website: Limit on foreign equity in Chinese banks not to change: insider. February 18, 2009, http://www.highbeam.com/doc/1G1193873675.html g Reuters website (Chinese version): http://cn.reuters.com/article/cnMktNews/idCNnCN133671120100930 h Bloomberg website: http://www.bloomberg.com/apps/news?pid¼newsarchive&sid¼aaieHFciz648 i Securities Times website: http://epaper.stcn.com/paper/zqsb/html/2009-04/29/content_84178.htm j Financial Times website: http://www.ft.com/cms/s/0/14ee5830-33b1-11de-88cd-00144feabdc0.html#axzz1DnWQjwNj k The Economist website: Money from another time; TPG exits Shenzhen Development Bank, June 18, 2009, http://www.economist.com/ node/13871146 l China Finance Information website: http://www.cfi.net.cn/p20090312000888.html

Divestment of Foreign Strategic Investment in China’s Banking Sector

101

any news releases that indicate UBS and BOC engaged in a specific cooperative project after the divestment. As mentioned, divestment (or whole divestment) is associated with the foreign financial institution’s deteriorating performance that is attributable to the global financial crisis. It is less likely that a foreign financial institution will sell all its shares in Chinese banks for the purpose of ending cooperative relationships. This finding provides robust evidence that shareholdings have a large impact on cooperation between foreign financial institutions and emerging countries’ banks. Indeed, Qingming Zhao said ‘‘Now that this premise (maintaining an equity stake) is completely gone, how will they push forward the next step of cooperation?’’14 When a foreign financial institution keeps shareholdings (partial divestments), the strategic investment agreement requires them to continue cooperation with Chinese banks. For example, BofA and CCB signed an agreement on January 15, 2009 jointly setting up 10 assistance projects and 53 experience-sharing projects. CCB will send 40 employees to take training in the United States.15

STOCK MARKET REACTIONS TO THE DIVESTMENT ANNOUNCEMENT Foreign divestments are likely to decrease Chinese banks’ shareholder wealth for some reasons. BOC stock prices are likely to decline because its cooperative relationship with RBS ended. Even if business cooperation is continued, the reduced shareholdings will give foreign financial institutions weaker incentives to improve Chinese bank performance. This section investigates how foreign divestment affects Chinese bank performance by using a standard event study method (MacKinlay, 1997; McWilliams & Siegel, 1997). All foreign divestments, excepting NB’s divestments in SDB, are executed on the Hong Kong Stock Exchange. It is likely that H-share prices show a negative stock price response to divestment announcements simply because a large shareholder (foreign financial institution) sells a block of shares. In our research, it is essential to use stock returns that are isolated from the effect of selling pressure by divesting financial institutions. Chinese listed firms issue several classes of shares: A- (for domestic investors), B- (for foreign investors), and H-shares. Foreign investors are principally prohibited from dealing with A-shares, which are traded only in Chinese Renminbi at Chinese domestic stock exchanges (Shanghai Stock Exchange and

102

YUHUA LI AND KONARI UCHIDA

Shenzhen Stock Exchange). In contrast, H-shares are traded on the Hong Kong Stock Exchange, and the Hong Kong dollar is used for settlements. Importantly, A- and H-share prices of a single Chinese company are not strongly interrelated. Yeh and Lee (2000) find evidence that the return volatility of the Chinese domestic stock markets reacts more to good news than to bad news, which is opposite to the result on the Hong Kong Stock Exchange. Wang and Jiang (2004) show evidence that A-share returns are sensitive only to Shanghai (Shenzhen) market risk although H-share returns have exposure both to the Hong Kong and Shanghai (Shenzhen) market risks. Several recent studies show weak evidence that A- and H-shares prices became interrelated after the Asian financial crisis (Groenewold, Tang, & Wu, 2004; Wang & Di Iorio, 2007). However, Hatemi-J and Roca (2004) find no significant interrelation between the Chinese domestic market return and the Hong Kong market return even after the crisis; it is still a prevailing view that A-share prices do not behave in the same manner as H-share prices. In our research, all sample banks issue A-shares; an event study of A-shares allows us to investigate stock returns that are not contaminated, at least to a certain degree, by the selling pressure effect. Stock price data is obtained from RESSET/DB.16 We use H-share prices as well as A-share prices for the sake of comparison. We use returns of the Shanghai Composite Index, the Shenzhen Component Index, and the Hang Seng Index as a proxy for the market return of each stock exchange. Among 10 events shown in Table 2, the exact event day is not available for the divestment by TH, which is a sovereign holding company. Also, Allianz and AE divested their stakes from the same Chinese bank on the same day, so we treat those two events as a single event. SDB is listed on the Shenzhen Stock Exchange and not on the Hong Kong Stock Exchange. As a result, we conduct an event study for eight divestment announcements by using A-stock price data, and seven divestment announcements by using H-stock price data. We use a standard market model to compute abnormal returns for divested banks. Some banks experience divestments several times over a few months. For example, divestments in CCB were announced on January 7, 2009 and May 12, 2009. We conduct market model estimations by using two different estimation windows to avoid the first event affecting the second announcement: [170, 21] and [240, 91] (the event day is denoted by day 0). Three cumulative abnormal returns (CARs) are used to examine stock price reaction to divestment announcements: CAR[1, þ1]; CAR[2, þ2]; CAR[10, þ10]. We do not present standard mean and median tests because of the limited number of observations. Instead, we present CARs for every single event (Table 7).

103

Divestment of Foreign Strategic Investment in China’s Banking Sector

Table 7.

Stock Market Reaction to Divestment Announcements.

Divested Bank/Divesting Financial Institution

Panel A: CAR[–1, þ1] BOC/UBS BOC/RBS BOC/LKS CCB/BofA (January 7, 2009) CCB/BofA (May 12, 2009) ICBC/GS ICBC/Allianz, AE SDB/NB Mean Median Panel B: CAR[–2, þ2] BOC/UBS BOC/RBS BOC/LKS CCB/BofA (January 7, 2009) CCB/BofA (May 12, 2009) ICBC/GS ICBC/Allianz, AE SDB/NB Mean Median Panel C: CAR[–10, þ10] BOC/UBS BOC/RBS BOC/LKS CCB/BofA (January 7, 2009) CCB/BofA (May 12, 2009) ICBC/GS ICBC/Allianz, AE SDB/NB Mean Median

Type

Estimation Window: [–170, –21]

Estimation Window: [–240, –91]

H-Share

A-Share

H-Share

A-Share

Whole Whole Partial Partial Partial Partial Partial Whole

–0.0846 0.1164 –0.0333 –0.0865 –0.0052 –0.0152 0.0304 NA –0.0111 –0.0152

–0.0195 –0.0178 –0.0239 –0.0422 –0.0342 0.0229 0.0186 0.1212 0.0031 –0.0187

–0.0712 0.1054 –0.0488 –0.0780 –0.0065 –0.0151 0.0285 NA –0.0123 –0.0151

–0.0160 –0.0163 –0.0234 –0.0444 –0.0366 0.0193 0.0156 0.1172 0.0019 –0.0161

Whole Whole Partial Partial Partial Partial Partial Whole

–0.1122 0.1308 –0.0550 –0.0938 –0.0525 –0.0344 –0.0054 NA –0.0318 –0.0525

–0.0399 –0.0190 –0.0393 –0.0596 –0.0244 0.0469 0.0057 0.1075 –0.0028 –0.0217

–0.0873 0.1141 –0.0621 –0.0871 –0.0552 –0.0363 –0.0116 NA –0.0322 –0.0552

–0.0321 –0.0157 –0.0348 –0.0594 –0.0302 0.0406 0.0006 0.1022 –0.0036 –0.0229

Whole Whole Partial Partial Partial Partial Partial Whole

–0.1132 0.0006 –0.0171 –0.0170 –0.0879 0.0029 –0.1089 NA –0.0487 –0.0171

–0.0512 –0.0914 –0.0345 0.0021 –0.0098 0.1140 0.0274 0.1896 0.0183 –0.0039

–0.1197 –0.0157 –0.0512 0.0131 –0.1088 0.0076 –0.1379 NA –0.0589 –0.0512

–0.0401 –0.0758 –0.0320 –0.0139 –0.0305 0.0839 0.0066 0.1682 0.0083 –0.0222

Note: This table shows event study results. We use the stock return data from day 240 to 91 as an estimation window as well as the period from day 170 to 21. The event day is denoted by day 0. SDB is not listed on the Hong Kong Stock Exchange. Bank names are denoted by their symbol. For formal names, see Table 1.

104

YUHUA LI AND KONARI UCHIDA

Panel A presents three-day CARs (CAR[1, þ 1]). When we estimate the market model for the estimation window [170, 21], BOC H-shares showed 8.5 percent abnormal reduction during the three days surrounding UBS’s divestment date. When we extend the event window to 21 days (CAR[10, þ 10]; Panel C), the H-share abnormal stock price reduction becomes 11 percent. However, the negative excess return is likely to come from selling pressure by UBS. Interestingly, BOC’s H-shareholders experienced 11.6 percent excess return during the three days surrounding the RBS’s divestment day (Panel A). A possible interpretation of this result is that some investors aggressively bought the BOC shares. Reportedly, the Hopu Fund bought the 30 percent of shares sold by RBS. China Investment Corporation also bought BOC’s H-shares during the period surrounding the divestment day.17 However, the positive abnormal return almost disappears when we adopt the 21-day event window (Panel C). Those volatile abnormal returns suggest that H-share prices are highly affected by investors’ buying or selling behaviors on the divestment day; it is not appropriate to use H-share prices to measure the impact of foreign divestments on the intrinsic value of divested banks. Importantly, BOC’s A-shares, which are traded separately from H-shares, received a negative stock price response from the three divestments. In aggregate, BOC lost 6.12 percent of its A-share value during the three days surrounding the divestment; the reduction in value increases to approximately 10 percent for the 5-day window (CAR[2, þ2]). Those figures suggest that foreign divestments that sell approximately 6.4 percent of total shares have a substantial negative impact on BOC’s A-shareholder value. Especially for the BOC case, the ending of business cooperation potentially has a substantial negative impact on the domestic shareholder value. As with BOC, CCB experienced large block share selling (8.1 percent of total shares) in 2 foreign divestments. In aggregate, CCB’s A-shareholders lost 7.6 percent (8.4 percent) of their value during 3 (5) days surrounding the two divestment days. Given that both CCB and BofA announced a continuous business cooperative relationship on the divestment date (Table 6), the negative abnormal return is attributable to nonbusiness cooperation factors; foreign financial institutions’ decreased incentive to maximize Chinese banks’ shareholder value could be a source of the value reduction. Contrary to those cases, relatively small blocks of shares (approximately 2 percent of total shares) were sold in the two ICBC divestments. In addition, ICBC successfully continued business relationships with the foreign financial institutions even after the divestment. As a

Divestment of Foreign Strategic Investment in China’s Banking Sector

105

result, ICBC A-shareholders did not experience excess stock price reduction during the few days surrounding divestment day. We also present event study results for the SDB divestment by NB, which was conducted in A-shares. Unfortunately, SDB is not listed on the Hong Kong Stock Exchange and thus we cannot accurately examine impacts of the divestment; investors’ buying and selling behaviors will have a substantial impact on abnormal returns. Table 7 suggests that SDB’s A-shareholders received a positive excess return during the few days surrounding the divestment. The result is probably because investors simultaneously knew that Ping An would buy SDB’s shares. The two right columns show event study results when using the alternative estimation window ([240, 91]). The presented results are qualitatively unchanged. In total, BOC lost approximately 8.3 percent of its A-share values during the 5 days surrounding the three divestments. Similarly, CCB lost about 9 percent of A-share values during the same period from the two divestments. Again, we do not find a negative excess return for ICBC and SDB divestments. Overall, the event study results suggest that BOC and CCB, which experienced large foreign divestments, suffered from economically large declines in their A-share values. Divestments in China’s banking sector provide a clear case that the banking sectors of emerging countries have been exposed to economic slumps in developed countries. Conversely, the evidence of declined stock prices suggests that business cooperation and advising, which are enhanced by foreign strategic investments, generated substantial value for Chinese banks (and their domestic shareholders) before the divestment. The result provides evidence that foreign investments play an important role in the developments of an emerging country’s banking sector.

CONCLUSIONS Previous studies show that foreign bank entry or investments play a key role in the banking sector of emerging countries (Berger et al., 2009; Bhattacharya et al., 1997; Bonin et al., 2005a, 2005b; Goldberg & Saunders, 1981; Levine, 1996; Walter & Gray, 1983). On the contrary, eclectic theory suggests that foreign firms are likely to exit emerging markets if they face severe economic conditions in their home country. Indeed, several empirical studies present evidence that supports this view (Hryckiewicz & Kowalewski, 2011; Hryckiewicz et al., 2010; Peek & Rosengren, 1997, 2000; Tschoegl, 2004; Williams, 1996). These facts naturally give rise to the

106

YUHUA LI AND KONARI UCHIDA

prediction that emerging countries’ banking sectors are vulnerable to economic slump in developed countries. This chapter examines the causes and consequences of divestments by foreign financial institutions in China’s banking sector. Chinese banks have received foreign strategic investments and cooperated with foreign financial institutions on some businesses. However, since the global financial crisis, some foreign financial institutions have divested their equity stake. Chinese banks have remained listed even after divestment and allowed us to investigate the impact of foreign divestments on divested banks. In addition, several classes of shares issued by Chinese banks allow us to use stock price data that is not contaminated, at least to a certain degree, by selling pressure of divesting financial institutions. To the best of our knowledge, this chapter is the first attempt to investigate the impact of foreign divestments on divested banks. Among the causes of divestments proposed in previous studies, we find that poor performance and regulated low-equity stakes of foreign divesting financial institutions are important drivers of divestment (or whole divestment) in China. In contrast, we do not find evidence that Chinese bank performance affects the probability of being divested. These results suggest divestment is an exogenous shock for Chinese banks. We find that business cooperation is usually terminated when a foreign financial institution fully divests equity stakes in a Chinese bank. In addition, the Bank of China and China Construction Bank, which experienced large divestments, suffered from economically large declines in their domestic shareholder values. These results illustrate that emerging markets’ banking sectors are vulnerable to economic downturn in developed countries. The decreased shareholder value also implies that foreign strategic investments created substantial shareholder value before divestment. Our analysis shows that foreign financial institutions’ investments play an important role in an emerging market’s banking sector by taking advantage of a research environment that successfully avoids endogeneity problems (the divestments are exogenous shock for divested Chinese banks) and measures pure effects of foreign divestments.

NOTES 1. In contrast, many empirical studies have examined the impact of divestments on divesting firms. Haynes, Thompson, and Wright (2002), who use an unbalanced panel of 132 UK quoted companies over the period 1985–1993, find that divestment

Divestment of Foreign Strategic Investment in China’s Banking Sector

107

has a positive, significant, and substantial effect on the profitability of divesting companies. Several existing literatures investigate US firms which invested in South Africa to find cost and benefits of their equity divestment. Rudd (1979) analyzes the impact of a particular divestment strategy on the projected risk and return of the divesting firm. Wagner, Emkin, and Dixon (1984) find that divestment restrictions have a substantial impact on the investment management activities of divesting firms. Grossman and Sharpe (1986) find that the magnitude of financial costs, effects on risk, and expected return, depend materially on the divestment strategy of divesting firms. 2. Boddewyn (1983) suggests that foreign divestments can be treated as a reverse process of foreign direct investment (FDI). Buckley and Casson (1976), Rugman (1980), and Casson (1982) imply that the reduction of transaction costs in armslength dealings is a potential cause of divestment (internalization theory). 3. Chen and Wu (1996) present evidence that supports eclectic theory for nonfinancial industrial companies. Specifically, they find that capital intensity and export proportion contribute to early withdrawal of foreign direct investments in Taiwan. 4. For example, Scotiabank refused to provide additional capital to its subsidiary in Argentina (Scotiabank Quilmes). Credit Agricole abandoned its subsidiaries in Argentina. 5. We do not find any other relevant news for Chinese divested banks during the five days surrounding the event day. On the contrary, we find news on quarterly financial statements of foreign divesting institutions during the few days or weeks after the event day. However, it is less likely that the news has a substantial impact on the Chinese divested banks’ stock returns during the event window. 6. Reuters website: RBS posts record loss as UK insures toxic assets, Feb. 26, 2009, http://www.reuters.com/article/2009/02/26/us-britain-banks-idUSTRE51O2G 220090226 7. As mentioned in endnote 5, some foreign financial institutions announced news of their quarterly financial statements a few days or weeks after the announcement of divestment. This fact suggests that foreign financial institutions divest Chinese banks to boost their accounting earnings. 8. The New York Times website: Goldman’s China trip, March 27, 2009, http:// dealbook.nytimes.com/2009/03/27/goldmans-china-trip/ 9. Reportedly, foreign financial institutions have anticipated a removal of the 20 percent cap, but it did not come true (see McKinsey Quarterly website: Jensen P., Kehoe C., & Ramanathan B., Why Asia’s banks underperform at M&A, May 2010, https://www.mckinseyquarterly.com/Why_Asias_banks_underperform_at_MA_2604). Foreign financial institutions have also felt upset by their limited influence over Chinese banks (see The Wall Street Journal website: Jason L. & Dan F., As foreigners pull up stakes, China reassesses banks’ path, January 8, 2009, http://online.wsj.com/article/ SB123135303986861431.html?mod¼rss_whats_news_us&utm_source ¼ feedburner& utm_medium¼feed&utm_campaign¼Feed%3A þ wsj%2Fxml%2Frss%2F3_7011þ %28WSJ.com%3A þ What%27s þ News þ US%29). 10. New Express Daily website, January 2, 2009: http://press.idoican.com.cn/ detail/articles/20090102150A175/ 11. It is difficult for foreign financial institutions to get an approval from CBRC to acquire equity stakes of Chinese banks. Instead, foreign financial institutions

108

YUHUA LI AND KONARI UCHIDA

usually set up business centers, which serve as an independent unit, jointly with a Chinese bank. For example, Citigroup and Shanghai Pudong Development Bank (SPDB) founded a joint credit center. 12. BOC announced an exclusive partnership with RBS PLC’s private banking arm in 2007. However, it finally started its own proprietary wealth management business. The Wall Street Journal website: Jason L. & Dan F., As foreigners pull up stakes, China reassesses banks’ path, January 8, 2009, http://online.wsj.com/article/ SB123135303986861431.html?mod ¼ rss_whats_news_us&utm_source ¼ feedburner& utm_medium ¼ feed&utm_campaign ¼ Feed%3Aþwsj%2Fxml%2Frss%2F3_7011þ %28WSJ.com%3AþWhat%27sþNewsþUS%29 13. Reuters website: Foreign banks may regret rush out of China, January 15, 2009, http://www.reuters.com/article/2009/01/15/us-china-banks-stakes-idUSTRE50 E1J920090115 14. Reuters website: Foreign banks may regret rush out of China, January 15, 2009, http://www.reuters.com/article/2009/01/15/us-china-banks-stakes-idUSTRE50 E1J920090115 15. Asia Pulse website: Limit on foreign equity in Chinese banks not to change: insider. February 18, 2009, http://www.highbeam.com/doc/1G1-193873675.html 16. The data is retrieved from http://www.resset.cn/en/ 17. Securities Times website: http://epaper.stcn.com/paper/zqsb/html/2009-01/20/ content_58033.htm

ACKNOWLEDGMENTS We would like to thank Masaharu Kuhara for his comments and advice. This research is financially supported by JSPS Grants-in-Aid for Scientific Research and Kyushu University Interdisciplinary Programs in Education and Projects in Research Development.

REFERENCES Berger, A. N., Hasan, I., & Zhou, M. (2009). Bank ownership and efficiency in China: What will happen in the world’s largest nation? Journal of Banking and Finance, 33(1), 113–130. Bhattacharya, A., Lovell, C. A. K., & Sahay, P. (1997). The impact of liberalization on the productive efficiency of Indian commercial banks. European Journal of Operational Research, 98(2), 332–345. Boddewyn, J. J. (1983). Foreign direct divestment theory: Is it the reverse of FDI theory? Review of World Economics, 119(2), 345–355. Bonin, J. P., Hassan, I., & Wachtel, P. (2005a). Bank performance, efficiency and ownership in transition countries. Journal of Banking and Finance, 29(1), 31–53. Bonin, J. P., Hassan, I., & Wachtel, P. (2005b). Privatization matters: Bank efficiency in transition countries. Journal of Banking and Finance, 29(8–9), 2155–2178.

Divestment of Foreign Strategic Investment in China’s Banking Sector

109

Buckley, P. J., & Casson, M. (1976). The future of the multinational enterprise. London: Macmillan Press. Calomiris, C. W., Klingebiel, D., & Laeven, L. (2005). A taxonomy of financial crisis restructuring mechanisms: Cross-country experience and policy implications. In: L. Laeven (Ed.), Systemic financial distress: Containment and resolution (pp. 25–75). Cambridge: Cambridge University Press. Casson, M. (1982). New theories of the multinational enterprise. In: A. M. Rugman (Ed.), Transaction costs and the theory of the multinational enterprises. London: Croom Helm. Chen, T. J., & Wu, G. (1996). Determinants of divestment of FDI in Taiwan. Review of World Economics, 132(1), 172–184. Claessens, S., Demirgu¨c- -Kunt, A., & Huizinga, H. (2001). How does foreign entry affect domestic banking markets? Journal of Banking and Finance, 25(5), 891–911. Denizer, C. (2000). Foreign entry in Turkey’s banking sector, 1980–1997. In: S. Claessens & M. Jansen (Eds.), The internationalization of financial services: Issues and lessons for developing countries (pp. 389–406). Boston: Kluwer Academic. Dunning, J. H. (1980). Toward an eclectic theory of international production: some empirical tests. Journal of International Business Studies, 11(1), 9–31. Dunning, J. H. (1988). The eclectic paradigm of international production: A restatement and some possible extensions. Journal of International Business Studies, 19(1), 1–31. Dunning, J. H. (2000). The eclectic paradigm as an envelope for economic and business theories of MNE activity. International Business Review, 9(2), 163–190. Garcı´ a-Herrero, A., & Santaba´rbara, D. (2008). Does the Chinese banking system benefit from foreign investors. BOFIT Discussion Papers, 11. Goldberg, L. G., & Saunders, A. (1981). The determinants of foreign banking activity in the US. Journal of Banking and Finance, 5(1), 17–32. Goldberg, L. S., Dages, B. G., & Kinney, D. (2000). Foreign and domestic bank participation in emerging markets: Lessons from Mexico and Argentina. NBER working paper. Groenewold, N., Tang, S. H. K., & Wu, Y. (2004). The dynamic interrelationships between greater China share markets. China Economic Review, 15(1), 45–62. Grossman, B. R., & Sharpe, W. F. (1986). Financial implications of South African divestment. Financial Analysts Journal, 42(4), 15–29. Hatemi-J, A., & Roca, E. D. (2004). Do birds of the same feather flock together? The case of the Chinese states equity markets. Journal of International Markets, Institutions, and Money, 14(3), 281–294. Haynes, M., Thompson, S., & Wright, M. (2002). The impact of divestment on firm performance: Empirical evidence from a panel of UK companies. Journal of Industrial Economics, 50(2), 173–196. Hope, N., & Hu, F. (2005). Can foreign entry transform China’s banking system? Stanford Institute for Economic Policy Research, Stanford University, Policy Brief. Hope, N., & Hu, F. (2006). Reforming China’s banking system: How much can foreign strategic investment help? In: J. Aziz, S. Dunaway & E. Prasad (Eds.), China and India: Learning from each other: Reforms and policies for sustained growth (pp. 33–83). Washington, DC: IMF Multimedia Services Division. Hryckiewicz, A., & Kowalewski, O. (2011). Why do foreign banks withdraw from other countries? International Finance, 14(1), 67–102. Hryckiewicz, A., Kowalewski, O., & Niepodleglosci, A. (2010). Predicting foreign bank exits? A logit and probit regression approach. SSRN Working Paper.

110

YUHUA LI AND KONARI UCHIDA

Jacobs, B. L. (2003). Pesification and economic crisis in Argentina: The moral hazard posed by a politicized Supreme Court. University of Miami Inter-American Law Review, 34(3), 391–434. Leigh, L. & Podpiera, R. (2006). The rise of foreign investment in China’s banks: Taking stock. International Monetary Fund Working Paper, No. 06/292. Levine, R. (1996). Foreign banks, financial development, and economic growth. In: C. E. Barfield (Ed.), International financial markets: Harmonization versus competition (pp. 224–254). Washington, DC: American Enterprise Institute Press. MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of Economic Literature, 35(1), 13–39. McWilliams, A., & Siegel, D. (1997). Event studies in management research: Theoretical and empirical issues. Academy of Management Journal, 40(3), 626–657. Peek, J., & Rosengren, E. S. (1997). The international transmission of financial shocks: the case of Japan. American Economic Review, 87(4), 495–505. Peek, J., & Rosengren, E. S. (2000). Collateral damage: Effects of the Japanese bank crisis on real activity in the United States. American Economic Review, 90(1), 30–45. Rudd, A. (1979). Divestment of South African equities. Journal of Portfolio Management, 5(3), 5–10. Rugman, A. M. (1980). Internalization as a general theory of foreign direct investment: A reappraisal of the literature. Review of World Economics, 116(2), 365–379. Tschoegl, A. E. (2004). The California subsidiaries of Japanese banks: A genealogical history. Journal of Asian Business, 20(2), 59–82. Tschoegl, A. E. (2005). Financial crises and the presence of foreign banks. In: P. Honohan & L. Laeven (Eds.), Systemic financial crises: Containment and resolution (pp. 197–231). New York: Cambridge University Press. Wagner, W. H., Emkin, A., & Dixon, R. L. (1984). South African divestment: The investment issues. Financial Analysts Journal, 40(6), 14–22. Walter, I., & Gray, H. P. (1983). Protectionism, and international banking: Sectorial efficiency, competitive structure and national policy. Journal of Banking and Finance, 7(4), 597–609. Wang, Y., & Di Iorio, A. (2007). Are the China-related stock markets segmented with both world and regional stock markets? Journal of International Financial Markets, Institutions, and Money, 17(3), 277–290. Wang, S. S., & Jiang, L. (2004). Location of trade, ownership restrictions, and market illiquidity: Examining Chinese A- and H-shares. Journal of Banking and Finance, 28(6), 1273–1297. Weill, L. (2003). Banking efficiency in transition economies. Economics of Transition, 11(3), 569–592. Williams, B. (1996). Determinants of the performance of Japanese financial institutions in Australia 1987–1992. Applied Economics, 28(9), 1153–1165. Yeh, Y-H., & Lee, T-S. (2000). The interaction and volatility asymmetry of unexpected returns in the greater China stock markets. Global Finance Journal, 11(1–2), 129–149. Zhu, Y., Zeng, Y., Li, P., & He, J. (2008). Zhongzi yinhang yinjin zhanglue touzizhe: Beijing, zhenglun ji pingshu (Chinese banks attract overseas strategic investors: Background, arguments and comments). Management World, No. 2008-1 (in Chinese).

DOMESTIC AND FOREIGN INSTITUTIONAL INVESTOR BEHAVIOR IN CHINA: FINANCIAL CHARACTERISTICS AND CORPORATE GOVERNANCE Don Bredin and Ningyue Liu STRUCTURED ABSTRACT Purpose – We study the investment behavior of foreign institutional investors operating in China. A detailed analysis of foreign institutional investors is examined, along with a comparison of domestic Chinese investors. Methodology/approach – We adopt annual Chinese stock market data for the period 2003–2009 for both foreign and domestic funds to analyze the industrial preference of foreign funds and compare the different preferences between foreign funds and domestic Chinese funds in relation to financial characteristic and corporate governance indicators. Findings – The analysis reveals that foreign funds have a preference for a range of sectors such as transportation, metals and nonmetals, and machinery, as opposed to industries with a requirement for local

Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 113–143 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012007

113

114

DON BREDIN AND NINGYUE LIU

knowledge. The portfolios of domestic Chinese funds are distributed more evenly across sectors, compared to foreign funds. The comparative analysis reveals that the companies foreign funds invest in are significantly different from those firms favored by domestic funds in terms of size, profit, and management compensation. Social implications – These empirical findings highlight the differences between foreign and domestic funds investment preferences and has implications for policy makers aiming to attract foreign investors to emerging markets. Originality/value of chapter – Our chapter not only provides an introduction on the QFII scheme in China, but also examines the impact of a comprehensive range of firm-level characteristics, financial and corporate governance indicators, on the investment decisions of foreign and domestic funds in emerging markets. Keywords: Qualified foreign institutional investor; corporate governance; China stock market JEL classifications: G11; G34

INTRODUCTION An important factor in global capital markets is the dramatic expansion of funds managed by institutional investors. According to the International Financial Services London (2007), total assets under management by major global institutional investors reached 81.9 trillion US dollars by the end of December 2007. Institutional investment grew from 6.1% of aggregate ownership of equities in 1950 to more than 50% by 2002 in the US market.1 Moreover, the role of institutional investors is also rapidly developing in emerging markets (Khorana, Servaes, & Tufano, 2005). In particular, the limited availability of external finance (Giannetti & Koskinen, 2010) and the recent liberalization to allow foreign institutional investors access to emerging markets had become a key source of financing (Bekaert, Harvey, & Lumsdaine, 2002). Although foreign direct investment (FDI) has had a dramatic impact on economic development in China, it is only since 2003 that the Chinese government has permitted foreign institutional investors to directly invest in

115

Domestic and Foreign Institutional Investor Behavior in China

Chinese securities market (Cheng & Kwan, 2000; Zhang, 2001). After China became a member of the World Trade Organization (WTO) in December 2001, it implemented numerous measures to liberalize its economy and improve its investment environment. One of the most significant measures has been the Qualified Foreign Institutional Investor (QFII) scheme, which is designed to allow the largest overseas institutions access to China’s debt and equities markets. The QFII scheme represents a significant departure from China’s traditional approach of strict capital controls. Us of September 30, 2010, 93 QFIIs had been approved by China Securities Regulatory Commission (CSRC).2 The total investment quota of QFIIs has grown from 425 million US dollars at the beginning of the scheme in 2003, to 19 billion US dollars by the end of September 2010 (Fig. 1). In spite of the exceptional growth, there is limited research on how QFIIs determine the allocations across different listed companies and the factors influencing their investment behavior. In this chapter, we provide an introduction to the QFII scheme in China and examine the firm-level characteristics of stocks that fund managers invest in. Specifically, we aim to address two questions. Firstly, what are the industrial preferences of the foreign funds? Is it the case that foreign funds 200

total investment quota of QFIIs (100 million US dollars)

180

160

140

120

100

80

60

40

20

0 Jun 03

Jan 04 Jun 04

Jan 05 Jun 05

Fig. 1.

Jan 06 Jun 06

Jan 07 Jun 07 month

Jan 08 Jun 08

Jan 09 Jun 09

Investment Quota of QFIIs in China.

Jan 10

Sep 10

116

DON BREDIN AND NINGYUE LIU

have a preference for manufacturing industries which are consistent with the preference of foreign institutional investors in developed markets (see Kang & Stulz, 1997) or does the Chinese market have special features? Secondly, how do these firm-level characteristics compare to that domestic funds invest in? The question of whether foreign and domestic institutional investors have the same investment behavior is increasingly controversial. One argument is that foreign institutional investors are more sophisticated investors than their domestic counterparts (Grinblatt & Keloharju, 2000). An alternative is that they are regarded as equally sophisticated but not as well informed (Covirg, Lau, & Ng, 2006). Therefore, foreign investors will invest in companies with different characteristics compared to those that domestic funds invest in. Conversely, Chang (2010) highlights that foreign investors in emerging markets might be expected to suffer from an informational disadvantage given a lack of local knowledge and contacts. For example, foreign investors need to take account of China’s unique characteristics, such as insider control and the influential role of government policy (Naughton, 2007).3 This information asymmetry, combined with China’s unique characteristics may motivate foreign institutional investors to choose the ‘‘free rider strategy’’ and follow the investment behavior of domestic funds. Following influential work of La Porta, Lopez-de-Silanes, Shleifer, & Vishny (1997, 1998, 2000) recent research on the preference of foreign investors has mainly focused on the impact of macroeconomic and other country-level factors. The authors find that stronger investor protection laws, high enforcement and high quality accounting disclosures have a positive impact on market development. Chan, Covrig, and Ng (2005) conclude that economic development, capital controls, and withholding tax variables have significant effects on foreign investors’ investment allocation. A number of studies have examined firm-level allocations, but the vast majority focus on developed countries. For example, Falkenstein (1996), Kang and Stulz (1997), and Dahlquist and Robertsson (2001) investigate foreign investment in the United States, Japan, and Swedish market, respectively. Given the short history of emerging stock markets and the difficulty of obtaining data, there is a considerable dearth of empirical evidence with regard to emerging markets. However, since developed markets have higher standards of information disclosure, the information asymmetry between foreign and domestic investors in these markets is less severe. It is therefore important to examine if investors in other markets, especially emerging markets, exhibit a different preference compared to domestic investors. The rate of economic growth in China in recent years has been dramatic, with an average annual growth rate of around 10% from 1978 to 2010.4

Domestic and Foreign Institutional Investor Behavior in China

117

China’s gross domestic product (GDP) reached 5.7 trillion US dollars in 2009 and is now the world’s third largest economy, second only to the Eurozone and the United States.5 The combined market capitalization of China’s two domestic stock exchanges has grown from 104 billion Chinese Yuan (CNY) in 1992 to more than 26 trillion CNY by the end of 2009.6 The number of listed companies has grown from 14 in 1991 to 2,063 by the end of 2010. Meanwhile, 133 million trading accounts had been established in China’s equity market.7 Covirg et al. (2006) believe that foreign fund managers are equally sophisticated as their domestic counterparts but might not be equally informed. This is likely to hold in their sample of developed countries, where they expect insider trading to be less prevalent. On the contrary, Naughton (2007) concludes that insider control and manipulation are one of the Chinese stock market’s characteristics. For example, some agents, like managers of state-owned firms, securities companies, regulators, are able to manipulate prices and obtain profits from them. In addition to insider control, weak disclosure and regulation and policy-driven characteristics are evident in the Chinese stock market (Naughton, 2007). These aspects of the Chinese market stand in stark contrast to the case of developed markets and point to a worthwhile avenue of research. Our chapter examines the impact of a comprehensive range of firm-level characteristics, financial and corporate governance indicators, on the investment decisions of foreign and domestic funds. Although financial characteristics have been investigated by previous studies, our comprehensive range of indicators offers a more robust analysis. In particular, a number of additional financial aspects are included, such as operating ability, which is an important indicator used to evaluate the financial condition of a company. Moreover, McKinsey and Company (2002) highlights in a recent survey the importance of governance to the investment decisions in the global market. The report also highlights that corporate governance considerations dominate any other issues when it comes to investment decisions in East Asia. It is therefore important to examine the impact of corporate governance on the investment decisions of foreign and domestic funds. Our results will be of value to policy makers in both developed and developing markets in gauging the important drivers of foreign investment. Given the increasing significance of foreign financing and the fact that access to foreign capital is uneven across firms, it is critical to fully understand the factors that influence investors’ behavior. Moreover, Huang and Shiu (2006) find that stocks with high foreign ownership outperform stocks with low

118

DON BREDIN AND NINGYUE LIU

foreign ownership. Our findings will be particularly relevant to policy makers and firms in creating an environment conducive to foreign investment. The chapter adopts annual Chinese stock market data for the period 2003–2009 for both foreign and domestic funds. Our results indicate that foreign funds prefer sectors such as transportation, metals and nonmetals, and machinery as opposed to industries requiring local knowledge, e.g. real estate, construction, and media and culture. The portfolios of domestic funds are distributed more evenly across sectors compared to foreign funds. This chapter examines the impact of a comprehensive range of firm-level characteristics, financial and corporate governance indicators, on the investment decisions of foreign and domestic funds. Characteristics, such as size, profit, and management compensation are significantly different for firms that foreign funds invest in, compared to companies preferred by domestic funds. However, there is no statistical difference in relation to the following characteristics: short-term liability paying capability, growth, operating capability, and management structure. These empirical findings highlight the differences between foreign and domestic funds investment preferences, and have implications for both policy makers and regulators alike. In particular our results have important implications for policy makers aiming to attract foreign investors to invest in emerging markets. The outline of the rest of this chapter is as follows. In the second section, a brief review of the literature on the preference of foreign institutional investors is provided, before introducing the case of China and the related research. In the third section we discuss the data and indicators used in this chapter. In fourth section, we present the empirical results including industrial distribution and comparative analysis of financial and corporate governance indicators between the companies foreign and domestic funds invest in. Finally, fifth section provides some concluding remarks.

LITERATURE REVIEW Preference of Foreign Investors Frenkel and Poterba (1991), Cooper and Kaplanis (1994), and Tesar and Werner (1995) document that although the barriers to international investment have declined dramatically, foreign ownership of shares is still extremely limited and much smaller than one would expect even in the absence of barriers to international investment. Several explanations have

Domestic and Foreign Institutional Investor Behavior in China

119

been suggested for this so-called home bias in portfolios. Typically, researchers compare the aggregated holdings of investors in foreign markets with their domestic holdings (Lewis, 1999). Kang and Stulz (1997) analyze this issue from a different perspective and study the shareholdings of foreigners in individual firms in a specific market. They adopt data for Japan from 1975 to 1991 and find that foreign investors hold disproportionately more shares of firms in manufacturing industries, large firms, and firms with good accounting performance, low unsystematic risk, and low leverage. Dahlquist and Robertsson (2001) analyze the determinants of foreign ownership in Swedish firms and find that foreigners have a preference for large firms, firms paying low dividends, and firms with large cash positions on their balance sheets. Foreign investors also tend to underweight firms with a dominant owner.8 Covirg et al. (2006) investigate a range of stock preferences of domestic and foreign fund managers from 11 developed countries and conclude that foreign fund managers have less information about the domestic stocks than their domestic counterparts. They find that ownership by foreign funds is related to the size of foreign sales, index membership and stocks with foreign listing. Li and Jeong-Bon (2004) find that foreign investors in Japan tend to avoid stocks with high crosscorporate holdings. The authors suggest that foreign institutional investors are likely to be efficient processors of public information and are attracted to Japanese firms with low-information asymmetry. Fast growing emerging markets are clearly attractive to foreign institutional investors and these were the most important source of capital for emerging markets, in particular during 1990s (Frenkel & Menkhoff, 2003). Aggarwal, Klapper, and Wysocki (2005) examine the portfolio holdings of 576 US mutual funds that invested in emerging markets (up to February 2002). The authors analyze both country and firm-level disclosure and the institutional policies that influence mutual funds’ allocation choices. At the firm level, US funds have a preference for firms that adopt discretionary policies such as greater accounting transparency and the issuance of an American Depositary Receipt (Dahlquist & Robertsson, 2001).

QFII: A Formal Classification and Recent Developments QFII is defined as ‘‘overseas fund management institutions, insurance companies, securities companies, and other assets management institutions which have been approved by CSRC to invest in China’s securities market

120

DON BREDIN AND NINGYUE LIU

and granted investment quota by SAFE’’.9 This definition can be found in ‘‘Provisional Measures on Administration (2002) of Domestic Securities Investments of Qualified Foreign Institutional Investors’’ issued by CSRC on November 5, 2002. Term 18 of the Provisional Measure states that QFIIs can invest in A-shares, treasuries, convertible bonds, and corporate bonds listed in China’s stock exchanges and other financial instruments as approved by CSRC.10 The requirement for QFII’s qualification states: for fund managers, they must have assets in excess of 10 billion US dollars during the latest accounting year and have operated for over 5 years. For insurance companies, they must have assets in excess of 10 billion US dollars during the latest accounting year and have operated for over 30 years. For commercial banks, they must be ranked in the world top 100 banks in terms of assets and have assets in excess of 10 billion US dollars during the latest accounting year as well. Hence only large foreign institutional investors are qualified to apply as QFIIs. Two important conditions of the regulations influencing the investment of QFIIs arise from Term 20 (1), which states that ‘‘shares held by each QFII in any one listed company should not exceed 10% of the total outstanding shares of the company’’ and Term 20 (2), which indicates that ‘‘total shares held by all QFIIs in one listed company should not exceed 20% of the total outstanding shares of the company.’’ The first three QFIIs approved in June 2003 were UBS, Nomura Securities, and Citigroup Global Markets and were soon followed by Morgan Stanley International and Goldman Sachs. QFIIs are all large international institutions from major developed countries, e.g. the United States, the United Kingdom, Germany, France, Japan, Canada, the Netherlands, and Switzerland. The details of QFIIs’ approved date and investment quota are described in Table 1. CSRC, People’s Bank of China (PBC) and SAFE jointly issued new regulations, entitled the ‘‘Measures for the Administration of Investment in Domestic Securities by Qualified Foreign Institutional Investors’’ (the ‘‘New QFII Rules’’) on August 24, 2006. The New QFII rules supersede the original QFII rules and took effect from September 1, 2006. The qualifying criteria in terms of assets under management for QFII applicants that are fund management institutions has been reduced from 10 billion US dollars to 5 billion US dollars during the latest accounting year. This will enable more fund management companies to apply on their own for QFII approval. The qualifying criteria for insurance companies is changed from at least 10 billion US dollars to 5 billion US dollars in securities assets held in the most recent accounting year. In addition, while the old rules did not

121

Domestic and Foreign Institutional Investor Behavior in China

Table 1. QFII’s Name

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

UBS AG Nomura Securities Co. Limited Citigroup Global Markets Limited Morgan Stanley & Co. International Limited Goldman, Sachs & Co. The Hongkong and Shanghai Banking Corporation Limited Deutsche Bank AG or Deutsche Bank Aktiengesellschaft ING Bank N. V. Jpmorgan Chase Bank Credit Suisse (HongKong) Limited Nikko Asset Management Co., Limited Standard Chartered Bank (Hong Kong) Limited Hangseng Bank Daiwa Securities SMBC Co., Limited Merrill Lynch International Lehman Brothers International (Europe) Bill & Melinda Gates Foundation ABN AMRO Bank N.V. Socit Gnrale Barclays Bank PLC BNP Paribas Dresdner Bank Aktiengesellschaft Fortis Bank SA/NV Power Corporation of Canada Calyon S.A. INVESCO Asset Management Limited Government of Singapore Investment Corporation Pte Limited Goldman Sachs Asset Management International Martin Currie Investment Management Limited Temasek Fullerton Alpha Investments Pte Limited

QFII List. Approved Date

Investment Quota

Investment Quota

(Approved Date)

(September 2010)

(100 million US dollars)

(100 million US dollars)

04/06/2003 04/06/2003 18/06/2003 01/07/2003

3.00 0.50 0.75 3.00

8.00 3.50 5.50 4.00

24/07/2003 26/08/2003

0.50 0.50

3.00 4.00

26/08/2003

0.50

4.00

16/10/2003 04/11/2003 28/11/2003 09/02/2004

1.00 0.50 0.50 0.50

4.00 1.50 5.00 4.50

19/05/2004

0.75

0.75

22/06/2004 05/07/2004 16/07/2004 16/08/2004

0.50 0.50 0.75 0.75

1.00 0.50 3.00 2.00

28/08/2004 17/09/2004 17/09/2004 15/10/2004 27/10/2004 08/11/2004 21/11/2004 21/11/2004 10/01/2005 08/03/2005 16/11/2005

1.00 0.75 0.50 0.75 0.75 0.75 1.00 0.50 0.75 0.50 1.00

3.00 1.75 0.50 4.00 2.00 0.75 5.00 0.50 0.75 2.50 3.00

16/11/2005

2.00

5.00

24/11/2005

1.20

1.20

12/12/2005

1.00

3.00

122

DON BREDIN AND NINGYUE LIU

Table 1. (Continued ) QFII’s Name

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

AIG Global Investment Corporation The Dai-ichi Mutual Life Insurance Company DBS Bank Limited JF Asset Management Limited KBC Financial Products UK Limited Scotia Bank or The Bank of Nova Scotia La Compagnie Financierr Edmond de Rothschild Banque Yale University AMP Capital Investors Limited Morgan Stanley Investment Management Inc. Prudential Asset Management (Hong Kong) Limited Stanford University United Overseas Bank Limited Schroder Investment Mangement Limited GE Asset Management Incorporated UBS Global Asset Management (Singapore) Limited Shinko Securities Co. Limited HSBC Investments (Hong Kong) Limited Sumitomo Mitsui Asset Management Company, Limited Norges Bank Pictet Asset Management Limited The Trustees of Columbia University in the City of New York Prudential Asset Management Co. Limited Robeco Institutional Asset management B.V. KBC Asset Management N.V. Mirae Asset Investment Management Co. Limited Platinum Investment Company Limited

Approved Date

Investment Quota

Investment Quota

(Approved Date)

(September 2010)

(100 million US dollars)

(100 million US dollars)

12/12/2005 22/02/2006

0.50 1.00

0.50 2.00

12/04/2006 12/04/2006 09/06/2006 09/06/2006

1.00 1.50 1.00 1.50

1.00 2.75 1.00 1.50

19/07/2006

1.00

1.00

01/08/2006 01/08/2006 05/09/2006

0.50 2.00 2.00

1.50 3.00 4.50

12/10/2006

2.00

3.00

07/11/2006 07/11/2006 11/12/2006

0.50 0.50 2.00

1.00 0.50 2.00

11/01/2007 11/01/2007

2.00 2.00

3.50 2.00

13/02/2007 13/02/2007

0.50 2.00

0.50 3.50

13/02/2007

2.00

3.50

24/01/2008 01/04/2008 07/04/2008

2.00 1.00 1.00

7.00 1.00 1.00

04/05/2008

0.75

0.75

20/06/2008

1.50

1.50

31/07/2008 02/09/2008

1.50 1.50

1.50 2.50

10/09/2008

1.50

1.50

123

Domestic and Foreign Institutional Investor Behavior in China

Table 1. (Continued ) QFII’s Name

58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

State Street Global Advisors Asia Limited Caisse de d pot et placement du Qu bec Samsung Investment Trust Management Co., Limited Oversea-Chinese Banking Corporation Limited AllianceBernstein Limited ACE INA International Holdings Limited President and Fellows of Harvard College T. Rowe Price International, Inc. DAIWA Asset Management Co. ABU Dhabi Investment Authority Allianz Global Investors Luxembourg S.A. Mitsubishi UFJ Securities Co. Limited Capital International inc. Credit Suisse Emerging Markets Management, L.L.C. First State Investment Management (UK) Limited Hanwha Investment Trust Management Co., Limited UOB Asset Management Limited Bank Negara Malaysia DWS Investment S.A. Lloyd George Management (Hong Kong) Limited The Korea Development Bank Templeton Investment Counsel, LLC Shell Asset Management Company B.V. BEA Union Investment Management Limited Woori Bank Co. Limited Korea Investment Trust Management Co. Limited The Sumitomo Trust & Banking Co. Limited

Approved Date

Investment Quota

Investment Quota

(Approved Date)

(September 2010)

(100 million US dollars)

(100 million US dollars)

03/11/2008

0.50

0.50

03/11/2008 07/11/2008

2.00 1.50

2.00 3.00

12/11/2008

1.50

1.50

12/11/2008 13/11/2008

0.50 1.50

1.50 1.50

14/11/2008

2.00

2.00

03/12/2008 26/12/2008 17/01/2009 04/03/2009

1.10 1.00 2.00 1.00

1.10 1.00 2.00 1.00

25/03/2009 31/03/2009 22/05/2009 03/06/2009

1.00 1.00 2.00 0.50

1.00 1.00 2.00 0.50

16/06/2009

1.20

1.20

10/08/2009

0.70

0.70

25/08/2009 04/09/2009 09/09/2009 06/11/2009

0.50 2.00 2.00 0.50

0.50 2.00 2.00 0.50

09/11/2009 08/12/2009 08/12/2009 08/12/2009

1.00 2.00 1.00 1.00

1.00 2.00 1.00 1.00

30/12/2009 30/12/2009

0.50 1.00

0.50 1.00

31/12/2009

0.50

0.50

124

DON BREDIN AND NINGYUE LIU

Table 1. (Continued ) QFII’s Name

86 87 88 89 90 91 92 93

Baring Asset Management Limited Ashmore Investment Management Limited Nomura Asset Management Co. Limited Manulife Asset Management (Hong Kong) Limited Tongyang Investment Trust Management Co., Limited Royal Bank of Canada Ivy Investment Management Company DIAM Co. Limited Total

Approved Date

Investment Quota

Investment Quota

(Approved Date)

(September 2010)

(100 million US dollars)

(100 million US dollars)

10/02/2010 10/02/2010

2.00 2.00

2.00 2.00

04/05/2010

2.00

2.00

31/05/2010

2.00

2.00

22/07/2010

1.00

1.00

19/08/2010 01/09/2010 01/09/2010

1.00 1.00 1.00

1.00 1.00 1.00

108.2

190.0

Notes: In this table, we report the list of QFIIs from June 2003 to September 2010. The source is from Investment Quota Approval Form of QFIIs issued by SAFE, November 10, 2010. The last two columns are the investment quota of each QFII at approved date and investment quota by the end of September 2010, respectively. The units are 100 million US dollars in each column. Up to September 30, 2010, there were 93 QFIIs approved by CSRC and SAFE and the investment quotas were 19 billion US dollars.

specifically cater for other categories of institutional investors, it is now provided in the Implementing Notice that other institutional investors (pension funds, charitable funds, donation funds, trust companies, and government investment companies) are subject to the qualifying criteria of having been established for at least five years and having assets under management or having a securities portfolio with at least 5 billion US dollars in the most recent accounting year. Furthermore, on September 29, 2009, SAFE released ‘‘the Provisions on Foreign Exchange Administration of Domestic Securities Investment by Qualified Foreign Institutional Investors’’ which came into effect on that date. The maximum accumulated investment quota of one single QFII has been increased to 1 billion US dollars from 800 million US dollars. The New QFII Rules contain significant improvements from the original rules, particularly for fund management companies wishing to invest in the Chinese domestic securities

Domestic and Foreign Institutional Investor Behavior in China

125

market through the QFII scheme. The positive changes under the New QFII Rules open the way to the development of more China investment fund products.

Empirical Evidence on QFII The QFII system was introduced in Taiwan in the late 1980s and began in 1991, when Taiwan’s stock market was particularly popular with foreign investors (Dean, 2003). Research on QFIIs in Taiwan is divided into two main categories. One area of research has examined the impact of QFIIs on Taiwan’s stock market and local companies’ performance (Huang & Shiu, 2006; Lin & Chen, 2006). The other avenue of research has investigated the behavior of QFIIs, e.g. the extent of momentum and herd behavior (Lai, Lou, & Shiu, 2008; Lin & Swanson, 2003; Lu, Fang, Wong, & Wang, 2009). Lin and Shiu (2003) investigate foreign ownership in the Taiwan stock market form 1996 to 2000 and find that foreign investors favor large firms and low book-to-market stocks. Results also indicate that foreign investors have a preference for firms with high export ratios with which they are more familiar, given their higher foreign sales. Furthermore, Korea announced guidelines for the limited opening up of its equities market to foreign investment in 1991. Using Korean data, Choe, Kho, and Stulz (2005) show that foreign institutional investors pay more than domestic counterparts when they buy and receive less when they sell for medium and large trades which indicate that domestic individual investors have an edge over foreign investors. The extent of research on QFIIs in China is limited given the difficulties in accessing data and the relatively short time period since QFII scheme began. Most of the papers to date provide an introduction to the scheme, e.g. Yeo (2003) introduces the QFII scheme in China and compares it with other Asian markets. An example of an empirical study is Ting Yen and Chiu (2008). The authors examine the relationship between audit opinions and default probability within the Chinese stock market and find audit opinions begin providing signals of potential default risk only after QFIIs entered the market. Chen and Yu (2003) investigate the market reaction around the announcement of the QFII scheme and find no significant abnormal returns in the market indices in the short-term period leading up to the announcement, negative abnormal returns in the short-term period following the announcement, and no significant abnormal returns in the long-term period thereafter.

126

DON BREDIN AND NINGYUE LIU

DATA ISSUES AND KEY CHARACTERISTICS Data Description The shareholder data of foreign and domestic funds are all sourced from the Wind database, whereas financial and corporate governance data are taken from the China Stock Market Accounting Research (CSMAR) at GTA Research Service Center.11 The data is annual and covers the period 2003–2009. In order to investigate the behavior of foreign and domestic funds, our sample only includes those companies whose stocks are held by either foreign or domestic funds or both. Table 2 presents the breakdown of both foreign and domestic funds for individual years. Over this period, the number of foreign and domestic funds grew dramatically. The average numbers of firms, foreign, and domestic funds invest in each year are 123 and 970, respectively. Following the global economic downturn, the number of firms whose shares are partially held by QFIIs fell in both 2007 and 2008.12 Industry Characteristics Firms are classified in 13 industrial sectors using code provided by CSRC. Given foreign investors traditionally hold more shares of firms in manufacturing industries (Kang & Stulz, 1997) and the large number of Table 2. Year

Distribution of Funds by Year.

QFII

Domestic Fund

Number of QFIIs

Number of Firms QFIIs Invest in

Number of Domestic Funds

Number of Firms Domestic Funds Invest in

2003 2004 2005 2006 2007 2008 2009

10 24 31 44 49 66 85

17 35 122 196 154 124 210

110 161 218 301 346 439 557

516 1049 1062 1113 893 887 1273

Mean

44

123

305

970

Notes: This table shows the distribution of QFIIs and domestic funds in terms of year. The information on amounts of QFIIs and domestic funds is obtained from Investment Quota Approval Form of QFIIs issued by SAFE, July 14, 2010 and monthly report issued by CSRC, respectively.

Domestic and Foreign Institutional Investor Behavior in China

127

manufacturing companies in China, we subdivide the manufacturing industry with CSRC industrial code C into 10 sublevel industries. There are 22 sectors in total. Dummy variables are employed to capture the industrial effect. For instance, the dummy variable for the agriculture, forestry, animal husbandry and fishery (A) industry will be assigned a value of one if a company belongs to that specific industry and a zero otherwise.

Financial Characteristics The firm-level characteristics are divided into two groups: the financial indicator group and the corporate governance indicator group. In each group, we choose a range of indicators to measure the firm-level characteristics according to previous literature and the characteristics of the Chinese capital market. Table 3 reports the detailed financial indicators definitions. The first column states the classification of financial characteristics, whereas suitable indicators are reported in column 2. For example, total assets and capitalization are adopted to indicate company size. In the last column, the formula is stated for each indicator. Drawing on existing studies, we use total assets and capitalization to measure size, current ratio, and quick ratio to measure short-term liability paying capability, asset-liability ratio, and long-term liability ratio to measure long-term liability paying capability, earning per share and price to earning ratio to measure return on shareholder equity and return on assets and return on equity to measure company profitability. In addition to these five firm-level characteristics, our analysis also includes four more characteristics, namely financial risk, cash flow, growth, and operating capability. Previous studies used market model indicators to measure the risk of a company, like the beta coefficient for the market model (Dahlquist & Robertsson, 2001; Kang & Stulz, 1997). However, such an indicator relied on the assumption of market efficiency. Naughton (2007) concludes the characteristics of Chinese stock market are weak disclosure and regulation, insider control, and manipulation. Therefore, rather than relying on models that assume market efficiency, we employ accounting indicators from the financial statement of listed firms, e.g. the degree of financial leverage and the degree of operating leverage. Using a similar argument, the growth rate of fixed assets and that of total assets are adopted in this chapter to measure the growth ability of a company rather than the market-to-book used in Kang and Stulz (1997) and Covirg et al. (2006). In order to investigate a comprehensive range of firm-level characteristics, we also add cash flow and

128

DON BREDIN AND NINGYUE LIU

Table 3. Classification

Financial Indicator Definition.

Indicator

Formula

Size

Total assets Capitalization

ln (total assets) ln (capitalization)

Short-term liability Paying capability

Current ratio

Current assets/current liability

Quick ratio

(current assets-inventory)/current liability

Assets-liability ratio

Total liability/total assets

Long-term liability ratio

Long-term liability/total liability

Degree of financial leverage

Earnings before interest and taxes (EBIT)/(EBITfinancial cost) Percent change in operating income/% change in sales

Long-term liability Paying capability Financial risk

Degree of operating leverage Return on shareholders

Earning per share Price to earning ratio

Net profit/total shares outstanding Price per share/earning per share

Profit

Return on assets Return on equity

Netprofit/average total assets Net profit/average total equity

Cash flow

Free cash flow

Net profit þ amortization/depreciation-changes in working capital-capital expenditures Net increase in cash and cash equivalents/ Total shares outstanding

Net cash flow per share

Growth

Growth rate of fixed assets Growth rate of total assets

Operating capability

Turnover rate of receivables Turnover rate of inventory

(ending fixed assets-beginning fixed assets)/beginning fixed assets (ending total assets-beginning total assets)/beginning total assets Operating income/average receivables Operating cost/average inventory

Notes: In this table, we explain the financial indicators employed in the chapter. We divide the comprehensive range of financial characteristics into nine subgroups, namely size, short-term liability paying capability, long-term liability paying capability, financial risk, return on shareholders, profit, cash flow, growth, and operating capability. In each subgroup, two indicators are used to measure this classification for robust. In the last column, we also introduce the way to calculate the indicator.

Domestic and Foreign Institutional Investor Behavior in China

129

operating capability which are important characteristics for a company, but are omitted by previous studies.

Corporate Governance Characteristics Table 4 reports the definition of the corporate governance indicators. Since the late 1980s, shareholder activism has played a predominate role in improving corporate governance structures and several studies have

Table 4. Classification

Corporate Governance Definition.

Indicator

Formula

Ownership structure

Percentage of state-owned shares Percentage of circulating shares

Number of state-owned shares/number of total shares Number of circulating shares/number of total shares

Ownership concentration

Ownership percentage of the largest circulating shareholder Z index

Number of shares held by the largest circulating shareholder/number of total shares Number of shares held by the largest circulating shareholder/number of shares held by the second largest circulating shareholder

Management structure

Number of directors Number of supervisors Duality of chairman and CEO Percentage of independent directors

Number of directors Number of supervisors A dummy variable equals to 1 if chairman and CEO is one person and 2 if otherwise Number of independent directors/number of directors

Management compensation

ln of sum of top three compensation of directors ln of sum of top three compensation of senior executives

ln of sum to top three compensation of directors ln of sum of top three compensation of senior executives

Notes: In this table, we explain the corporate governance indicators employed in the chapter. We divide the comprehensive range of corporate governance characteristics into four subgroups, including ownership structure, ownership concentration, management structure, and management compensation. In each subgroup, some indicators are used to measure this classification for robust. In the last column, we also introduce the way to calculate the indicator.

130

DON BREDIN AND NINGYUE LIU

investigated this issue (Gillan & Starks, 2000; Karpoff, Malatesta, & Walkling, 1996). However, there is an endogeneity problem, suggesting that institutions are good at investing in the firms with better corporate governance structure, leading to the observed relationship between institutional presence and better-governed firms without any active participation (Chen, Harford, & Li, 2007). The authors focus on independent long-term institutions and find the corporate governance index (G-score) (Gompers, Ishii, & Metrick, 2003) has no effect on institutions’ shareholding decisions. Alternatively, Leuz, Lins, and Warnock (2009) study 4,409 firms from 29 countries and find foreigners invest less in firms due to governance problems. For ease of comparison, we first choose the same ownership concentration attributes as Dahlquist and Robertsson (2001) and management structure as Bushee, Carter, and Gerakos (2009). In addition to the two classifications, our analysis also includes two additional measures, namely ownership structure and management compensation. Naughton (2007) indicates that circulating and noncirculating shares is a notable characteristic of the Chinese equity market. Therefore, we employ the percentage of state-owned shares and percentage of circulating shares in the ownership structure subgroup. Finally, Hartzell and Starks (2003) find that institutions influence executive compensation through their preferences. However, there is also an endogeneity problem suggesting that institutions have a preference for certain executive compensation. Therefore, we investigate the impact of executive compensation on the investment decisions of funds, an issue which has been omitted by previous studies in the literature to date.

EMPIRICAL RESULTS Industrial Distribution Table 5 presents the difference between the industry’s value-weighted foreign and domestic funds ownership and the industry’s weight in the Chinese market portfolio for a total of 22 industries. All figures are in percentage terms. A 1% indicates that QFIIs invest 1% more of their Chinese A-share portfolio in that industry relative to the Chinese market portfolio. Column one lists the industry sectors. Column two and column three (Panel A) report the number of firms (in %) broken down by industry. Finally, in parentheses, the numbers of firm-year observations for each industry are reported. Columns four and five (Panel B) present the number of shares (in %) for each industry from 2003 to 2009.13 Columns two and

131

Domestic and Foreign Institutional Investor Behavior in China

Table 5. Industry Allocations of Foreign and Domestic Funds in China. Industry

Firm Number (Panel A)

Shareholding Number (Panel B)

Foreign

Domestic

Foreign

Domestic

Agriculture(A)

0.29 (21)

0.08 (142)

0.01

0.26

Mining(B)

1.80 (36)

0.18 (172)

4.26

0.85

Food & Beverage(C0)

1.46 (48)

0.39 (304)

2.38

1.23

Textiles & Apparel(C1)

1.22 (25)

0.44 (252)

0.61

0.74

Timber & Furnishings(C2)

0.01 (3)

0.07 (20)

0.01

0.04

Paper & Printing(C3)

0.30 (15)

0.20 (125)

0.30

0.13

Petrochemicals(C4)

3.04 (64)

0.16 (705)

0.95

0.57

Electronics(C5)

1.52 (25)

0.37 (277)

0.51

0.52

Metals & nonmetals(C6)

3.47 (105)

0.56 (628)

14.02

6.11

Machinery(C7)

0.02 (140)

0.93 (1040)

5.10

2.71

Pharmaceutical(C8)

0.35 (58)

0.27 (452)

0.10

1.46

0.15 (7)

0.17 (54)

0.04

0.05

1.86 (50)

0.71 (315)

2.24

0.41

0.37 (17)

0.04 (157)

1.39

0.12

Other Manufacturing(C9)

Utilities(D)

Construction(E)

132

DON BREDIN AND NINGYUE LIU

Table 5. (Continued ) Industry

Firm Number (Panel A)

Shareholding Number (Panel B)

Foreign

Domestic

4.82 (78)

0.99 (351)

4.17

0.38

Information Technology(G)

2.75 (37)

1.05 (410)

0.85

2.15

Wholesale & Retail(H)

0.19 (47)

0.68 (431)

0.35

2.36

Finance & Insurance(I)

0.39 (19)

0.14 (132)

19.44

17.81

2.76 (25)

0.19 (402)

0.23

2.94

1.37 (40)

0.20 (207)

1.03

0.55

Media & Culture(L)

0.84 (0)

0.03 (55)

0.23

0.40

Conglomerates(M)

2.68 (17)

0.37 (293)

0.96

1.33

Total

(877)

(6925)

Transportation(F)

Real Estate(J)

Social Services(K)

Foreign

Domestic

Notes: This table shows statistics of industry allocations by foreign and domestic funds. Column 1 lists the industries in China according to the classification of CSRC. Columns 2 and 3 (Panel A) report the deviation in percent of each industry’s weight in the portfolio held by foreign and domestic funds from its weight in the Chinese market portfolio in term of the number of firms. For example, the first row shows that 0.29% means that foreign funds invest 0.29% more of their portfolio in Agriculture industry than they would if their investment weights were those of the Chinese market portfolio. It indicates the fund’s over/under investment in an industry relative to the Chinese market portfolio. In parentheses, the numbers of firm-year observations for each industry are reported. Columns 4 and 5 (Panel B) report the deviation in term of the number of shares. The percentage of each industry’s weight in the Chinese market portfolio is the value by the end of 2009. The full name of industry A is agriculture, forestry, animal husbandry and fishery and F is communication, transportation & storage.

Domestic and Foreign Institutional Investor Behavior in China

133

four report QFIIs’ ownership and the domestic institutional investors’ ownership are presented in columns three and five. QFIIs hold disproportionately more of the transportation sector with an additional weight of 4.82% in relation to the number of firms and 4.17% in relation to the number of shareholders. The ownership of domestic institutional investors in the transportation sector is similar with the Chinese market portfolio (0.99% and 0.38% weight in their portfolio in two panels, respectively). We find consistent results to those reported by Kang and Stulz (1997), in relation to foreign investors having a preference for manufacturing industry (see machinery, metals and nonmetals results). In Panel A, QFIIs’ allocations in the finance and insurance sector (0.39%) are consistent with Chinese market portfolio, which means QFIIs invest in most of firms in this industry. However, QFIIs hold disproportionately less of the finance and insurance sector and the deviation is extremely large in Panel B.14 Between 2003 and 2009, QFIIs invested 19.44% less of their Chinese portfolio in the finance and insurance sector than in the market portfolio. This is likely to be due to some QFIIs being strategic investors in Chinese banks.15 Strategic investor is an alternative channel to invest in finance and insurance sectors in China. Furthermore, the ownership of each strategic investor can not be less than 10% whereas shares held by each QFII in any one listed company should not exceed 10% of the total outstanding shares of the company. Therefore, QFIIs are more likely to invest more in finance and insurance sector as strategic investors in order to play a dominate role as a big shareholder rather than invest in this sector as a QFII. Our results indicate that foreign investors have little preference for real estate, in particular relative to domestic investors. The lower levels of investment in real estate by QFIIs are consistent with the empirical evidence from developed markets that finds a considerably larger role played by domestic investors, given the requirement of local knowledge (Dahlquist & Robertsson, 2001). QFIIs also hold fewer allocations of construction, media and culture, other manufacturing and conglomerates. The portfolios of domestic funds are distributed more evenly across sectors compared to QFIIs. Domestic funds invest 0.56% and 6.11% more in metals and nonmetals in the two panels respectively which is consistent with the industrial preferences of QFIIs. However, they hold disproportionately more of wholesale and retail and real estate sectors which is inconsistent with QFIIs. The requirement for local knowledge is again a likely explanation for such a finding. Finally, domestic funds also hold disproportionately lower levels of the finance and insurance with 17.81% underweight their portfolios in Panel B.

134

DON BREDIN AND NINGYUE LIU

Foreign and Domestic Funds: An Empirical Comparison In this section, we examine evidence of different investment characteristics between foreign and domestic funds. In order to take account of outliers, we winsorize the data before computing the statistics.16 All data below the 1st percentile are set to the 1st percentile, and data above the 99th percentile are set to the 99th percentile. We use the t-test and the median test to investigate the mean and median differences between the two groups of nonfinancial firms. Foreign and domestic funds’ portfolio holdings consist of 858 and 6,793 firm-year observations, respectively. The sample size of domestic funds’ portfolio holdings is almost eight times of that of QFIIs. This is not surprising because of the short history and the small number of QFIIs in China compared to domestic funds. The comparative results of financial characteristics are reported in Table 6. The two groups are statistically different in terms of size and profit. Consistent with prior research, we find that the size of the firms in foreign funds’ portfolios are larger than those in domestic funds’ portfolios (see

Table 6. Classification

Size

Financial Indicator Comparison.

Indicator

ln of total assets ln of capitalization

Test

QFII

Domestic Fund

p-Value

Mean Median Mean Median

22.05 21.87 22.52 22.44

21.65 21.51 22.04 21.90

0.000 0.000 0.000 0.000

Mean Median Mean Median

1.69 1.20 1.27 0.78

1.82 1.27 1.36 0.84

0.161 0.014 0.294 0.009

Short-term liability paying capability

Current ratio

Long-term liability paying capability

Assets-liability ratio Long-term liability ratio

Mean Median Mean Median

0.47 0.48 0.19 0.12

0.48 0.49 0.16 0.09

0.305 0.060 0.000 0.002

Financial risk

Degree of Financial leverage Degree of operating leverage

Mean Median Mean Median

1.20 1.13 1.76 1.64

1.31 1.14 1.98 1.78

0.033 0.129 0.008 0.001

Quick ratio

135

Domestic and Foreign Institutional Investor Behavior in China

Table 6. (Continued ) Classification

Test

QFII

Domestic Fund

p-Value

Mean Median Mean Median

0.49 0.39 65.76 29.53

0.33 0.26 68.24 32.27

0.000 0.000 0.689 0.002

Mean Median Mean Median

0.07 0.06 0.14 0.12

0.05 0.04 0.09 0.09

0.000 0.000 0.000 0.000

Net cash flow per share

Mean Median Mean Median

8.94 2.09 0.34 0.09

4.86 1.09 0.29 0.07

0.000 0.000 0.245 0.081

Growth

Growth rate of fixed assets Growth rate of total assets

Mean Median Mean Median

0.29 0.08 0.24 0.15

0.31 0.06 0.23 0.12

0.756 0.020 0.557 0.002

Operating capability

Turnover rate of receivables Turnover rate of inventory

Mean Median Mean Median

125.79 9.20 31.89 4.85

86.79 7.74 17.75 4.22

0.303 0.000 0.089 0.001

Return on shareholders

Indicator

Earning per share Price to earning ratio

Profit

Return on assets Return on equity

Cash flow

Free cash flow

Notes: In this table, we compare a range of financial indicators between companies in two groups. The first column is the classification of financial characteristics. In the second column, we choose some indicators to measure this characteristic. T-test and median test are employed to compare the differences between these two groups. The mean and median of each indicator of these two groups are in the fourth and fifth column respectively. P-value of test is in the last column. , , and  represent significance at 1%, 5%, and 10% levels, respectively. The unit of total assets, capitalization, earning per share, and net cash flow per share is CNY while the unit of free cash flow is 100 million CNY.

Aggarwal et al., 2005; Dahlquist & Robertsson, 2001). The mean of both the current and quick ratio in the two groups are statistically nonsignificant different, which is consistent with the findings of Kang and Stulz (1997).17,18 The earning per share in foreign funds’ group is greater than that in domestic funds’ group which highlights the success of foreign investors. A higher price to earning ratio means that investors are paying more for each

136

DON BREDIN AND NINGYUE LIU

unit of net income, so the stock is more expensive compared to one with lower price to earning ratio. Moreover, the price to earning ratio in China is several times greater than those in many developed countries.19 Therefore, it is not a surprise that QFIIs have a preference for firms with lower price to earning ratio according to the median test. Another notable difference is both ROA and ROE in foreign funds’ group are larger than those in domestic funds’ group which is consistent with the results in Kang and Stulz (1997) using ROA, but inconsistent with the findings in (Dahlquist & Robertsson, 2001) and Aggarwal et al. (2005) using ROE, where no significant relationship is found. The ratio of assets-liability in foreign funds’ group is not significantly larger than that in domestic funds’ group at the 5% significant level. Even the long-term liability ratio in foreign funds’ group is significantly greater than that in domestic funds’ group at 1% level. Given the questions over market efficiency, we employ accounting indicators to model company risk, rather than the standard market indicators. The results indicate that there is significant difference between the two groups. The mean and median values of the degree of operating leverage in domestic funds’ group are larger than those in foreign funds’ group which indicates that foreign funds have a preference for the firms with lower risk level. Our finding is inconsistent with the findings in previous studies (Dahlquist & Robertsson, 2001; Kang & Stulz, 1997) adopting the market proxy (b). Moreover, the t-test results using accounting indicators to measure growth indicate there is no significant difference between these two groups which is inconsistent with those reported in prior research (Aggarwal et al., 2005; Dahlquist & Robertsson, 2001) employing the market proxies like book-to-market ratio or price-to-book ratio. One possible reason for the inconsistency is the potential lack of market efficiency. Our findings have an important implication for further research which would like to use indicators to measure the financial characteristics of companies in emerging markets. Table 6 also reports our findings on two new firm-level characteristics. There is significant difference between the two groups in relation to free cash flow, which partially indicates that the firms foreign funds invest in hold more free cash flow compared those domestic funds invest in. However, there is no significant robust difference between the two groups in terms of operating capability. Finally, the comparative results of corporate governance characteristics are reported in Table 7. Unlike the results reported in Dahlquist and Robertsson (2001), the larger mean and median of ownership percentage of the largest circulating shareholder in foreign funds’ group, indicate that the

137

Domestic and Foreign Institutional Investor Behavior in China

Table 7. Classification

Corporate Governance Comparison. Test

QFII

Domestic Fund

p-Value

Mean Median Mean Median

0.29 0.30 0.54 0.50

0.28 0.28 0.51 0.46

0.645 0.308 0.000 0.000

Ownership Ownership percentage of the concentration largest circulating shareholder Z index

Mean Median Mean Median

8.72 3.11 9.25 1.60

6.23 1.95 6.18 1.57

0.000 0.000 0.014 0.554

Management structure

Mean Median Mean Median Mean Median Mean

9.74 9.00 4.16 3.00 1.85 2.00 0.35

9.48 9.00 4.11 3.00 1.85 2.00 0.35

0.002 0.011 0.469 0.493 0.903 0.900 0.565

Mean Median Mean Median

13.49 13.52 13.71 13.73

13.26 13.30 13.43 13.46

0.000 0.000 0.000 0.000

Ownership structure

Indicator

Percentage of state-owned shares Percentage of circulating shares

Number of directors Number of supervisors Duality of chairman and CEO Percentage of independent directors

Management ln of sum to top three compensation compensation of directors ln of sum of top three compensation of senior executives

Notes: In this table, we compare some corporate governance indicators between the companies in two groups. The first column is the classification of corporate governance characteristics. In the second column, we choose some indicators to measure this characteristic. T-test and median test are employed to compare the differences between these two groups. The mean and median of each indicator of these two groups are in the fourth and fifth column respectively. P-value of test is in the last column. , , and  represent significance at 1%, 5%, and 10% levels, respectively. The unit of compensation is CNY.

firms foreign funds invest in are more likely to have a dominant owner. Furthermore, our results also suggests that foreign funds have a preference for firms with a higher percentage of circulating shares. In contrast, however, there are no significant differences between the two groups with regard to most indicators in the management structure aspect, except for the number of directors and management compensation. Both indicators of management compensation are significantly different between the two

138

DON BREDIN AND NINGYUE LIU

groups at the 1% level, which means that foreign funds have a preference for the firms with greater management’s compensation, which is inconsistent with the findings in developed markets.20 Our findings have highlighted a number of issues between foreign and domestic funds in particular in relation to financial and corporate governance characteristics. Consistent with prior evidence from developed markets, e.g. Kang and Stulz (1997) and Dahlquist and Robertsson (2001) and emerging countries, e.g. Aggarwal et al. (2005), we find that the size of firms foreign institutional investors invest in is larger than those that domestic funds invest in. Although the importance of profit is consistent with the results reported in Kang and Stulz (1997) using ROA, it is inconsistent with the findings in Dahlquist and Robertsson (2001) and Aggarwal et al. (2005) using ROE. Recent studies have highlighted the greater explanatory power of accounting measures, relative to market measures, in relation to the cost of equity in markets where information is limited. Furthermore, Toms, Salama, and Nguyen (2005) confirm the importance of the degree of operating leverage in the determination of systematic risk. Given the potential for limited information, we employ accounting indicators, the degree of financial leverage and the degree of operating leverage, to measure risk. Our results stand in stark contrast to previous studies, e.g. Kang and Stulz (1997) and Dahlquist and Robertsson (2001) using the beta of the market model, albeit for developed markets. There is significant difference between the companies in the two groups in relation to the aspect of corporate governance, e.g. management compensation, which indicates that corporate governance is also an important consideration when foreign funds make investment decisions which is consistent with the findings of McKinsey and Company (2002). Policy makers could use this characteristic of QFIIs to improve the governance mechanisms of Chinese listed firms. Finally, our results indicate that QFIIs do not follow the ‘‘free rider strategy’’ even though they are not familiar with the market.

CONCLUSIONS This chapter employs a unique data set to analyze the firm-level characteristics of firms QFIIs invest in, including financial and corporate governance indicators, and identifies the similarities and differences between foreign and domestic funds. In the first part of the study we analyze the industrial distribution of the companies foreign and domestic funds invest

Domestic and Foreign Institutional Investor Behavior in China

139

in. Furthermore, we use the t-test and median test to compare the mean and median of financial and corporate governance indicators of the companies held by foreign and domestic funds, respectively. Results of our analysis indicate that QFIIs have a preference for industries like transportation, metals and nonmetals, and machinery. However, there is very little foreign institutional investment in real estate, construction, and media and culture, where there is a requirement for local knowledge. This result is consistent with evidence from developed countries (Dahlquist & Robertsson, 2001; Kang & Stulz, 1997). The portfolios of domestic Chinese funds are distributed more evenly compared to the QFIIs’ portfolio. The characteristics of firms that foreign and domestic funds invest in are significantly different in terms of size and profit which are consistent with the findings of developed markets (Kang & Stulz, 1997). Our results highlight the implications of adopting market indicators, rather than accounting indicators, for volatile, illiquid, and speculative emerging markets. Moreover, consistent with the finding in McKinsey and Company (2002), corporate governance is an important issue when it comes to investment decisions for QFIIs. The dramatic growth of the QFII scheme may by itself force improvement in the area of corporate governance in China. Finally, the comparative analysis of firm-level characteristics between firms held by foreign and domestic funds partly indicates that QFIIs do not follow the investment decisions of domestic funds, even taking account of the limited information associated with the Chinese market. Future research could investigate several issues that our study has raised. Since QFIIs tend to rely on some aspects of corporate governance structure when they make the investment decisions, it might be worthwhile to examine the impact of QFIIs on the corporate governance of China’s listed companies during their holding periods. For example, Ferguson and McGuinness (2004) have highlighted that the QFIIs scheme could improve the corporate governance structure of Chinese listed companies. An extension to our approach could provide an empirical context to this prediction and a further validation of the QFII scheme. Finally, our study could be further developed to the multivariate setting and so provide further insight.

NOTES 1. Source: Board of Governors of the Federal Reserve System, 2003. 2. From Investment Quota Approval Form of QFIIs issued by State Administration of Foreign Exchange (SAFE), November 10, 2010.

140

DON BREDIN AND NINGYUE LIU

3. Naughton (2007) indicates that the Chinese stock market is heavily influenced by changes in government policy, along with limitations on individual firm information and shareholder control. 4. From National Bureau of Statistics of China. 5. International Monetary Fund, World Economic Outlook Database, October 2010. 6. China’s two domestic stock exchanges, Shanghai Stock Exchange and Shenzhen Stock Exchange, were established in December 1990 and July 1991, respectively. 7. From the data released by CSRC on January 26, 2011. 8. Dahlquist and Robertsson (2001) find there is a negative relationship between foreign institutional ownership and ownership concentration, implying that foreigners avoid companies with a dominant owner. 9. Source from ‘‘Provisional Measures on Administration of Domestic Securities Investments of Qualified Foreign Institutional Investors,’’ Chapter 1, Page 1. 10. A Shares in Shanghai and Shenzhen stock exchanges refer to those that are traded in CNY. Currently only mainlanders and QFIIs are allowed to trade A-shares. B-shares are traded in foreign currencies in the two mainland Chinese stock exchanges. In the past, only non-Chinese were allowed to trade B-shares. Mainlanders can trade B-shares as well from March 2001. QFIIs are not permitted to invest in B-shares. 11. Wind Info is a financial data provider in Mainland China. It provides largesized financial database to academic researchers as well as financial organizations, including Merrill Lynch. Data from the Wind database has been examined by Poon and Chan (2008) and Wu, Xu, and Yuan (2009). 12. Shanghai Stock Exchange Composite Index fell from the peak of 6,092 points on October 16, 2007 to the bottom of 1,719 on November 3, 2008. 13. In some cases, investors invest in large number of firms in one industry but hold small number of shares in each firm. Shareholding numbers represent the preference of investors in each industry. Therefore, we also report the number of shares (in %) for each industry. 14. The companies in finance and insurance sector will not be examined in later sections. This is due to data limitations. A number of indicators used in this chapter do not exist in this industry, e.g. current ratio and quick ratio, as there are no current assets and current liability terms in banks’ balance sheets. Kang and Stulz (1997) also employ the data from nonfinancial Japanese firms. 15. Strategic investor is another form of foreign institutional investors with access to China’s market. They invest A-shares through negotiations and initial public offerings rather than trading under the QFII scheme. Most strategic investors acquire interests in Chinese banks. For example, Temasek, through its wholly owned subsidiary Asia Financial Holdings (AFH), acquired a 10% interest in Bank of China in August 2005. Meanwhile, Temasek is also a member of QFIIs. 16. Winsorising or Winsorization is the transformation of statistics by transforming extreme values in the statistical data, and is named after the engineer-turnedbiostatistician Charles P. Winsor (1895–1951). Winsorised estimators are usually more robust to outliers than their unwinsorised counterparts. 17. We use the current and quick ratio as a proxy for short-term liability paying capability. They are calculated as current assets divided by current liability and current assets minus inventory divided by current liability respectively.

Domestic and Foreign Institutional Investor Behavior in China

141

18. In Kang and Stulz (1997), the coefficient on current ratio in the regression of foreign ownership on several explanatory variables is significantly positive in 1 year out of 16 years. Therefore, we believe it is almost nonsignificant. 19. Source from P/E Ratio Global Stock Markets Analysis and Technical Outlook, November 07 issued on 18th November 2007 by The Market Oracle http:// www.marketoracle.co.uk/. A high price to earning ratio signifies high expectations on the growth. Many developed countries have low price to earning ratios, but they also have low GDP growth, while developing countries may have higher market valuations as well as stronger GDP growth, in particular the high price to earning ratio and strong GDP growth in China. 20. David, Kochhar, and Levitas (1998) using US data find that institutional owners that have only an investment relationship with a firm reduce the level of CEOs’ pay.

REFERENCES Aggarwal, R., Klapper, L., & Wysocki, P. (2005). Portfolio preferences of foreign institutional investors. Journal of Banking and Finance, 29(12), 2919–2946. Bekaert, G., Harvey, C. R., & Lumsdaine, R. (2002). Dating the integration of world capital markets. Journal of Finance, 65(2), 203–249. Bushee, B., Carter, M. E., & Gerakos, J. (2009). Institutional investor preferences for corporate governance mechanisms. SSRN Working Paper Series. Chan, K., Covrig, V., & Ng, L. (2005). What determines the domestic bias and foreign bias? Evidence from mutual fund equity allocations worldwide. Journal of Finance, 60(3), 1495–1534. Chang, C. (2010). Information footholds: Isolating local presence as a factor in analyst performance and trading. Journal of International Money and Finance, 29(6), 1094–2218. Chen, M. & Yu, W. (2003). Stock market liberalization and market returns in China: Evidence from QFII announcement. SSRN Working Paper. Chen, X., Harford, J., & Li, K. (2007). Monitoring: Which institutions matter? Journal of Financial Economics, 86(2), 279–305. Cheng, L. K., & Kwan, Y. K. (2000). What are the determinants of the location of foreign direct investment? The Chinese experience. Journal of International Economics, 51(2), 379–400. Choe, H., Kho, B., & Stulz, R. M. (2005). Do domestic investors have an edge? The trading experience of foreign investors in Korea. Review of Financial Studies, 18(3), 795–829. Cooper, I. A., & Kaplanis, E. (1994). What explains the home bias in portfolio investment. Review of Financial Studies, 7(1), 45–60. Covirg, V., Lau, S. T., & Ng, L. K. (2006). Do domestic and foreign fund managers have similar preferences for stock characteristics? A cross country analysis. Journal of International Business Studies, 37(3), 407–429. Dahlquist, M., & Robertsson, G. (2001). Direct foreign ownership, institutional investors, and firm characteristics. Journal of Financial Economics, 59(3), 413–440. David, P., Kochhar, R., & Levitas, E. (1998). The effect of institutional investors on the level and mix of CEO compensation. Academy of Management Journal, 41(2), 200–208. Dean, J. (2003). Taiwan’s stock market could get a lift. Wall Street Journal (Eastern Edition), C14.

142

DON BREDIN AND NINGYUE LIU

Falkenstein, E. G. (1996). Preferences for stock characteristics as revealed by mutual fund portfolio holdings. Journal of Finance, 51(1), 111–135. Ferguson, M. J., & McGuinness, P. B. (2004). Chinese securities reform: The role of QFIIs in the corporate governance process. Business Horizons, 47(2), 53–61. Frenkel, K. R., & Poterba, J. M. (1991). Investor diversification and international equity markets. American Economic Review, 81(2), 222–226. Frenkel, M., & Menkhoff, L. (2003). Are foreign institutional investors good for emerging markets?. The World Economy, 27(8), 1275–1293. Giannetti, M., & Koskinen, Y. (2010). Investor protection, equity returns, and financial globalization. Journal of Financial and Quantitative Analysis, 45(1), 135–168. Gillan, S. L., & Starks, L. T. (2000). Relationship investing and shareholder activism by institutional investors. Journal of Financial Economics, 57(2), 275–305. Gompers, P., Ishii, J., & Metrick, A. (2003). Corporate governance and equity prices. Quarterly Journal of Economics, 118(1), 107–155. Grinblatt, M., & Keloharju, M. (2000). The investment behavior and performance of various investor types: A study of Finland’s unique data set. Journal of Financial Economics, 55(1), 43–67. Hartzell, J. C., & Starks, L. T. (2003). Institutional investors and executive compensation. Journal of Finance, 58(6), 2351–2374. Huang, R. D., & Shiu, C. (2006). Local effects of foreign ownership in an emerging financial market: Evidence from qualified foreign institutional investors in Taiwan. Financial Management, 38(3), 567–602. Kang, J., & Stulz, R. (1997). Why is there a home bias? An analysis of foreign portfolio equity ownership in Japan. Journal of Financial Economics, 46(1), 3–28. Karpoff, J. M., Malatesta, P. H., & Walkling, R. A. (1996). Corporate goverance and shareholder initiatives: Empirical evidence. Journal of Financial Economics, 42(3), 365–395. Khorana, A., Servaes, H., & Tufano, P. (2005). Explaining the size of the mutual fund industry around the world. Journal of Financial Economics, 78(1), 145–185. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1997). Legal determinants of external finance. Journal of Finance, 52(3), 1131–1150. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1998). Law and finance. Journal of Political Economy, 106(6), 1113–1155. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (2000). Investor protection and corporate governance. Journal of Financial Economics, 58(5), 3–28. Lai, C., Lou, K., & Shiu, C. (2008). Does foreign investors trade predict stock return? Evidence from Taiwan. Information and Management Sciences, 19(3), 477–494. Leuz, C., Lins, K. V., & Warnock, F. E. (2009). Do foreigners invest less in poorly governed firms?. Review of Financial Studies, 22(8), 3245–3285. Lewis, K. K. (1999). Trying to explain home bias in equities and consumption. Journal of Economic Literature, 37(2), 571–608. Li, J., & Jeong-Bon, K. V. (2004). Foreign equity ownership and information asymmetry: Evidence from Japan. Journal of International Financial Management and Accounting, 15(3), 185–211. Lin, A., & Chen, C. (2006). The impact of qualified foreign institutional investors on Taiwan’s stock market. Journal of Chinese Management Review, 9(2), 1–27.

Domestic and Foreign Institutional Investor Behavior in China

143

Lin, A. Y., & Swanson, P. E. (2003). The behavior and performance of foreign investors in emerging equity markets: Evidence from Taiwan. International Review of Finance, 4(3–4), 189–210. Lin, C. H., & Shiu, C. (2003). Foreign ownership in the Taiwan stock market: An empirical analysis. Journal of Multinational Financial Management, 13(1), 19–41. Lu, Y., Fang, H., Wong, J., & Wang, Y. (2009). Information contents of QFIIs’ cascades in the Taiwan stock market. SSRN Working Paper. McKinsey and Company. (2002). Global Investor Opinion Survey, July. Naughton, B. (2007). The Chinese economy transitions and growth. Cambridge, MA: MIT Press. Poon, W. P. H., & Chan, K. C. (2008). An empirical examination of the informational content of credit ratings in China. Journal of Business Research, 61(7), 790–797. Provisional Measures on Administration of Domestic Securities Investments of Qualified Foreign Institutional Investors, China Securities Regulatory Commissionon, 5th November 2002. Tesar, L., & Werner, I. M. (1995). Home bias and high turnover. Journal of International Money and Finance, 14(4), 467–493. Ting, W., Yen, S., & Chiu, C. (2008). Risk and audit opinions: Evidence form the Chinese stock market. Corporate Governance: An International Review, 16(5), 400–415. Toms, S., Salama, A., & Nguyen, D. T. (2005). The association between accounting and marketbased risk measures. University of York Working Paper. Wu, S., Xu, N., & Yuan, Q. (2009). State control, legal investor protection, and ownership concentration: Evidence from China. Corporate Governance: An International Review, 17(2), 176–192. Yeo, S. (2003). The PRC qualified foreign institutional investors market. China Economic Review, 14(4), 443–450. Zhang, K. H. (2001). China’s inward FDI boom and the greater Chinese economy. Chinese Economy, 34(1), 74–88.

INSTITUTIONAL INVESTORS’ PARTICIPATION IN FOREIGN FIRMS: EVIDENCE FROM ADRs Narjess Boubakri, Olfa Hamza and Maher Kooli STRUCTURED ABSTRACT Purpose – Study the firm-level and country-level determinants of US institutional investors’ holdings in American Depositary Receipts (ADRs) from emerging markets. Methodology/approach – We use a sample of 112 firms from emerging markets that listed as ADRs between 1990 and 2005. Rather than adopting the issuer’s perspective, we take in this study the point of view of the investor and we focus on the US institutional investors’ participation in ADR firms. Findings – We find that institutional investors hold higher stakes in foreign firms that are listed on more restrictive exchanges, in large, privatized, more liquid, and more transparent firms. Mutual investors and other institutional investors also prefer firms from countries with weaker institutional environments and from civil law legal tradition. Controlling for country-level determinants increases significantly the explanatory power of the model. Social implications – Our results have important implications for firms from emerging markets seeking to attract foreign institutional investors. Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 145–168 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012008

145

146

NARJESS BOUBAKRI ET AL.

Originality/value of the chapter – We focus on the motivations of investors when they choose to invest in the ADR, rather than on the ADR issuer motivation. In addition, we consider all types of institutional investors that acquire a participation in an ADR firm. Keywords: Emerging markets; allocation; cross-listed stocks; ADRs JEL classifications: G11; G15; G18; G23

INTRODUCTION American Depositary Receipts (ADRs) and cross-listing have drawn a lot of attention lately from academicians and researchers. Cross-listing on US markets and issuing ADRs offer foreign firms a variety of advantages. Cross-listing allows investors to circumvent investment barriers and market segmentation (direct costs and information problems, among others as suggested by Merton, 1987). Other benefits include the possibility to raise funds to finance investment projects, especially for firms with growth opportunities. Foreign firms are also attracted by deep and liquid markets (Pagano, Roell, & Zechner, 2002). Indeed, ADR issuers typically enjoy an increase in liquidity (i.e., higher trading volumes or lower bid-ask spreads) as shown by Bekaert, Harvey, and Lundblad (2007) and Errunza and Miller (2003). In this chapter, we examine the firm-level and the country-level determinants of US institutional investors’ holdings in ADRs from emerging markets. The role of institutional investors as important financial intermediaries is widely documented in the literature, and their importance as major players in the recent market history has called our attention: for instance, Aggarwal, Dahiya, and Klapper (2007) report that according to the US Federal Reserve’s Flow of Funds, US investors allocated 11–12% of their total equity portfolio to non-US equities in 2003 (equivalent to $ billion 1,300). Most importantly for our purposes, institutional investors are responsible for the large interest in the trading of ADRs (Aggarwal et al., 2007). Informational asymmetries including poor visibility of the firm on foreign markets as well as the lower quality of accounting information may explain the ‘‘domestic bias’’ defined as the preference of investors for domestic firms compared to foreign firms (Ahearne, Griever, & Warnock, 2004). Edison

Institutional Investors’ Participation in Foreign Firms

147

and Warnock (2004) show there is no ‘‘domestic bias’’ for firms from emerging markets that have issued ADRs. Foreign large firms that are crosslisted on US markets and that have higher analyst following are shown to attract more foreign capital (Ahearne et al., 2004; Covrig, Lau, & Ng, 2006; Edison & Warnock, 2004; Kang & Stulz, 1997). In addition, once cross-listed, the firm must comply with the Security and Exchange Commission (SEC) disclosure requirements, thus increasing the confidence of US investors in the financial information provided by these companies. In fact, factors related to the visibility of the firm (i.e., firm size), analyst following, and being an ADR are shown to be significant determinants of the investment choices of US institutional investors (Edison & Warnock, 2004; Kang & Stulz, 1997). Accounting standards are also likely to affect the perception and credibility of the quality of accounting information by outsiders, helping to decrease the cost of acquiring information in a context of informational asymmetries (Barth, Clinch, & Shibano, 1999; Sunder, 2002). Complying with US generally accepted accounting principles (US GAAP) provides a positive signal of the credibility of financial information (Bradshaw & Miller, 2006; Dye & Sunder, 2001; Krishnan, 2003). Bradshaw, Bushee, and Miller (2004) in particular predict that institutional investors are more likely to invest in firms that comply with US GAAP. Gillian and Starks (2003) test this hypothesis and show that institutional investors prefer foreign firms with better quality of financial information when making their investment decisions. In conclusion, issuing an ADR allows the firm to alleviate informational asymmetries, thus providing investors with better accounting information and higher disclosure, thus affecting both the firm valuation and the investor’s perception of the firm’s disclosure environment. Rather than adopting the issuer’s perspective, we take in this chapter the point of view of the investor and we focus on the US institutional investors’ participation in ADR firms. Unlike Aggarwal, Klapper, and Wysocki (2005) who consider only mutual funds, we include all types of institutional investors that participate in ADRs. Investing in ADRs offers several benefits including a better investor rights protection: Indeed, according to the ‘‘bonding’ hypothesis, if a firm originating from a country with weak investor protection lists in the United States, US stringent securities laws will provide additional protection to investors (Coffee, 1999). Other advantages offered to investors through an ADR program include higher liquidity, higher transparency (i.e., resulting from the compliance to US GAAP), and ease in trading (Aggarwal, Erel, Ferreira, & Matos, 2011). Examining the determinants of institutional investors’ participation in foreign firms is important on different critical dimensions: first, institutional

148

NARJESS BOUBAKRI ET AL.

investors are often seen as important channels ‘‘for promoting better governance and convergence in governance across countries’’ (Aggarwal et al., 2011; Gillian & Starks, 2003). The literature shows that foreign investors typically play an active role in promoting investments and improve corporate governance practices and performance. Second, establishing the characteristics of those firms favored by institutional investors is likely to help managers of domestic firms that seek to attract foreign capital to restructure their firms. Third, examining the country-level determinants of foreign institutional investors’ allocations is related to the debate on the effect of liberalization on domestic economic growth. As the country attracts more foreign investors, it becomes more liberalized, thus spurring economic growth (Bekaert, Harvey, & Lundblad, 2005). Finally, countries and firms are interested in attracting foreign capital because ‘‘it helps to create liquidity for both the firms’ stock and the stock market in general’’ (Ferreira & Matos, 2008).1 This in turn will directly benefit the economy and the country. Our chapter differs from previous studies on ADRs on several grounds: first, as previously stated, we focus on the motivations of investors when they choose to invest in the ADR, rather than on the ADR issuer motivation. We also use a long study period covering the years from 1990 to 2005. In addition, rather than focusing on institutional holdings from just one class of institutions (mutual funds as in Aggarwal et al., 2005, 2007; Chen, Hong, & Stein, 2002; Covrig et al., 2006), we consider all types of institutional investors that acquire a participation in an ADR firm. Nevertheless, we analyze in the empirical section whether mutual funds differ from other investors in their allocation choice. We finally measure portfolio investment allocation of institutional investors on a raw basis and on an adjusted basis (i.e., in comparison with the industry average institutional investors’ participation). This allows us to identify which firms these investors over or underweigh in terms of participation. Our study adds to the growing literature on the determinants of foreign investment allocations on the following grounds. First, we complement Aggarwal et al. (2005, 2007) who study the choice of institutional investors (mutual funds) between investing in ADRs and investing in major stock market indices, and between investing in the ADR (traded in the United States) and investing in the underlying stock that trades in the domestic market, respectively. By examining all types of institutional investors’ allocations in ADRs, we are able to provide a more exhaustive analysis of their investment choices. Also instead of running separate regressions for firm-level and country-level variables as in Aggarwal et al. (2005), we

Institutional Investors’ Participation in Foreign Firms

149

consider a more exhaustive approach that incorporates both. Furthermore, we contribute to the literature on corporate ownership determinants by focusing on the factors underlying US institutional investors’ participation in foreign firms. By considering ADRs, we are able to incorporate potential explanatory variables at the country level. Using a sample of 112 firms from emerging markets that listed as ADRs between 1990 and 2005, we find that institutional investors hold higher stakes in foreign firms that are listed on more restrictive exchanges, in large, privatized, more liquid, and more transparent firms. Mutual investors and other institutional investors also prefer firms from countries with weaker institutional environments and from civil law legal tradition. Controlling for country-level determinants increases significantly the explanatory power of the model. The remainder of the chapter is structured as follows: in the second, we review the literature on the determinants of institutional investors’ ownership and draw our hypotheses. In the third section, we describe our sample and variables and discuss descriptive statistics. In the fourth section, we present our univariate and multivariate analyses before we conclude in the fifth section.

THE DETERMINANTS OF INSTITUTIONAL INVESTORS’ OWNERSHIP: LITERATURE REVIEW AND HYPOTHESES Firm-Level Determinants Institutional investors are perceived as being sophisticated and enjoy an informational advantage compared to other types of investors. Ke and Petroni (2004) for instance provide evidence on the sources and informational advantages of institutional investors by investigating their capacity to predict a decline in the quarterly earnings of firms after successive positive earnings. The authors show that institutional investors are indeed able to predict this reversal of profits one quarter in advance. They also show that one of the primary sources of institutional investors’ information advantage is their private communication with managers. The authors find no evidence that there is a predictive capacity of reversals for other investors. Similar results are reported in Walther (1997) and El Gazzar (1998). In the same vein, Utama and Cready (1997) show that the volume of transaction around

150

NARJESS BOUBAKRI ET AL.

earnings announcements is higher in firms where institutional investors own around 50% of outstanding shares. This result suggests that institutional investors enjoy more precise private information than other investors. As ADR firms originate from foreign countries, it is very unlikely that US investors benefit from this informational advantage. However, previous studies have shown that firms with ADRs appear to have better information environments that lead to higher valuation and significantly higher market reactions to earnings announcements (Doidge, Karolyi, & Stultz, 2002). Several studies have also focused on the characteristics of firms targeted by foreign institutional investors. Kang and Stulz (1997) provide evidence on institutional investors’ investments in Japanese firms and show that they tend to avoid smaller companies with heavy debt. The participation stake is shown to be higher in more internationalized companies (i.e., higher level of exports). Following Kang and Stulz (1997), Dahlquist and Robertsson (2001) try to identify the firm characteristics of Swedish firms that attract foreign investors. They specifically focus on firm size, dividend, stock return, systematic risk, book to market, debt ratio, firm liquidity, stock turnover, and export ratio as well as ownership concentration and whether the firm is cross-listed on foreign markets. Results show that foreign investors prefer large firms that pay lower dividends and those with high liquidity. The authors show that the level of exports (internationalization), cross-listing, and liquidity add to the explanatory power of the model of foreign participation. In addition, institutional investors prefer firms with more diffuse ownership (i.e., less agency problems). In the first to our knowledge multinational study related to this issue, Aggarwal et al. (2005) consider firms operating in emerging markets and examine the investment allocation choices of US mutual funds in these firms, relative to major stock market indices. This allocation decision is conditioned by firm- and country characteristics. The authors follow Falkenstein (1996) for US firms, Kang and Stulz (1997) for Japanese firms, and Dahlquist and Robertsson (2001) for Swedish firms and examine the following characteristics: firm size, dividend yield, leverage, firm growth, profitability, and annual stock return. They confirm previous findings that institutional investors prefer larger firms, with low leverage, higher profitability, and better growth opportunities. Aggarwal et al. (2005) also use the number of analysts following the firm on the ground that these play an important role in alleviating informational asymmetries by gathering information and monitoring management’s actions. The authors further distinguish between listed and unlisted ADRs on the grounds that listed ADRs (exchange listings) have higher disclosure requirements and must

Institutional Investors’ Participation in Foreign Firms

151

reconcile their financial statements according to US GAAP. Finally, the authors use another proxy for disclosure quality, namely, auditor quality. Consistent with previous studies, institutional investors prefer firms with more visibility (larger size) and higher analyst following. Also, proxies for better disclosure (i.e., listed ADRs) are significant determinants of institutional investors’ allocation. Country-Level Determinants The literature on institutional investors’ preferences or participation in foreign firms was very scarce before the study of Aggarwal et al. (2005) in the context of ADRs from emerging markets. US institutional investors, as argued by the authors, are the major force behind the interest in ADRs. According to the ‘‘bonding’ hypothesis, ADRs provide domestic investors with higher investor protection: if a firm originating from a country with weak investor protection lists in the United States, US stringent securities laws will provide investors with additional protection (Coffee, 1999).2 In this vein, Aggarwal et al. (2005) examining whether investor protection is a determinant in the allocation choice of institutional investors provide evidence that accounting standards (capturing the quality of accounting information) and shareholder rights (investor protection) play a significant role of the investment decision of US mutual funds in emerging markets. A related study by Ferreira and Matos (2008) examines the investment preferences of three pools of professional investors (US, non-US, and domestic institutions). The authors confirm previous evidence about firmlevel characteristics that attract institutional investors, as these seem to share a preference for large, liquid, and widely held stocks as well as those that cross-listed on US exchange by way of ADRs. At the country level, Ferreira and Matos (2008) reveal a strong preference by all institutional investors for stocks of countries with strong disclosure standards and that are geographically close to their home market. On the basis of the previous review of the literature on the determinants of foreign institutional investors’ participation in firms, we can conclude that institutional investors’ allocations are determined not only by firm-level characteristics but also by the characteristics of the country of origin of the firm. In summary, using our sample of foreign firms listed as ADRs, we will test the following hypotheses: H1. Institutional investors’ allocation in ADRs will be higher in large firms, privatized firms, those with lower leverage, better performance, better

152

NARJESS BOUBAKRI ET AL.

growth opportunities, better accounting standards, higher liquidity, lower dividends, and lower systematic risk. H2. Institutional investors are more likely to prefer countries with higher growth rates, better investor protection, and legal institutions.

DATA AND VARIABLES Our sample consists of 112 ADRs for the period 1990–2004. Country- and firm-level data are taken from various sources including Bankscope, Emerging Market Database, Money Central web site, Datastream, Disclosure, Mergent online, and ICRG (International Country Risk Guide). In the following, we briefly describe our dependent and independent variables. Dependent Variables 1. TOP50 is the percentage of ADR allocated to the top 50 institutional investors. 2. TOP50-ADJUSTED is the industry-adjusted percentage of ADR allocated to the top 50 institutional investors. 3. MFALL is the percentage of ADR allocated to mutual funds. Firm-Level Variables As year variable:  ADR-YEARS denotes the number of years since ADR. As market listing variable:  NYSE: a dummy variable that takes the value 1 if it is a NYSE ADR and 0 otherwise. As accounting standard variables:  LEVEL denotes the accounting standards level of each ADR.

Institutional Investors’ Participation in Foreign Firms

153

 LEVEL 1: a dummy variable that takes the value 1 if the ADR is under level 1 and 0 otherwise.  LEVEL 2: a dummy variable that takes the value 2 if the ADR is under level 2 and 0 otherwise. As privatization variable:  PRIVATE: a dummy variable that takes the value 1 if it is a privatized firm and 0 otherwise. As liquidity variable:  LIQUIDITY denotes market capitalization divided by trading volume. As firm performance variables:  ROE is the return on equity.  ROA is the return on assets. As leverage variable:  LEVERAGE 1 is total debt divided by the book value of common stock.  LEVERAGE 2 Is total debt divided by the firm’ asset. As firm visibility variables:  MV denotes the logarithm of market capitalization.  SIZE denotes the logarithm of total assets. As growth opportunity variables:  MB denotes market to book ratio.  SALES-G denotes percent change in total sales.  PER denotes price to earnings ratio. As payout variable:  DIVY denotes the dividend yield of the firm.  DIVP denotes dividend payout ratio measured by the ratio annual dividends per share to earnings per share. As stock return variable:  MR denotes the average annual stock market return of the firm.

154

NARJESS BOUBAKRI ET AL.

As risk variable:  BETA denotes market beta of the firm. As industries variables:  TECHNOLOGY is a dummy variable that takes the value 1 if the firm is reported to operate in the high technology industry and 0 otherwise; FINANCE is a dummy variable that takes the value 1 if the firm is reported to operate in the finance services industry and 0 otherwise; NON FIN SERVICES is a dummy variable that takes the value 1 if the firm is reported to operate in the nonfinancial services industry and 0 otherwise; PRODUCTION is a dummy variable that takes the value 1 if the firm is reported to operate in the production industry and 0 otherwise. Country-Level Variables As macroeconomic variables:  GDP denotes gross domestic product.  GDP-G denotes gross domestic product growth rate. As legal system variable:  COMMON: a dummy variable that takes the value 1 if the country’s legal structure is based on the English common law and 0 otherwise. As investor protection variables:  CORRUPTION denotes the International Country Risk’s (ICR) assessment of corruption in government. Scale from 0 to 6, with lower scores for higher levels of corruption.  LAW&ORDER denotes the assessment of the strength and impartiality of the legal system and of the popular observance of the law in the country produced by the country-risk rating agency ICR. Scale from 0 to 6, with lower scores for less tradition for law and order. Table 1 (panels A and B) report sample distribution by countries and industries, respectively. We observe that 22% of our ADR sample is from Brazil and 41% from technology issuers. Table 1, panel C, reports sample distribution by stock markets. We find that, as expected, NYSE ADRs represent the largest part of our sample. Table 2 provides descriptive statistics for our variables. We find that 39% of ADRs are capital raising level 2, 30% of issuers are privatized, and 21% of ADRs are from common

155

Institutional Investors’ Participation in Foreign Firms

Table 1.

Panel A: Sample Distribution by Countries.

Country Argentina Brazil Chile China Greece Hungary India Indonesia Israel Korea Mexico Peru Philippines Portugal Russia Singapore South Africa Taiwan Turkey Venezuela Total

Number of firms 8 25 14 8 2 1 8 2 5 5 13 1 1 1 4 1 6 5 1 1 112

Panel B: Sample Distribution by Industries Industries Technology Financial services Transport Production Consumer Total

Number of firms 46 17 7 30 12 112

Panel C: Sample Distribution by Stock Markets Markets NYSE Nasdaq Amex Total

Number of Firms 100 11 1 112

156

Table 2. Variable

Mean

Standard Deviation

Minimum

Maximum

0.1376 0.0719 0.01853 0.8928 5.6785 0.3928 0.3035 167.6205 0.08823 0.0566 0.6667 3.8692 21.9374 0.4727 0.0156 1.2954 0.1524 25.8599 2.6887 1.64e07 22.0116 8.3812 0.2343 0.2142

0.1777 0.0973 0.1813 0.3106 3.5290 0.4905 0.4618 642.9979 0.5823 0.1084 0.8045 5.1510 1.3962 1.0015 0.0240 0.8567 0.3159 45.5962 2.4963 7.66e07 1.5193 0.9326 0.3061 0.4121

0 0 0 0 0 0 0 1.26 4.9899 0.6179 0 0.5 18.7284 0 0 2 0.458 69.8 0.0011 4.91e08 17.2277 6.4134 0.4742 0

0.8898 0.3891 0.6064 1 14 1 1 5276.16 0.6614 0.2801 4.89 39.8 25.3962 8.13 0.113 3.5 0.965 266 12.997 91.2 25.7766 10.18074 1.0109 1

NARJESS BOUBAKRI ET AL.

TOP50 MFALL TOP50-ADJUSTED NYSE ADR-YEARS LEVEL PRIVATE LIQUIDITY ROE ROA LEVERAGE 1 LEVERAGE 2 MV DIVPR DIVY BETA MR PER MB SALES-G SIZE GDP GDP-G COMMON

Descriptive Statistics.

3.4107 2.9732 0.4107 0.1517 0.0625 0.2678

1.3189 0.9905 0.4941 0.3604 0.2431 0.4448

2 1 0 0 0 0

6 5 1 1 1 1

Notes: This table provides descriptive statistics for all our dependent and independent variables. TOP50 is the percentage of ADR allocated to the top 50 institutional investors; TOP50-ADJUSTED is the industry-adjusted percentage of ADR allocated to the top 50 institutional investors; MFALL is the percentage of ADR allocated to mutual funds; ADR-YEARS denote the number of years since ADR; NYSE is a dummy variable that takes the value 1 if it is a NYSE ADR and 0 otherwise. PRIVATE is a dummy variable that takes the value 1 if it is a privatized firm and 0 otherwise; LEVERAGE 1 is total debt divided by the book value of common stock; LEVERAGE 2 is total debt divided by the firm’ asset; ROE is return on equity; ROA is return on asset; MV denotes the logarithm of market capitalisation; SIZE denotes the logarithm of total assets; LEVEL denotes the accounting standards level of each ADR; LEVEL 1 is a dummy variable that takes the value 1 if the ADR is under level 1 and 0 otherwise; LEVEL 2 is a dummy variable that takes the value 2 if the ADR is under level 2 and 0 otherwise; LIQUIDITY denotes market capitalization divided by trading volume; DIVY denote the dividend yield of the firm; MR denotes average annual stock market return; SALES-G denotes percent change in total sales; BETA denotes market beta of the firm; TECHNOLOGY is a dummy variable that takes the value 1 if the firm is reported to operate in the high-technology industry and 0 otherwise; FINANCE is a dummy variable that takes the value 1 if the firm is reported to operate in the finance services industry and 0 otherwise; NON FIN SERVICES is a dummy variable that takes the value 1 if the firm is reported to operate in the nonfinancial services industry and 0 otherwise; PRODUCTION is a dummy variable that takes the value 1 if the firm is reported to operate in the production industry and 0 otherwise; COMMON is a dummy variable that takes the value 1 if the country’s legal structure is based on the English common law and 0 otherwise; PER denotes price to earnings ratio; DIVP denotes dividend payout ratio measured by the ratio annual dividends per share to earnings per share; GDP denotes gross domestic product; GDP-G denotes gross domestic product growth rate; MB denotes market to book ratio; CORRUPTION denotes the International Country Risk’s assessment of the corruption in government. Scale from 0 to 6, with lower scores for higher levels of corruption; LAW&ORDER denotes the assessment of the law and order tradition in the country produced by the country-risk rating agency International Country Risk (ICR). Scale from 0 to 6, with lower scores for less tradition for law and order.

Institutional Investors’ Participation in Foreign Firms

LAW&ORDER CORRUPTION TECHNOLOGY FINANCE NON-FIN SERVICES PRODUCTION

157

158

NARJESS BOUBAKRI ET AL.

law countries. Our descriptive analysis also confirms the presence of certain outliers. For this reason, we will rely on quintile regressions rather than OLS ones in fourth section.

EMPIRICAL RESULTS Univariate Analysis Results Table 3 reports univariate results by firm characteristics and Table 4 reports those by country characteristics. Overall, the univariate analysis reveals many significant results. For instance, when we split our sample according to the stock exchange for the ADR listing, results show that the industryadjusted institutional investors and mutual funds participations are significantly higher for NYSE interlisted ADRs (0.5387 and 0.3496, respectively) than non-NYSE interlisted ones (0.504 and 0.0430, respectively). These differences are significant at the 10% level for both cases. This result confirms Baker, Nofsinger, and Weaver’s (2002) observation that international firms listing their shares on the NYSE are those who experience the highest increase in visibility, as proxied by analyst coverage and print media attention (The Wall Street Journal or Financial Times). When we partition our sample by size, results are more significant. The percentage allocations of institutional investors and mutual funds in large firm ADRs are significantly higher than those in small firms. The t-statistics and the z-Wilcoxon tests for the differences in the percentages of allocation are significant at the 1%, 5%, and 10% levels. This confirms evidence in Aggarwal et al. (2005) for mutual funds investors in ADRs following the financial crises of the late 1990s. Regarding stock liquidity, results show that institutional investors and mutual funds prefer to invest in more liquid stocks (0.1560 and 0.5349 versus 0.1152 and 0.0907). The z-statistics for the difference in the ADR allocation percentages are significant at the 10% level confirming earlier evidence in Dahlquist and Robertsson (2001). When it comes to profitability (ROE), we find that institutional investors’ ADR participations are higher in more profitable firms (0.1717 for ROE Z median versus 0.1035 ROE o median), t-statistic ¼ 2.0606). Moreover, mutual funds and institutional investors’ ADR participations are higher in firms with more leverage (using TOP50: 0.1717 for LEVERAGE 1 Z median versus 0.1002 for LEVERAGE 1 o median and using MFALL: 0.0919 for LEVERAGE 1 Z median versus 0.0520 for

159

Institutional Investors’ Participation in Foreign Firms

Table 3. Variable

Non-NYSE NYSE z-Statistic t-Statistic ADR-YEARS (omedian) ADR-YEARS (Zmedian) z-Statistic t-Statistic LEVEL 1 LEVEL 2 z-Statistic t-Statistic SIZE (omedian) SIZE (Zmedian) z-Statistic t-Statistic Non-PRIVATE PRIVATE z-Statistic t-Statistic LIQUIDITY (omedian) LIQUIDITY (Zmedian) z-statistic t-statistic ROE (omedian) ROE (Zmedian) z-Statistic t-Statistic LEVERAGE 1 (omedian) LEVERAGE 1 (Zmedian) z-Statistic t-Statistic DIVY (omedian) DIVY (Zmedian) z-Statistic t-Statistic BETA (omedian) BETA (Zmedian) z-Statistic t-Statistic MR (omedian) MR (Zmedian) z-Statistic

Univariate Analysis: Firm Level. Test 1: TOP50

Test 2: MFALL

Test 3: TOP50-ADJUSTED

0.0794 (12) 0.1445 (100) 1.4910 1.2010 0.1308 (60) 0.1454 (52) 0.3530 0.4337 0.1340 (68) 0.1430 (44) 0.6490 0.2594 0.1049 (56) 0.1702 (56) 2.4330 0.1968 0.1299 (78) 0.1551 (34) 0.1840 0.6872 0.1152 (55) 0.1560 (57) 1.7990 0.3624 0.1035 (56) 0.1717 (56) 1.4200 2.0606 0.1002 (56) 0.1749 (56) 1.8420 2.2647 0.0974 (58) 0.1702 (54) 2.7430 2.5392 0.1255 (52) 0.1223 (60) 1.7160 0.9733 0.1530 (56) 0.21 (56) 0.3000

0.0430 (12) 0.3496 (100)  1.7460 0.3862 0.5276 (60) 0.0735 (52) 0.6270 0.9251 0.4725 (68) 0.0761 (44) 0.3340 0.7902 0.0528 (56) 0.0911 (56) 2.6400 2.1174 0.4255 (78) 0.0743 (34) 0.2980 0.6527 0.0907 (55) 0.5349 (57) 1.9100 0.9070 0.5481(56) 0.0854(56) 0.8910 0.9452 0.0520 (56) 0.0919 (56) 1.6010 2.2055 0.0452 (58) 0.6085 (54) 3.53 1.1523 0.8370 (52) 0.0618 (60) 2.0630 1.1909 0.079(56) 0.5557(56) 0.099

0.5041(12) 0.5387(100) 1.8810 0.3998 0.8649 (60) 0.0255 (52) 0.0440 0.9158 0.7657(68) 0.0262 (44) 0.7000 0.7893 0.0175 (56) 0.0546 (56) 2.9620 2.1416 0.6632 (78) 0.0437 (34) 0.4430 0.6219 0.0356 (55) 0.8993 (57) 0.8470 0.9448 0.0591 (56) 0.8912 (56) 2.2080 0.9102 0.0149 (56) 0.0520 (56) 1.2770 1.9794 0.0202 (58) 1.0073 (54) 2.5970 1.1254 0.4073 (52) 0.0007 (60) 1.3650 1.2085 0.0247 (56) 0.9256 (56) 0.5000

160

NARJESS BOUBAKRI ET AL.

Table 3. (Continued ) Variable

t-Statistic SALES-G (omedian) SALES-G (Zmedian) z-Statistic t-Statistic

Test 1: TOP50

Test 2: MFALL

Test 3: TOP50-ADJUSTED

0.9188 0.1213 (56) 0.1538 (56) 1.8970 0.9686

0.9764 0.0579 (56) 0.0860 (56) 2.0580 1.5377

0.9861 0.0061 (56) 0.0432 (56) 1.9990 1.4483

Notes: TOP50 is the percentage of ADR allocated to the top 50 institutional investors. TOP50ADJUSTED is the industry-adjusted percentage of ADR allocated to the top 50 institutional investors. MFALL is the percentage of ADR allocated to mutual funds. ADR-YEARS denote the number of years since ADR. NYSE is a dummy variable that takes the value 1 if it is a NYSE ADR and 0 otherwise. PRIVATE is a dummy variable that takes the value 1 if it is a privatized firm and 0 otherwise. LEVERAGE 1 is total debt divided by the book value of common stock. ROE is return on equity. SIZE denotes the logarithm of total assets. LEVEL denotes the accounting standards level of each ADR. LEVEL 1 is a dummy variable that takes the value 1 if the ADR is under level 1 and 0 otherwise. LEVEL 2 is a dummy variable that takes the value 2 if the ADR is under level 2 and 0 otherwise. LIQUIDITY denotes market capitalization divided by trading volume. DIVY denote the dividend yield of the firm. MR denotes average annual stock market return. SALES-G denotes percent change in total sales. BETA denotes market beta of the firm. The z-statistics for the Wilcoxon–Mann–Whitney test and t-statistics for the Student tests are reported to compare each two groups of firms. Numbers of observations in each group are shown in parentheses. Statistical significance at the 1% level. Statistical significance at the 5% level. Statistical significance 10% level.

LEVERAGE 1omedian) and more sales growth (using TOP50: 0.1538 for SALES-GZmedian versus 0.1213 for SALES-G omedian and using MFALL: 0.0860 for SALES-GZmedian versus 0.0579 for SALES-G omedian). The leverage result contrasts with previous evidence in Kang and Stulz (1997) on Japan and Dahlquist and Robertsson (2001) on Sweden. Finally, results in Table 3 note that institutional investors and mutual funds prefer investing in high dividend yield firms and less risky ADRs. The dividend result is consistent with Aggarwal et al. (2005) that argue that firms in the US mutual funds portfolio holdings that overlap with the MSCI Index have higher dividend yield. Table 4 reports the results of the univariate analysis at the country level. We find that institutional investors seem to prefer investing in countries with high GDP growth. GDP growth is indeed good indicator for future prosperity and firm’s profitability. However, we do not find that corruption, the country’s legal structure, and investor protection affect institutional investors’ participation in ADRs from emerging markets.

161

Institutional Investors’ Participation in Foreign Firms

Table 4. Variable

GDP (omedian) GDP (Zmedian) z-Statistic t-Statistic GDP-G(omedian) GDP-G (Zmedian) z-Statistic t-Statistic CIVIL COMMUN z-Statistic t-Statistic LAW&ORDER (omedian) LAW&ORDER (Zmedian) z-Statistic t-Statistic CORRUPTION (omedian) CORRUPTION (Zmedian) z-Statistic t-Statistic

Univariate analysis: country level. Test 1: TOP50

Test 2: MFALL

Test 3: TOP50-ADJUSTED

0.1279(57) 0.1476(55) 1.4090 0.5836 0.1884 (54) 0.0902 (58) 0.2040 3.0289 0.1495(88) 0.0938(24) 0.0210 1.3670 0.1497(78) 0.1096(34) 0.0380 1.0992 0.1290 (79) 0.1580 (33) 0.5970 0.7837

0.0721(57) 0.5703(55) 1.2140 1.0185 0.0955 (54) 0.5228 (58) 0.2010 0.8718 0.3883(88) 0.0545(24) 0.2270 0.5582 0.4321(78) 0.0522(34) 0.2500 0.7125 0.4167(79) 0.0775(33) 0.0800 0.6305

0.0082(57) 0.9591(55) 0.7220 .1.0410 0.0703 (54) 0.8521 (58) 1.4970 0.8541 0.6125(88) 0.0283(24) 0.6590 0.5740 0.6859(78) 0.0083(34) 0.7720 0.6972 0.6571(76) 0.0396(33) 0.3450 0.6145

Notes: TOP50 is the percentage of ADR allocated to the top 50 institutional investors; TOP50ADJUSTED is the industry-adjusted percentage of ADR allocated to the top 50 institutional investors; MFALL is the percentage of ADR allocated to mutual funds; GDP denotes gross domestic product; GDP-G denotes gross domestic product growth rate; COMMON is a dummy variable that takes the value 1 if the country’s legal structure is based on the English common law and 0 otherwise; CIVIL is a dummy variable that takes the value 1 if the country’s legal structure is based on the civil law and 0 otherwise; CORRUPTION denotes the International Country Risk’s assessment of the corruption in government. Scale from 0 to 6, with lower scores for higher levels of corruption; LAW&ORDER denotes the assessment of the law and order tradition in the country produced by the country-risk rating agency International Country Risk (ICR). Scale from 0 to 6, with lower scores for less tradition for law and order. The z-statistics for the Wilcoxon–Mann–Whitney test and t-statistics for the Student tests are reported to compare each two groups of firms. Numbers of observations in each group are shown in parentheses. Statistical significance at the 1% level. Statistical significance at the 5% level. Statistical significance 10% level.

Table 5. 1 ADR-YEARS LEVEL PRIVATE LIQUIDITY ROE ROA LEVERAGE 1 LEVERAGE 2 MV SIZE PER DIVY BETA MR INDUSTRYCONTROLS NYSE GDP GDP-G COMMON LAW&ORDER CORRUPTION Constant Observations Pseudo R2

Multivariate Analysis: Institutional Investors’ Allocation (1).

Model 1 Firm 0.0022 (0.444) 0.0216 (0.444) 0.0249 (0.289) 8.3C.06 (0.257) 0.0014 (0.896)

Model 2 Firm 0.0004 0.0147 0.0006 0.0001 0.0078

Model 1 Country

Model 2 Country

Model 1 Firm þ Country 0.0037 (0.007) 0.0236 (0.018)

(0.873) (0.398) (0.973) (0.114) (0.401)

0.0013 (0.893) 0.0001 (0.000) 0.0074 (0.880) 0.0055 (0.316)

0.0020 (0.361) 0.0211 (0.010)

0.0013 (0.532)

0.0001 (0.566) 1.2477 (0.008) 0.0112 (0.0372) 0.0064 (0.822) included

0.01137 (0.128) 0.0003 (0.167) 0.6955 (0.076) 0.0251 (0.021) 0.0262 (0.294) included

0.0099 (0.755)

0.0311 (0.260) 0.0107 0.0275 0.0119 0.0148

0.4222 (0.021) 105 0.0624

0.1809 (0.258) 106 0.0600

(0.437) (0.533) (0.704) (0.154)

0.0491 (0.673) 112 0.0102

0.0091 (0.536) 0.0058 (0.895) 0.0121 (0.242) 0.0055 (0.694) 0.0372 (0.750) 112 0.0102

Model 2 Firm þ Country 0.0024 0.0328 0.0217 0.0001 0.0056

(0.000) (0.000) (0.000) (0.000) (0.000)

0.0318 (0.000)

0.0008 (0.000) 0.0356 (0.000)

0.0004 (0.000) 0.1883 (0.342) 0.5197 (0.000) 0.0535 (0.000) included

0.0005 (0.000) 0.2453 (0.000) 0.0574 (0.000) 0.0505 (0.000) included

0.0342 (0.039) 0.0365 0.000) 0.0710 (0.000) 0.0508 (0.000) 0.0216 (0.000) 0.0014 (0.783) 0.8949 (0.000) 106 0.1147

0.0232 (0.000) 0.0367 (0.000) 0.0931 (0.000) 0.0621 (0.000) 0.0276 (0.000) 0.0114 (0.000) 0.0114 (0.000) 105 0.0970

Notes: This table reports results from OLS regressions for our base model of institutional investors’ allocation. Independent variable is TOP50 (the percentage of ADR allocated to the top 50 institutional investors). Dependent variables are ADR-YEARS, LEVEL, PRIVATE, LIQUIDITY, ROE, ROA, LEVERAGE 1, LEVERAGE 2, MV, SIZE, PER, DIVY, BETA, MR, NYSE, GDP, GDP-G, COMMON, LAW&ORDER, and CORRPTION. INDUSTRY CONTROLS are included but not reported. Variables are defined in Table 4. For each independent variable, we report the estimated coefficient, and the p-values for the coefficients, corrected for heteroskedasticity, are reported in parentheses. Significant at 1%. Significant at 5%. Significant at 10%.

Institutional Investors’ Participation in Foreign Firms

163

Multivariate Analysis Results We consider in this section several models to examine the determinants of institutional investors’ ADR allocation. Table 5 reports results for institutional investors’ participation, using TOP50 as a dependent variable and various models for each set of independent variables. In model 1, we consider firm characteristics. Results show that the major common drivers of institutional investors’ ADR allocation are firm size, dividend yield, and risk. Specifically, we find positive and significant relations between TOP50 and both market capitalization and dividend yield and a negative and significant relationship with beta. Thus, institutional investors prefer large and less risky firms with a high dividend yield. Model 2, is a variant of model 1, and uses alternative measures for both profitability and leverage. Results of model 2 confirm the results found earlier in model 1 and the univariate analysis. When we incorporate both firm and country characteristics (models 5 and 6), we find that the coefficient of ADR-YEARS, LIQUIDITY, MV, PER, MR, GDP, and NYSE are positive and significant. In other words, institutional investors prefer investing in large and performing firms, listed on the NYSE, and established in high-GDP countries. However, we find that the coefficient of LEVEL, BETA, COMMON, LAW&ORDER, and GDP-G are negative and significant. The diversification purpose could explain these observations. In contradiction with Aggarwal et al. (2005), we note that institutional investors are seeking to generate returns by investing in risky environments with week shareholders protection. The pseudo R2 for all models ranges from 0.01 to 0.11. Table 6 reports results for mutual funds using the percentage of ADR held by US mutual funds as a dependent variable. As explained earlier, model 1 considers only firm characteristics. Results show that the major common drivers of mutual funds ADR allocation are LEVEL, ROE (with positive and significant coefficients), and BETA (negative and significant). Thus, US mutual funds seem to prefer investing in risky but profitable firms and those with high accounting standards. Model 2 also confirms these observations. When we incorporate both firm and country variables (model 5), we confirm that mutual funds consider investing in firms established in countries that show high economic growth (GDP coefficient positive and significant at the 1% level) with high institutional deficiencies (law and order coefficient negative and significant at the 1% level) and showing high systematic risk (beta-coefficient negative and significant at the 1% level), probably for diversification purposes. The pseudo R2 for the models

Table 6. Variables ADR-YEARS LEVEL PRIVATE LIQUIDITY ROE ROA LEVERAGE 1 LEVERAGE 2 MV SIZE PER DIVY BETA MR INDUSTRY-CONTROLS NYSE GDP GDP-G COMMON LAW&ORDER CORRUPTION Constant Observations Pseudo R2

Multivariate Analysis: Mutual Funds’ Allocation.

Model 1 Firm

Model 2 Firm

0.0005 (0.716) 0.0191 (0.075) 0.0167 (0.182) 3.46e.06 (0.523) 0.0072 (0.057)

0.0002 (0.822) 0.0229 (0.006) 0.0047 (0.609) 2.39e.06 L(0.461) 0.0055 (0;055)

0.0011 (0.836) 0.0073 (0.103) 0.0001 (0.362)

0.0006 (0.911) 0.0051 (0.096)

0.0012 (0.750) 0.0146 (0.039) 0.0235 (0.140) included 0.0199 (0.266)

Model 1 Country

0.0037 (0.024) 0.4378 (0.004) 0.0169 (0.001) 0.0102 (0.409) included 0.0033 (0.816) 0.0076 (0.322) 0.0031 (0.893) 0.0096 (0.088)

0.1094 (0.252) 107 0.0711

Model 2 Country

0.8726 (0.201) 107 0.0807

0.0092 (0.886) 112 0.0031

0.0052 (0.542) 0.0067 (0.779) 0.0078 (0.670) 0.0095 (0.114) 0.0035 (0.655) 0.0175 (0.794) 112 0.0040

Model 1 Firm þ Country

Model 2 Firm þ Country

0.0006 (0.537) 0.0037 (0.597) 0.0074 (0.335) 6.39e.06 (0.022)

0.0005 (0.246) 0.0038 (0.296) 0.0076 (0.055) 6.15e.06 (0.000) 0.0063 (0.000)

0.0053 (0.841) 0.0072 (0.147) 0.0031 (0.298)

0.0065 (0.011) 0.0028 (O.062)

0.0059 (0.000)

0.0056 (0.000)

0.1101 (0.387) 0.0143 (0.001) 0.0115 (0.282) included 0.0019 (0.864) 0.0154 (0.000) 0.0029 (0.788) 0.0019 (0.815) 0.0099 (0.001) 0.0026 (0.505) 0.1531 (0.034) 107 0.1020

0.0841 (0.246) 0.0140 (0.000) 0.0114 (0.039) included 0.0014 (0.809) 0.0157 (0.000) 0.0037 (0.506) 0.0009 (0.826) 0.0096 (0.000) 0.0028 (0.175) 0.1486 (0.000) 107 0.1065

Notes: This table reports results from OLS regressions for our base model of mutual funds’ allocation. Independent variable is MUALL (the percentage of ADR allocated to mutual funds). Dependent variables are ADR-YEARS, LEVEL, PRIVATE, LIQUIDITY, ROE, ROA, LEVERAGE 1, LEVERAGE 2, MV, SIZE, PER, DIVY, BETA, MR, NYSE, GDP, GDP-G, COMMON, LAW&ORDER, and CORRPTION. INDUSTRY CONTROLS are included but not reported. Variables are defined in Table 4. For each independent variable, we report the estimated coefficient, and the p-values for the coefficients, corrected for heteroskedasticity, are reported in parentheses. Significant at 1%. Significant at 5%. Significant at 10%.

Table 7. Variables ADR-YEARS LEVEL PRIVATE LIQUIDITY ROE ROA LEVERAGE 1 LEVERAGE 2 MV SIZE PER MB DIVPR DIVY BETA MR INDUSTRY-CONTROLS NYSE GDP GDP-G COMMON LAW&ORDER CORRUPTION Constant Observations Pseudo R2

Multivariate Analysis: Institutional Investors’ Allocation (2).

Model 1 Firm 0.0011 (0.598) 0.0218 (0.142) 0.0142 (0.408) 1.02e.06 (0.887) 0.0176 (0.136)

0.0093 (0.000) 0.0071 (0.215) 0.0002 (0.152) 0.0091 (0.380) 0.0125 (0.191) 0.0417 (0.060) included 0.0398 (0.086)

Model 2 Firm

Model 1 Country

0.0092 (0.852) 0.0109 (0.124)

(0.263) (0.490) (0.000) (0.006) (0.653)

0.0024 (0.003) 0.0172 (0.000)

0.0147 (0.001) 0.0002 (0.069)

0.0119 (0.000) 0.0014 (0.747)

0.8779 (0.000) 0.0081 (0.244) 0.0226 (0.187) included 0.0097 (0.594)

0.4501 (0.000) 107 0.1057

Model 1 Firm þ Country 0.0010 0.0044 0.0310 0.0001 0.0014

0.0067 (0.547) 0.0094 (0.472) 5.35e.06 (0.183)

0.0011 (0.938) 0.0116 (0.759)

0.3069 (0.016) 105 0.0909

Model 2 Country

0.0058 (0.000)

0.0115 (0.217) 0.0009 (0.937) 0.0085 (0.934) 112 0.0025

0.0061 (0.646) 0.0069 (0.862) 0.0034 (0.907) 0.0003 (0.982) 0.0191 (0.858) 112 0.0006

0.0287 (0.000) 0.0247 (0.014) included 0.0291 (0.014) 0.0335 (0.000) 0.0234 (0.045) 0.0576 (0.000) 0.0172 (0.000) 0.0064 (0.091) 0.7704 (0.000) 105 0.1412

Model 2 Firm þ Country 0.0065 0.0263 0.0081 0.0001 0.0001

(0.000) (0.000) (0.111) (0.000) (0.000)

0.0026 (0.277) 0.0293 (0.000) 0.0003 (0.000) 0.0009 (0.579) 0.0278 (0.000) 0.0555 (0.000) included 0.0252 (0.002) 0.0289 (0.000) 0.0310 (0.000) 0.0471 (0.000) 0.0104 (0.000) 0.0026 (0.307) 0.9607 (0.000) 105 0.1111

Notes: This table reports results from OLS regressions for our base model of institutional investors’ allocation. Independent variable is TOP50ADJUSTED (the industry adjusted percentage of ADR allocated to the top 50 institutional investors). Dependent variables are ADR-YEARS, LEVEL, PRIVATE, LIQUIDITY, ROE, ROA, LEVERAGE 1, LEVERAGE 2, MV, SIZE, PER, DIVY, BETA, MR, NYSE, GDP, GDP-G, COMMON, LAW&ORDER, and CORRPTION. INDUSTRY CONTROLS are included but not reported. Variables are defined in Table 4. For each independent variable, we report the estimated coefficient, and the p-values for the coefficients, corrected for heteroskedasticity, are reported in parentheses. Significant at 1%. Significant at 5%. Significant at 10%.

166

NARJESS BOUBAKRI ET AL.

ranges from 0.003 to 0.10. It increases as we add the country level explanatory variables to the firm level determinants identified by Kang and Stulz (1997). Table 7 reports results for all types of institutional investors, using TOP50-ADJUSTED as a dependent variable and several models for each set of independent variables. Results of Table 7 confirm the previous results reported in Table 5.

SUMMARY AND CONCLUSION In this chapter, we examine the firm-level and country-level determinants of US institutional investors’ holdings in ADRs. Specifically, we focus on the motivations of investors when they choose to invest in the ADR from emerging markets. In addition, rather than focusing on institutional holdings from just one class of institutions, we consider all types of institutional investors that acquire a participation in an ADR firm. We measure portfolio investment allocation of institutional investors on a raw basis and on an adjusted basis (i.e., in comparison with the industry average institutional investors’ participation). Using a sample of 112 firms from emerging markets that listed as ADRs between 1990 and 2005, we find that institutional investors hold higher stakes in foreign firms that are listed on more restrictive exchanges, in large, privatized, more liquid, and more transparent firms. Mutual investors and other institutional investors also prefer firms from countries with weaker institutional environments and from civil law legal tradition. Controlling for country-level determinants increases significantly the explanatory power of the model. Our findings highlight that institutional investors’ allocations are determined not only by firm-level characteristics but also by the characteristics of the country of origin of the firm. Our results also provide insights to policymakers in emerging markets willing to improve governance mechanisms and attract US institutional investment.

NOTES 1. Adverse effects on domestic markets following ADR issuance are however found in some studies that show that issuing firms’ domestic markets suffer from a reduction in size, liquidity, and growth (Karolyi, 2006; Levine & Schmukler, 2006). 2. This hypothesis was later questioned by Siegel (2005, 2009).

Institutional Investors’ Participation in Foreign Firms

167

ACKNOWLEDGMENTS We would like to acknowledge the useful comments of our colleagues. Special thanks to Najia Saikouk for her excellent research assistance.

REFERENCES Aggarwal, R., Dahiya, S., & Klapper, L. (2007). ADR holdings of US-based emerging market funds. Journal of Banking & Finance, 31, 1649–1667. Aggarwal, R., Erel, I., Ferreira, M., & Matos, P. (2011). Does governance travel around the world? Evidence from institutional investors. Journal of Financial Economics, 100, 154–181. Aggarwal, R., Klapper, L., & Wysocki, P. (2005). Portfolio preferences of foreign institutional investors. Journal of Banking & Finance, 29, 2919–2946. Ahearne, A. G., Griever, W. L., & Warnock, F. E. (2004). Information costs and home bias: An analysis of U.S. holdings of foreign equities. Journal of International Economics, 62, 313–336. Baker, H. K., Nofsinger, J. R., & Weaver, D. G. (2002). International cross-listing and visibility. Journal of Financial and Quantitative Analysis, 37, 495–521. Barth, M. E., Clinch, G., & Shibano, T. (1999). International accounting harmonization and global equity markets. Journal of Accounting and Economics, 26, 201–235. Bekaert, G., Harvey, C. R., & Lundblad, C. (2005). Does financial liberalization spur growth? Journal of Financial Economics, 77, 3–55. Bekaert, G., Harvey, C. R., & Lundblad, C. (2007). Liquidity and expected returns: Lessons from emerging markets. The Review of Financial Studies, 20, 1783–1831. Bradshaw, M. T., Bushee, B. J., & Miller, G. S. (2004). Accounting choice, home bias, and U.S. investment in non-U.S. firms. Journal of Accounting Research, 42, 795–841. Bradshaw, M., & Miller, G. S. (2006). Will harmonizing accounting standards really harmonize accounting? Evidence from non-U.S. fans adopting US GAAP. Manuscript, Harvard University, Cambridge, MA. Chen, J., Hong, H., & Stein, J. C. (2002). Breadth of ownership and stock returns. Journal of Financial Economics, 66, 171–205. Coffee, J. (1999). The future as history: The prospects for global convergence in corporate governance and its implications. Northwestern University Law Review, 93, 641–708. Covrig, V., Lau, S., & Ng, L. (2006). Do domestic and foreign fund managers have similar preferences for stock characteristics? A cross country analysis. Journal of International Business Studies, 37, 407–429. Dahlquist, M., & Robertsson, G. (2001). Direct foreign ownership, institutional investors, and firm characteristics. Journal of Financial Economics, 59, 413–440. Doidge, C., Karolyi, G. A., & Stultz, R. M. (2002). Why are foreign firms listed in the US. Worth more? Journal of Financial Economics, 71, 205–238. Dye, R. A., & Sunder, S. (2001). Why not allow the FASB and the IASB standards to compete in the US? Accounting Horizons, 15, 257–271. Edison, H. J., & Warnock, F. E. (2004). U.S. investors’ emerging market equity portfolios: A security-level analysis. The Review of Economics and Statistics, 86, 691–704.

168

NARJESS BOUBAKRI ET AL.

El Gazzar, S. (1998). Predisclosure information and institutional ownership: A cross-sectional examination of market revaluations during earnings announcement periods. The Accounting Review, 29, 119–129. Errunza, V., & Miller, D. P. (2003). Valuation effects of seasoned global equity offerings. Journal of Banking & Finance, 27, 1611–1623. Falkenstein, E. G. (1996). Preferences for stock characteristics as revealed by mutual fund portfolio holdings. Journal of Finance, 51, 111–135. Ferreira, M. A., & Matos, P. (2008). The colors of investors’ money: The role of institutional investors around the world. Journal of Financial Economics, 88, 499–533. Gillian, S., & Starks, L. T. (2003). Corporate governance, corporate ownership, and the role of institutional investors: A global perspective. Journal of Applied Finance, 13, 4–22. Kang, J. K., & Stulz, R. (1997). Why is there a home bias? An analysis of foreign portfolio equity ownership in Japan. Journal of Financial Economics, 46, 3–28. Karolyi, G. A. (2006). The world of cross listings and cross listings of the world: Challenging conventional wisdom. Review of Finance, 10, 99–152. Ke, B., & Petroni, K. (2004). How informed are actively trading institutional investors? Evidence from their trading behavior before a break in a string of consecutive earnings increases. Journal of Accounting Research, 42, 895–927. Krishnan, G. (2003). Which is more useful for predicting future cash flows, IAS earnings or U.S. GAAP earnings. Working Paper. City University of Hong Kong, Hong Kong. Levine, R., & Schmukler, S. L. (2006). Internationalization and stock market liquidity. Review of Finance, 10, 153–187. Merton, R. (1987). A simple model of capital market equilibrium with incomplete information. Journal of Finance, 42, 483–510. Pagano, M., Roell, A. A., & Zechner, J. (2002). The geography of equity listing: Why do companies list abroad? Journal of Finance, 57, 2651–2694. Siegel, J. (2005). Can foreign firms bond themselves effectively by renting US securities laws? Journal of Financial Economics, 75, 319–359. Siegel, J. (2009). Is there a better commitment mechanism than cross-listings for emergingeconomy firms? Evidence from Mexico. Journal of International Business Studies, 40, 1171–1191. Sunder, S. (2002). Regulatory competition among accounting standards within and across international boundaries. Journal of Accounting and Public Policy, 21, 219–234. Utama, S., & Cready, W. M. (1997). Institutional ownership, differential predisclosure precision and trading volume at announcement dates. Journal of Accounting and Economics, 24, 129–150. Walther, B. (1997). Investor sophistication and market earnings expectations. Journal of Accounting Research, 35, 157–179.

DO FOREIGN INSTITUTIONAL INVESTORS (FIIs) EXHIBIT HERDING AND POSITIVE FEEDBACK TRADING IN INDIAN STOCK MARKETS? Mangesh Tayde and S. V. D. Nageswara Rao STRUCTURED ABSTRACT Purpose – The aggregate investment by foreign institutional investors (FIIs) in the Indian stock market is significant compared to that by domestic institutions and individual (retail) investors. The question of whether FIIs exhibit herding and positive feedback trading while investing in the Indian stock markets has not been examined so far. This study is an attempt to fill the gap and contribute to the existing evidence on foreign portfolio investment in India. Methodology/approach – We have analyzed the daily data on purchases and sales of securities by FIIs sourced from the Securities and Exchange Board of India (SEBI), and the Bombay Stock Exchange (BSE). We have adopted the approach of Lakonishok et al. (1992), and Wermers (1999) to examine herding and positive feedback trading by foreign investors. Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 169–185 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012009

169

170

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

Findings – Our results suggest that FIIs exhibit herding and positive feedback trading during different phases of the stock market. This observed behavior is prominent in but not restricted to large cap stocks as they enjoy better liquidity. Social implication – The herding and positive feedback trading by FIIs is a cause for concern for government of India, capital market regulator (SEBI), and the country’s central bank (RBI) as it adversely affects stock prices and volatility. They are required to formulate and implement a suitable policy response given their objective of protecting the interests of small investors in the market. They may also have to monitor the purchases and sales of equities by FIIs in general and of better performing large cap stocks in particular. Keywords: Foreign portfolio investment (FPI); herding; positive feedback trading; volatility; destabilize; foreign institutional investors (FIIs) JEL classification: F32

INTRODUCTION Background International capital flows have significant benefits for economies around the world. Countries with sound macroeconomic policies and well functioning institutions are in a better position to reap the benefits and minimize the risks. The surge of foreign capital flows into emerging markets was reversed following the Mexican crisis in 1994, and the economic and financial turmoil in Asia, Brazil, and Russia during 1997–1998. This reversal of capital flows is evidence that emerging economies that rely on private capital flows for external financing are vulnerable to the turbulent nature of global capital markets. After the international debt crisis of the 1980s, private capital flows into many emerging economies surged during the 1990s, reflecting renewed access to international capital markets. The average annual net private capital inflow to emerging markets was US$ 150 billion during 1991–1998, compared to US$ 11 billion and US$ 22 billion during the 1970s and the 1980s, respectively. After the Asian financial crisis, the net private capital inflows plunged to an annual average of US$ 57 billion during 1998–2002.

Do FIIs Exhibit Herding and Positive Feedback Trading

171

The composition also shifted significantly from the traditional debt financing through loans and credits during the 1970s and the 1980s to nondebt financing such as foreign direct investment (FDI) and foreign portfolio investment (FPI) during the 1990s (Baek, 2006). There are different explanations of the nature and impact of foreign capital flows on developing economies. It is essential to understand the causes and behavior of foreign capital flows to implement appropriate economic policies to achieve the objective of growth in output and employment. The evidence on the impact of foreign portfolio investment flows on prices, volumes, and volatility in the Indian stock market is limited. The question of whether foreign institutional investors (FIIs) exhibit herding and positive feedback trading while investing in the Indian stock markets has not been examined so far. This study is an attempt to fill the gap and contribute to the existing evidence on foreign portfolio investment in India.

Herding A group of investors is said to exhibit industry (stock) herding if they follow one another into and out of the same industry (stock) during a period (Choi and Sias, 2009). Herding is also defined as investors buying or selling the same industry (or security) at the same time (Sias, 2004). Since investors cannot buy (sell) the same stock at the same time as trades occur sequentially, stock herding has a temporal effect.

Positive Feedback Trading According to Bikhchandani and Sharma (2000), a momentum based strategy is a tendency of investors to buy and sell stocks based on past returns, i.e., to buy recent winners and sell recent losers. This strategy will not yield (positive) abnormal returns if the market is weak form efficient. Such trading based on analysis of past prices (returns) is known as positive feedback trading, which can exacerbate price movements and add to volatility.

LITERATURE REVIEW Following the Asian financial crisis of 1997, many researchers and policy makers argued that foreign institutional investors exert a destabilizing

172

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

influence on emerging markets. The trades by foreign investors were found to be highly correlated, and they (foreign investors) were found to exhibit herding, which could destabilize the market they enter or exit due to the possibility of their trades creating panic in the market. DeLong, Shleifer, Summers, and Walmann (1990) offer an explanation of the potential destabilizing effects of positive feedback trading. The role of volatile private capital flows across borders has been debated widely in other countries, but the impact of such flows on the Indian economy is yet to be fully evaluated (Khanna, 2002). The discussion in India is mostly limited to the impact of foreign portfolio investment on macroeconomic variables. There is no evidence on herding and positive feedback trading behavior of FIIs investing in the Indian stock markets. The analysts and policy makers believe that herding by market participants increases volatility, destabilizes markets, and weakens the financial system. The reasons for herding by institutional investors, among others, include underlying flow of investors, positive feedback trading, attempts to preserve reputation by acting like other fund managers (reputational herding), inferring information from each other’s trades (information cascades), and following correlated signals. Lakonishok, Shleifer, and Vishny (1992) examined the impact of trades by pension fund managers on stock prices. They focused on herding, and positive feedback trading. They also examined if such trades destabilize stock prices. The results suggest that pension fund managers did not pursue these potentially destabilizing practices. Dornbusch and Park (1995), among others, contend that the trades of foreign investors are affected by past returns, as they buy when prices go up and sell when they fall. The evidence suggests that positive feedback trading can exert a destabilizing influence on the stock market. Grinblatt, Sheridan, and Wermer (1995) analyzed the extent to which mutual funds exhibit herding and positive feedback trading. They found the level of herding and positive feedback trading to be statistically significant. The funds, which followed momentum strategies, experienced significant abnormal returns. It is widely believed that foreign portfolio investment flows destabilize stock markets in the host nations as they are short-term flows. But, Choe, Kho, and Stulz (1999) concluded that foreign portfolio investors did not destabilize the (South) Korean stock market. They analyzed the impact of foreign portfolio investment (FPI) flows on stock returns in Korea using order and trade data. They found strong evidence of positive feedback trading and herding before the East Asian economic crisis. But, during the

Do FIIs Exhibit Herding and Positive Feedback Trading

173

crisis period, herding declined and even positive feedback trading was negligible. Kim and Wei (1999) studied the behavior of foreign portfolio investors before and during a crisis in Korea. Their results suggest that Korean institutional investors following negative feedback trading strategy, turned positive feedback traders during the crisis period. But, the nonresident institutional investors were continuously following a strategy of positive feedback trading. Nofsinger and Sias (1999) explored the relationship between institutional ownership and stock returns. They found that institutional investors engage in intra-year positive feedback trading to a greater extent than individual investors. The herding behavior of institutional investors had a larger impact on stock returns than that of individual investors. Bennett, Sias, and Starks (2003) addressed the question of whether institutional preferences change over time, reasons for such changes, and their impact on securities markets. Their results revealed that institutional investors’ preference for large, safe stocks changed in favor of smaller, riskier stocks over time as they offered ‘‘greener pastures’’ The results also suggested that institutional investors are momentum investors. Bowe and Domuta (2004) studied the behavior of individual and foreign institutional investors in the Jakarta stock exchange before and after the Asian financial crisis of 1997. The herding behavior of foreign investors was more pronounced than that of individual investors. There was no evidence of positive feedback trading in either class of investors. They conclude that the investor behavior was not inherently destabilizing and positive feedback trading did not exacerbate stock market movements in Indonesia at the time of the Asian crisis in 1997. Hwang and Salmon (2004) adopted a new approach to identify and measure herding based on the cross-sectional dispersion of factor sensitivity of assets in a given market. They implemented this approach in the United States and South Korean stock markets and found significant movements and persistence in herding during both rise and fall in stock prices. Sias (2004) did not find any evidence of herding by institutional investors during the period from March 1983 to December 1997 on NYSE, AMEX, and NASDAQ. Bohl and Brzeszczynski (2005) employed asymmetric GARCH model to test the popular belief that institutional investors engage in herding and positive feedback trading, thus contributing to autocorrelation and excess volatility. They found that foreign pension fund investors in Poland contributed to reduction in autocorrelation and volatility of returns.

174

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

Demirer and Kutan (2006) explored the presence of herding in Chinese stock markets. They analyzed the behavior of return dispersions during periods of unusually large upward and downward changes in the market index. Their findings suggest that herding does not exist in the Chinese market. Lin and Swanson (2008) examined the herding behavior and investment performance of foreign investors in the United States. After studying the strategies of foreign investors from 38 emerging and developed countries who invested in US equity markets, they reported very weak evidence of herding. Shu (2008) studied the effect of positive feedback trading by institutional investors on stock return momentum. The findings suggest that positive feedback trading moves stock prices further away from their fundamentals, thus destabilizing stock prices, and adversely affecting the efficiency of the market. Tan, Chiang, Mason, and Nelling (2008) analyzed herding behavior in each of the Chinese stock markets. The results indicate that herding exists in both A- and B-shares markets, when stock prices rise and when volumes and volatility are high. Several theories of reputation suggest that managers’ incentives affect their propensity to engage in herding behavior. Boyson (2009) examined these theories by tracking the herding behavior of hedge fund managers during their careers. There is evidence of managerial incentives for herding, which suggests that managers who deviate from the herd are more likely to fail than their junior counterparts. The implicit incentives should encourage managers to herd more as their careers progress. Choi and Sias (2009) studied the impact of Institutional industry herding on industry valuation. Their results suggest that institutional investors follow one another into and out of the same industry. Consistent with reputational herding, most institutions are more likely to follow similarly classified institutions. However, they found that investors who should be most concerned about their reputation (mutual funds and independent advisors) are more likely to herd than other investors. Duasa and Kassim (2009) examined the presence of herd behavior among foreign investors in the Malaysian stock market by employing vector error correction model. They conclude that herding by large portfolio investors destabilized the market. Goodfellow, Bohl, and Gebka (2009) compared individual and institutional herding in the Polish stock market. The herding by individual

Do FIIs Exhibit Herding and Positive Feedback Trading

175

investors was observed during the bear phase of the market, whereas there was no herding by institutional investors. Zhou and Lai (2009) analyzed the herding behavior and informational cascades of investors in the Hong Kong stock exchange. They found consistently higher frequency of herding on sell side than on the buy side. They conclude that herding still exists and is largely due to informational cascades even in a transparent market. Chang (2010) studied herding by qualified foreign institutional investors (QFIIs). This study examined how trading is determined by key players in the market, specifically, foreign institutions, dealers, margin traders, and mutual funds. The trading by these four market participants was closely interrelated with foreign institutions leading the way. Chiang and Zheng (2010) explored herding behavior in global markets. They found that most investors in the world herd with the US market in addition to their domestic markets. With the exception of US and Latin American markets, herding was present in both developed and developing markets, whereas herding asymmetry was more profound in Asian markets. Economou, Kostakis, and Philippas (2010) examined herd behavior in volatile market conditions using daily data from Greek, Italian, Portuguese, and Spanish stock markets during 1998–2008. The results reveal that herding was present in the Italian and Greek stock markets, which was stronger during bull phase of the market. Herding in Portuguese market was observed during the bear phase of the market while there was no evidence of herding in the Spanish market.

Indian Evidence The aggregate FII investment in the Indian stock market is significant compared to that by domestic institutions and individual (retail) investors. There are 1,713 FIIs registered with the capital market regulator (SEBI) at the end of March 2010. The cumulative net FII investment at the end of September 2010 is US$ 108.584 billion. Batra (2004), Trivedi and Nair (2006), Gorden and Gupta (2003), and Chakrabarti (2001) examined the impact of FPI on stock returns, and macrovariables in India. Mukherjee, Bose, and Coondoo (2002) suggested that FPI flows are caused by developments in the Indian markets and not by those in markets abroad. It is an indication that they are less likely to destabilize the Indian markets.

176

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

Batra (2003), Ananthanarayanan, Krishnamurti, and Sen (2004), and Shah and Patnaik (2006) analyzed the impact of FPI flows on market volatility in India. They concluded that FPI flows did not destabilize the Indian stock markets.

RESEARCH GAP The evidence on the impact of foreign portfolio investment flows on prices, volumes, and volatility in the Indian stock market is limited. The question of whether FIIs exhibit herding and positive feedback trading while investing in the Indian stock markets has not been examined so far. This study is an attempt to fill the gap and contribute to the existing evidence on foreign portfolio investment in India.

RESEARCH DESIGN Data The daily data on purchases and sales of securities by FIIs is sourced from the Securities and Exchange Board of India (SEBI), and the Bombay Stock Exchange (BSE). The names of FIIs are not disclosed by the capital market regulator (SEBI), and the exchange (BSE). The sample period is from January 1, 2003 to December 31, 2009. Methodology Analysis of Herding We adopted the approach of Lakonishok et al. (1992), and Wermers (1999) to examine herding by foreign investors. We estimated the herding measure with a horizon of one week, and analyzed the daily trades during all the weeks by different FIIs. Specifically, the herding measure is estimated as:   BðiÞ  ½pðtÞ  ½AFðiÞ HðiÞ ¼ ½BðiÞ þ SðiÞ

(1)

jpit  Eðpit Þj  Ejpit  Eðpit Þj

(2)

Do FIIs Exhibit Herding and Positive Feedback Trading

177

Where ‘‘B(i)’’ is the number of foreign institutional investors (FIIs) which increased their holdings (net buyers) in the stock in that week; ‘‘S(i)’’ is the number of FIIs which decreased their holdings (net sellers); ‘‘p(t)’’ is the expected proportion of FIIs buying in that week relative to the number of active ones; ‘‘E|pit  E(pit)|’’ is an adjustment factor estimated assuming that in the absence of herding, the number of FIIs with net purchases follows a binomial distribution as in Wermers (1999). We estimate the herding measure for each of the selected stocks listed on the Bombay Stock Exchange (BSE). The number of buyers and sellers changes every week, and hence the measure ‘‘p(t)’’ will also be different. Analysis of Positive Feedback Trading According to Lakonishok et al. (1992), at any level of herding, institutional investors have more potential to destabilize asset prices if they follow strong positive feedback trading strategies. We estimate net buying (aggregated across all FIIs in a given stock) in a week conditional on performance of the stock during the preceding week. We use two measures of excess demand in a week, Pratio (Rupee ratio) and Nratio (numbers ratio). For a given stock-week, i, ‘‘Pratio’’ is defined as

PratioðiÞ ¼

½buyðRsÞ  sellsðRsÞ ½buyðRsÞ þ sellðRsÞ

(3)

Where ‘‘buy(Rs)’’ denotes the increase in total rupee investment by all FIIs in the given stock week (valued at the average price prevailed during that week) while ‘‘sell(Rs)’’ is similarly defined as the decrease in total rupee investment by all FIIs. Similarly, ‘‘Nratio’’ is defined as Nratio ¼

#buyðiÞ #activeðiÞ

(4)

Where ‘‘#buy(i)’’ denotes the number of FIIs increasing the holding (in a stock) in week ‘iu’, whereas ‘‘#active(i)’’ denotes the number of FIIs changing their holdings. The mean ‘‘Pratio’’ and ‘‘Nratio’’ are estimated for all the stocks for the sample period. We employed two measures as they are expected to yield different results. For example, most FIIs might have engaged in positive feedback trading, but those following negative feedback might have

178

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

executed larger trades. In that case, trend chasing will reflect in the ‘‘Nratio’’ but not in the ‘‘Pratio.’’

RESULTS Herding The mean herding measure of the sample is 8%. It indicates the presence of herding in the Indian market though to a very limited extent. It also indicates that 58% of the FIIs are on the sell side of the market. During a period of seven years, 58% herding on an average is not uncommon. One reason for this can be equal number of transactions around the same time to offset the impact of herding (Table 1). There is no evidence of significant herding in individual stocks as FIIs may be interested in a select group of securities based on market capitalization, past performance, etc. We have divided our sample on the basis of market capitalization, and past (week) performance. The sample of listed firms is divided into small (S1), medium (S2), and large (S3) based on market capitalization (Table 2). We observe that herding in large (cap) stocks is significantly more than that in small caps due to liquidity differences between the two. Investing in small caps can be risky as they are less liquid. There is more information on large cap firms as they are tracked by many analysts. The stocks are divided into three groups based on their performance during the preceding week. We formed three portfolios of stocks, P1, P2, Table 1.

Herding for the Period (2003–2009). All Cases

Mean Median

0.081 (0.005) 0.0556

Table 2.

Mean

Classification Based on Market Capitalization. S1

S2

S3

0.10 (0.002)

0.05 (0.003)

0.30 (0.003)

179

Do FIIs Exhibit Herding and Positive Feedback Trading

Table 3.

Mean

Classification based on Past Performance. P1

P1

P3

0.062 (0.003)

0.094 (0.003)

0.095 (0.004)

Table 4. Classification based on Market Cap and Past Performance.

P1 P2 P3

S1

S2

S3

0.0190 (0.013) 0.0028 (0.0020) 0.0022 (0.003)

0.0571 (0.39) 0.0736 (0.306) 0.0742 (0.27)

0.1826 (0.0007) 0.01987 (0.00861) 0.1993 (0.0019)

S1 – Portfolio of small cap firms. S2 – Portfolio of mid-cap firms. S3 – Portfolio of large cap firms. P1 – Portfolio of stocks with lower returns during the preceding week. P2 – Portfolio of stocks with average returns during the preceding week. P3 – Portfolio of stocks with higher returns during the preceding week.

and P3 comprising stocks with lower, average, and higher returns during the preceding week, respectively. The herding in stocks which performed better in the preceding week is higher than that in others (Tables 3 and 4). The market is classified into bull and bear phase following the control chart approach, to analyze herding by FIIs when conditions in the stock market are different. We find evidence of herding during bull and bear phases of the market, which is more pronounced in large cap stocks (Tables 5 and 6). Herding during Bull Phase of the Market

Table 5.

P1 P2 P3

Herding during Bull Phase.

S1

S2

S3

0.01 (0.018) 0.05 (0.017) 0.05 (0.019)

0.17 (0.021) 0.21 (0.019) 0.21 (0.022)

0.43 (0.019) 0.46 (0.017) 0.47 (0.02)

180

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

Herding during Bear Phase of the Market Table 6.

P1 P2 P3

S1

S2

S3

0.064 (0.058) 0.031 (0.058) 0.018 (0.058)

0.094 (0.064) 0.0127 (0.064) 0.14 (0.064)

0.327 (0.068) 0.360 (0.068) 0.37 (0.068)

Table 7.

P1 P2 P3

Pratio during Normal Phase.

S1

S2

S3

0.0208 (0.01) 0.1044 (0.001) 0.1466 (0.0023)

0.0739 (0.009) 0.1570 (0.009) 0.1997 (0.008)

0.038 (0.008) 0.1221 (0.008) 0.1643 (0.007)

Table 8.

P1 P2 P3

Herding during Bear Phase.

Nratio during Normal Phase.

S1

S2

S3

0.5106 (0.002) 0.5523 (0.0026) 0.5996 (0.0023)

0.5784 (0.0025) 0.5734 (0.306) 0.5188 (0.0025)

0.5817 (0.002) 0.5367 (0.0026) 0.5605 (0.002)

Positive Feedback Trading The FIIs also follow positive feedback trading strategies in the Indian market. The results (Pratio and Nratio) suggest that positive -feedback trading was significantly high in the stocks which performed well in the past (Tables 7). Pratio The ‘‘Pratio’’ of the worst performing small cap stock is only 2% compared to that of the best performing large cap stock at 16%. This indicates that there are 66% buyers for the best performing large cap firm compared to 52% buyers (and 48% sellers) in the case of the worst performing small cap company.

181

Do FIIs Exhibit Herding and Positive Feedback Trading

Nratio The results in Table 8 suggest that there is no evidence of positive feedback trading in small cap stocks which did not perform well as 51% of FIIs are buyers in the worst performing small cap stocks compared to 59% in the best performing small cap stocks. The ‘‘Nratio’’ increases with past performance, and positive feedback trading is evident in the best performing stocks. Pratio during Bull Phase The ‘‘Pratio’’ for the worst performing small cap stocks is 30% compared to that for the best performing large cap stocks at 43%. These results clearly indicate the presence of positive feedback trading (Table 9). Pratio during Bear Phase The ‘‘Pratio’’ for the worst performing small cap stocks during the bear phase is negative (9%) while that for the best performing large cap stocks is positive (6%). The positive feedback trading increases with market capitalization and past performance during the bear phase of the market (Table 10). We have examined the ‘‘Pratio’’ during bull and bear phases of the market. We find that FIIs follow positive feedback trading in large cap stocks during the bull phase while there is no evidence of the same during the bear phase of the market. Table 9.

P1 P2 P3

S1

S2

S3

0.30 (0.02) 0.37 (0.02) 0.39 (0.03)

0.34 (0.03) 0.41 (0.02) 0.43 (0.03)

0.33 (0.02) 0.40 (0.02) 0.43 (0.03)

Table 10.

P1 P2 P3

Pratio during Bull Phase.

Pratio during Bear Phase.

S1

S2

S3

0.09 (0.03) 0.01 (0.03) 0.05 (0.03)

0.04 (0.02) 0.04 (0.02) 0.10 (0.03)

0.08 (0.02) 0.00 (0.02) 0.06 (0.02)

182

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

Nratio during Bull Phase The ‘‘Nratio’’ for the worst performing small cap stocks during the bull phase is 56% while that for the best performing large cap stocks is 61% (Table 11). Nratio during Bear Phase The ‘‘Nratio’’ for the worst performing small cap stocks during the bear phase is 46% while that for the best performing large cap stocks is 53%. The positive feedback trading by FIIs is more in the best performing large cap stocks during the bull and bear phases of the market (Table 12). We have analyzed the value (in million rupees) of FII trades in small, medium, and large cap stocks during the sample period. The trading by FIIs in large cap stocks is significantly higher than that in other stocks. This supports the earlier finding that herding by FIIs is present in large cap stocks (Table 13). Table 11.

P1 P2 P3

S1

S2

S3

0.56 (0.01) 0.59 (0.016) 0.60 (0.01)

0.58 (0.01) 0.62 (0.013) 0.63 (0.01)

0.57 (0.01) 0.60 (0.01) 0.61 (0.011)

Table 12.

P1 P2 P3

Nratio during Bull Phase.

Nratio during Bear Phase.

S1

S2

S3

0.46 (0.07) 0.50 (0.07) 0.53 (0.07)

0.48 (0.05) 0.52 (0.05) 0.55 (0.05)

0.46 (0.04) 0.50 (0.04) 0.53 (0.04)

Table 13.

Value of FII Trades.

Small Cap

Medium Cap

Large Cap

7,988.4938

3,6771.7327

402,763.2705

Do FIIs Exhibit Herding and Positive Feedback Trading

183

CONCLUSIONS The aggregate FII investment in the Indian stock market is significant compared to that by domestic institutions and individual (retail) investors. There are 1,713 FIIs registered with the capital market regulator (SEBI) at the end of March 2010. The cumulative net FII investment at the end of September 2010 is US$ 108.584 billion. We have examined the behavior of FIIs in the Indian stock markets. Our results suggest that they exhibit herding and positive feedback trading during different phases of the market. This observed behavior is prominent in but not restricted to large cap stocks as they enjoy better liquidity. The herding and positive feedback trading by FIIs is a cause for concern for government of India, capital market regulator (SEBI), and the country’s central bank (RBI) as it adversely affects stock prices and volatility. They are required to formulate and implement a suitable policy response given their objective of protecting the interests of small investors in the market. They may also have to monitor the purchases and sales of equities by FIIs in general and of better performing large cap stocks in particular. There may be a justification to retain some of the restrictions on foreign capital flows into India given their potential to destabilize the capital market. The RBI saved India from the Asian financial crisis thanks to such measures when it encouraged Indian firms to borrow (from abroad) long, and discouraged them from borrowing short, thus ensuring that the country’s short term (external) liabilities as a proportion of total liabilities was kept low.

REFERENCES Ananthanarayanan, S., Krishnamurti, C., & Sen, N. (2004). Foreign institutional investors and security returns: Evidence from Indian stock exchanges 5–6. Retrieved from http:// www.isb.edu/ISB-F/htmls/Sandhya&Sen.pdf Baek, I-M. (2006). Portfolio investment flows to Asia and Latin America: Pull, Push or market sentiment? Journal of Asian Economics, 17, 363–373. Batra, A. (2003). The dynamics of foreign portfolio inflows and equity returns in India. Indian Council for Research on International Economic Relations, New Delhi, Working Paper No. 109. Batra, A. (2004). Stock return volatility patterns in India. Indian Council for Research on International Economic Relations, New Delhi, Working Paper No. 124. Bennett, J., Sias, R., & Starks, L. (2003). Greener pastures and the impact of dynamic institutional preferences. Review of Financial Studies, 16, 1203–1238.

184

MANGESH TAYDE AND S.V.D. NAGESWARA RAO

Bikhchandani, S., and Sharma, S. (2000). Herd behavior in financial markets: A review. IMF Working Paper, WP/00/48. Bohl, M., and Brzeszczynski, J. (2005). Do institutional investors destabilize stock prices? Evidence from an emerging market. CERT Discussion Papers 0501, Centre for Economic Reform and Transformation, Heriot Watt University. Bowe, M., & Domuta, D. (2004). Investor herding during financial crisis: A clinical study of the Jakarta stock exchange. Pacific – Basin Finance Journal, 12, 387–418. Boyson, N. (2009). Implicit incentives and reputational herding by hedge fund managers. Journal of Empirical Finance. Retrieved from http://www.northeastern.edu/nicoleboyson/ repherding.pdf Chakrabarti, R. (2001). FII flows to India: Nature and causes. Money & Finance, 2(7), 1–31. Chang, C. (2010). Herding and the role of foreign institutions in emerging equity markets. Pacific – Basin Finance Journal, 18, 175–185. Chiang, T., & Zheng, D. (2010). An empirical analysis of herd behavior in global stock markets. Journal of Banking & Finance, 34, 1911–1921. Choe, H., Kho, B., & Stulz, R. (1999). Do foreign investors destabilize stock markets? The Korean experience in 1997. Journal of Financial Economics, 54, 227–264. Choi, N., & Sias, R. (2009). Institutional industry herding. Journal of Financial Economics, 94, 469–491. DeLong, J. B., Shleifer, A., Summers, L. H., & Walmann, R. J. (1990). Positive feedback investment strategies and destabilizing rational speculators. Journal of Finance, 45, 379–395. Demirer, R., & Kutan, A. (2006). Does herding behavior exist in Chinese stock markets? Journal of International Financial Markets, Institutions & Money, 16, 123–142. Dornbusch, R., & Park, Y. C. (1995). In: Financial opening: Policy lessons for Korea. Seoul, Korea: Korea Institute of Finance. Duasa, J., & Kassim, S. (2009). Herd behavior in Malaysian capital market: An empirical analysis. Journal of Applied Economic Sciences, 4(1(7)), 45–57. Economou, F., Kostakis, A., & Philippas, N. (2010). An examination of herd behavior in four Mediterranean stock markets. European economics and finance society conference paper. Retrieved from http://www.eefs.eu/conf/Athens/Papers/511.pdf Goodfellow, C., Bohl, M., & Gebka, B. (2009). Together we invest? Individual and institutional investors’ trading behavior in Poland. International Review of Financial Analysis, 18, 212–221. Gorden, J., & Gupta, P. (2003). Portfolio flows into India: Do domestic fundamentals matter? IMF working paper. Grinblatt, M., Sheridan, T., & Wermer, R. (1995). Momentum investment strategies, portfolio performance, and herding: A study of mutual fund behavior. American Economic Review, 85, 1088–1105. Hwang, S., & Salmon, M. (2004). Market stress and herding. Journal of Empirical Finance, 11, 585–616. Khanna, S. (2002). Has India gained from capital account liberalization? Private capital flows and the Indian economy in the 1990s. IDEAs conference, 16–19. Kim, W., & Wei, S. (1999). Foreign portfolio investors before and during crisis. NBER Working paper No. 6968. Retrieved from http://www.nber.org/papers/w6968 Lakonishok, J., Shleifer, A., & Vishny, R. W. (1992). The impact of institutional trading on stock prices. Journal of Financial Economics, 32, 23–43.

Do FIIs Exhibit Herding and Positive Feedback Trading

185

Lin, A., & Swanson, P. (2008). Foreigners’ perception of US markets: Do foreigners exhibit herding tendencies? Journal of Economics and Business, 60, 179–203. Mukherjee, P., Bose, S., & Coondoo, D. (2002). Foreign institutional investment in the Indian equity market: An analysis of daily flows during January 1999–May 2002. Money & Finance, 21–47. Nofsinger, J., & Sias, R. (1999). Herding and feedback trading by institutional and individual investors. Journal of Finance, 54, 2263–2295. Shah, A., & Patnaik, I. (2006). The interplay between capital flows and the domestic Indian financial system. Technical Report. Background paper for World Bank’s ABCDE. Shu, T. (2008). Does positive feedback trading by institutions contribute to stock return momentum? AFA 2007 Chicago meetings paper. Retrieved from http://papers.ssrn.com/ sol3/papers.cfm?abstract_id¼831364 Sias, R. (2004). Institutional herding. Review of Financial Studies, 17, 165–206. Tan, L., Chiang, T., Mason, J., & Nelling, E. (2008). Herding behavior in Chinese stock markets: An examination of A and B shares. Pacific – Basin Finance Journal, 16, 61–77. Trivedi, P., & Nair, A. (2006). Determinants of FII investment inflows in India. ICFAI Journal of Applied Finance, 12, 5–20. Wermers, R. (1999). Mutual fund herding and the impact on stock prices. Journal of Finance, 54, 581–622. Zhou, R., & Lai, R. (2009). Herding and information based trading. Journal of Empirical Finance, 16, 388–393.

DO FINANCIAL CONGLOMERATES HAVE AN INCENTIVE TO PREVENT MANAGERS OF OTHER FIRMS FROM PURSUING THEIR OWN INTEREST?$ Carlos Alves and Victor Mendes STRUCTURED ABSTRACT Purpose – We develop a theoretical model to analyze the role that financial conglomerates may play in reducing agency costs in target firms. Methodology/Approach – We develop a model to analyze the activism of a financial conglomerate (that includes investment banking besides mutual fund management activities) in monitoring the managers of a listed firm. The specific problem we study is this: should the managers of a listed company undertake a new project within the firm or should they develop it outside of the firm with the help of a bank? Should or not the financial conglomerate help the managers undertake the project outside of the

$

The views stated in this chapter are those of the author and are not necessarily those of the Portuguese Securities Commission. CEF.UP and CEFAGE-UE are supported by FCT through POCTI of the QCAIII, which is financed by FEDER and Portuguese funds.

Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 187–204 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012010

187

188

CARLOS ALVES AND VICTOR MENDES

existing firm at the expenses of the investors of the mutual fund that it manages, but collecting fees from the investment banking activities? Findings – It will be attractive to both the financial conglomerate and the managers to develop the project outside of the firm if the fees charged by the financial conglomerate for the provision of investment banking services are within a certain range. However, a more intense reaction to performance from the fund investors will translate to a greater space of converging interests between the conglomerate shareholders and mutual fund investors. Additionally, if fees earned by the mutual fund company are a large source of income for the conglomerate, then the lower will be its tendency to assist the managers. Social implications – From a regulatory standpoint, the implementation of measures aimed at transferring capital between funds without cost would allow mutual fund investors to intensify their reaction to fund performance, therefore increasing the likelihood of lower agency costs. We also conclude that supervisory authorities should pay special attention to the banking relationships of firms and banks to whom the asset management component is secondary and with smaller direct stakes in the said firm. Originality/Value of paper – We develop a theoretical framework to explain the absence of activism of institutional investors integrated in financial conglomerates in the governance of listed firms. Keywords: Corporate governance; institutional investor activism; financial conglomerates Jel classifications: G10; G21; G28; G30; L20

INTRODUCTION This chapter focuses on the participation of financial conglomerates (also called universal banks) in governing corporate interests.1 It broaches issues related to conflicts of interest between the mutual funds investors and the financial conglomerate responsible for the fiduciary management of these instruments.

Do Financial Conglomerates Have an Incentive to Prevent Managers?

189

The activism of institutional investors in the governance of firms is considered a means of minimizing agency costs and safeguarding the interests of a firm’s shareholders. This impetus originates as much from the academic field (Porter, 1992; Yuan, Xiao, & Zoub, 2008) as from the codes of best practice (Cadbury Report, 1992; Hampel Report, 1998) and various international organizations operating in the regulatory and supervisory fields (IOSCO, 2003). Nevertheless, it is not guaranteed that institutional investors are motivated toward performing this role (Short & Keasey, 1997; Suto & Toshino, 2005), nor is it guaranteed that they do indeed perform it (Gillan & Starks, 2000) or are effective in performing it (Carleton, Nelson, & Weisbach, 1998; Del Guercio & Hawkins, 1999; Karpoff, Malatesta, & Walkling, 1996; Prevost & Rao, 2000; Smith, 1996; Wahal, 1996). For example, the relationship that some institutional investors have with firms in which a stake is held often focuses more on the interests of the respective managers than of the mutual fund investors (Murphy & van Nuys, 1994; Romano, 1993). There is controversy in the literature as to whether share ownership by institutional investors prevents managers from pursuing their own interests (at the expense of the shareholders) because of economies of scale and diversification (Diamond, 1984), gains from monitoring (Admati, Pfeiderer, & Zechner, 1994), relevance of resources invested (Agrawal & Knoeber, 1996; Shleifer & Vishny, 1997), market liquidity (Maug, 1998), or high monitoring costs.2 Previous evidence on the monitoring role of institutional investors shows that institutional behavior can indeed depend on the sensitivity to managerial pressure (Gordon & Pound, 1993), but there is no consensus on the effectiveness of shareholder activism. Karpoff et al. (1996) find weak evidence of improved operating performance of companies that are the target of pension funds proposals, but Smith (1996) reports that companies targeted by large pension funds increase their performance. In the United Kingdom, Faccio and Lasfer (2000) find that occupational pension funds are not effective monitors. However, the evidence presented in work by Yuan et al. (2008) supports recent regulatory efforts to promote mutual funds as a corporate governance mechanism and suggests that pooling diffuse minority interests of individual shareholders who are prone to free-rider problems via mutual funds is beneficial. In the context of a universal banking system (the scenario for this study), safeguarding the interests of a financial conglomerate’s shareholders may clash with the interests of shareholders of firms in which a stake is held (as well as with the interests of the investors of mutual funds that are fiduciary managed by the financial conglomerate). Consequently, instead of

190

CARLOS ALVES AND VICTOR MENDES

preventing managers from pursuing their own interest at the expense of the shareholders, the financial conglomerate may not oppose (and may even support) management. However, there may even be an alignment of interests between the financial conglomerate and management due to the efficiency of investor protection and the performance reaction of mutual fund investors. This can then steer the financial conglomerate to ensure adequate oversight of firms in which the mutual fund has a stake. For example, Payne, Millar, and Glezen (1996) found that the voting behavior of banks as fiduciaries differ for banks with and without business relationships. When income-related relationships exist, banks tend to vote in favor of management; when they do not exist, banks tend to vote against management. In this chapter, we develop a model to analyze the activism of a financial conglomerate in monitoring the managers of a listed firm. The specific problem we study is this: should the managers of a listed company undertake a new project within the firm or should they develop it outside of the firm with the help of a bank? For example, suppose the firm is an energy utility and the project is a new wind plant. The managers could decide to undertake this project within the existing energy firm, thus expanding the company into a new business area. Or, the managers could undertake the project themselves, as private citizens, and start a new company unrelated to the original energy firm (provided that, in general, legislation does not prevent managers from having their own business). If the project is developed inside the firm, then managers will share the project’s profits with the other shareholders of this firm. If the project is developed outside the existing company, then the managers will keep all the profits for themselves, without any payouts to the shareholders of the listed company. Under these circumstances, the cost of redirecting the funds from the listed company to the firm to be created (as well as other costs) needs to be weighted against the benefit of the full profit keeping, vis-a`-vis the profits that the managers as shareholders receive if the new project is developed inside the existing firm. Thus, managers would undertake the project outside the firm and fail to share the profits with the other shareholders if the profits they receive when they are 100% shareholders of the new firm are larger than those they would receive as (partial) shareholders of the existing firm. The financial conglomerate faces a conflict of interest as well: should it help the managers undertake the project outside of the existing firm at the expenses of its own mutual fund investors? If it helps the managers to develop the project outside the firm, the action will generate fees for the

Do Financial Conglomerates Have an Incentive to Prevent Managers?

191

conglomerate but the conglomerate will not receive the profits it would otherwise get as a firm shareholder. The financial conglomerate will also sacrifice some of the fees on the mutual fund it manages. However, if the project is undertaken by the company, then the funds with the firm’s stock in the portfolio would exhibit higher performance. Assuming, as we do in this chapter, that investors react to past fund performances,3 then the higher performance of the mutual fund would result in larger inflow of mutual fund capital and bigger fees for the conglomerate (Chevalier & Ellison, 1997). Our model also takes into account an investor protection mechanism. We assume that some profits of the project are lost if the project is developed outside the firm due to inefficient schemes used by managers to redirect funds. These costs are presumably higher in environments with better protection of investors (La Porta, Lopez-De-Silanes, Shleifer, & Vishny, 1998) because better investor protection mechanisms will make it more difficult for managers to redirect funds. Thus, the investor protection environment and the behavior of the supervisory authorities will play a decisive role in preventing conflicts of interests. Based on our model, we conclude that it will be attractive to both the financial conglomerate and the managers to develop the project outside of the firm if the fees charged by the financial conglomerate for the provision of banking services are within a certain range. However, if the conglomerate has a strong asset management activity (i.e., fees earned by the mutual fund company are a large source of income for the conglomerate), then the lower will be its tendency to assist the managers. Furthermore, the other firm’s shareholders and the supervisory authorities should pay special attention to the income-related banking relationships between financial conglomerates that have reduced asset management activity and companies in which a small direct stake is held. These relationships also depend on the intensity of fund investors’ performance reaction and on the weight of the asset in the fund’s portfolio. The topic of investigation of this chapter is therefore well suited with the research theme of the present volume of the International Finance Review. In fact, institutional investors do play a vital role in the governance of companies, all the more so in a global capital market where there are diffuse minority interests of individual shareholders that are not able or lack the expertise to monitor the managers of listed firms. The chapter is structured as follows. The model is presented in the following section. We then deduce the equilibrium conditions, and conclude the chapter with some final remarks.

192

CARLOS ALVES AND VICTOR MENDES

THE MODEL The Model Architecture Assumptions and Sequence of Events We assume an economy with risk-neutral economic agents. There are multiple (N) financial assets and multiple intermediaries that fiduciary manage capital. A financial conglomerate headed by a bank (B – the ‘‘bank’’) holds a 100% stake in a mutual fund company (MC – the ‘‘mutual fund company’’)4 that manages one mutual fund (F – the ‘‘fund’’). Firm J is a listed company (the ‘‘company’’) with a standardized quantity of 1 share being issued. These shares are held by J’s managers – the ‘‘managers’’ – (qA), by B (qB), and by fund F (qF). The remaining shares (1 – qA – qB – qF) are scattered among other investors (see the model architecture in Fig. 1). The complete series of events is summarized in Fig. 2. At the start of each investment cycle (date 0), the mutual fund company allocates the total value (V0) of the funds under management among the N available assets. In particular, it decides the weight (w) of asset J. This decision is not influenced by the bank that the fund uses a buy and hold strategy, which means that we do not need to account for transaction costs, and that the fund holds the

Fig. 1.

Architecture of the Model.

Do Financial Conglomerates Have an Incentive to Prevent Managers?

Fig. 2.

193

Sequence of Events.

same number of the shares of J at the end of the period. Lastly, we assume that both the mutual fund company and the bank seek to maximize the consolidated value of their businesses. On date 1, the managers detect a business opportunity (the ‘‘project’’) that would bring about a payout of p (p W0) on date 2. As an alternative, these managers may decide to develop the project outside of the company using the help of bank B (the ‘‘deal’’). In general, legislation does not prevent managers from having their own firm. If the managers decide to undertake the project outside the company, then they will not share the profits of the project with other shareholders. The deal may include financing, the implementation of one or more financial operations, entering into a partnership in an offshore business deal, or a mix of any of the usual services the bank provides in its day-to-day activity, none of which are at risk of infringing legislation.5 The managers chose B for reasons of trust. The bank, on the other hand, knows that if it refuses the deal, the managers will not look for another bank to carry out the deal. In fact, as soon as the managers contact the bank, there is no longer asymmetry of information between the managers and the bank. Thus, the bank will not be willing to take a loss (as a shareholder of the company) without receiving a bank-related asset in return. Therefore, there is no danger of suffering the consequences of the deal as a shareholder and fund manager and not benefiting from it as a supplier of banking services. The opposite is not plausible; the bank would oppose managers to undertake the investment outside the company because it possesses information regarding the managers’ plans and has a direct and indirect interest in the company. The bank has to decide whether to fully provide the services requested. It is possible that the bank may only support partial solutions, that is, solutions that lead to a certain part of the project’s payout being received by J’s managers; the remaining payout would be received by the company

194

CARLOS ALVES AND VICTOR MENDES

itself, thereby influencing its earnings (and dividends). In other words, the managers may only receive a certain proportion d(0rdr1) of the project’s profits depending on the bank’s course of action. Thus, dp monetary units are distributed to the managers, while the remaining (1d)p monetary units are received by J’s shareholders. If the bank supports solutions with d ¼ 1, then it fully supports managers to redirect funds. If the best solution for the bank is such that d ¼ 0, then it will oppose the managers. Finally, if the best solution for the bank is such that 0odo1, then it will exhibit a Salomonic behavior; it will partially defend the interests of J’s shareholders, but will also leave some space for the managers to extract private benefits.6 However, the managers will not collect all the diverted dp profits; they will only receive ZdpU Since 0oZo1, this implies that if the project is undertaken outside the company, then some profits of the project are lost due to inefficient managerial schemes when redirecting funds. Besides, developing the project outside of the company may not be as efficient as developing it inside the firm. The value of Z could be influenced by the activity, project, and quality of investor protection mechanisms that characterize the operating environment of the firm (La Porta et al., 1998). Presumably, better investor protection mechanisms will make it more difficult (and therefore more inefficient) for managers to redirect funds. Consequently, Z will be lower. However, we assume here that Z is constant, given that we have only one project, one regulatory environment, and one firm (J). On the other hand, the bank B needs to be paid for the services rendered; l represents the payment made to B for each monetary unit received after the deduction of additional costs. Thus, the payout p of the project is distributed as such: – (1 d)p is delivered to J’s shareholders as dividends; – (1 Z)dp are lost funds (i.e., additional costs) due to the inefficient schemes used by managers to redirect funds; – lZdp are the fees received by the bank for services provided; – (1 l)Zdp is received by the managers. The product (1 l)Zdp is the so-called ‘‘personal benefit obtained by the managers’’ or, if the managers are also larger shareholders, the ‘‘personal control benefit.’’ The total agency costs is dp. On date 2, the company’s financial statements are disclosed. The payment of an extraordinary dividend will or will not be made, depending on the course of action decided on at date 1. Likewise, the distribution of the project’s profits will be implemented at this time, depending again on the course of action decided on at date 1. In addition, the bank will be paid on

Do Financial Conglomerates Have an Incentive to Prevent Managers?

195

this date for the services provided. Lastly, the returns of the fund are disclosed, the mutual fund company receives its fees, and a new decision regarding the distribution of the funds for the following cycle is made by investors. In the following paragraphs we discuss the behavior and performance of the different investors, managers, and the subsequent impact on the equilibrium of the model. The Behavior of the Mutual Fund Investors On date 2, the financial results of firms and funds are disclosed, and investors decide on which funds to invest their savings. We assume that mutual fund investors allocate their savings to the best-performing funds, that is, investors react to past fund performances (Christoffersen, 2001; Goetzmann & Peles, 1997; Ippolito, 1992; Sirri & Tufano, 1998). Therefore, we use the rule V tþ2 ¼ V t ð1 þ g1 þ g2 atþ2 þ tþ2 Þ

with t ¼ 0; 2; 4; . . . ; 1;

(1)

where Vt represents the amount invested in fund F for the cycle between dates t and t þ 2, g1 (g1W0) is a growth rate that does not depend on the fund’s performance, g2 (g2W0 because investors react to performance) is a performance bonus coefficient, at þ 2 represents the fund’s performance in the previous cycle (between dates t and t þ 2), and e is a random disturbance (i.i.d.). Thus, the amount invested in the fund is a random variable that depends on the previous net assets under the management of the fund, a constant growth rate, the performance of the fund, the investors’ reaction to fund performance, and a random error term. The Performance of the Mutual Fund On date 2, the stock of the company produces an abnormal payout of (1 d)p/P0 (where P0 is the standardized market price of the stock on date 0). Using P0 as the purchase price, the marginal contribution of this payout to the performance of the fund is w(1  d)p/P0. This means that the deal has an impact on the performance of the fund. If the bank refuses to provide the services (d ¼ 0), then it maximizes the fund’s performance; if it fully supports management (d ¼ 1), then the project’s contribution to the performance of the fund is zero. Thus, the opportunity cost borne by the fund is Da2 ¼

wdp P0

(2)

196

CARLOS ALVES AND VICTOR MENDES

The Behavior of J’s Managers If the managers decide to develop the project outside of the company, then they contact the bank to determine what technical solutions this can offer and to negotiate the value of the fees (l). The managers bear (as shareholders of the company) a marginal cost equivalent to the percentage of the stock held (qA) for each monetary unit. Thus, the full dp distributed to the managers implies a marginal cost of qAdp. On the other hand, the managers’ marginal income is (1  l)Z per monetary unit and the total benefit is (1  l)Zdp. As a result, the payout for the managers (RA) is given by Eq. (3) (i.e., the difference between total benefits and total costs),7 and it is only advantageous to implement the deal with the bank when the payout is positive (RAW0): RA ¼ ð1  lÞZdp  qA dp

(3)

The Behavior of the Bank After analyzing the deal, the bank proposes a technical solution. The bank puts forward d (the proportion of the project that is financed) and negotiates the fees for those services. To this end, the bank has to consider the effect that its decisions will have on its assets, either directly (through the commissions it receives) or indirectly (through the inherent effect of its shareholding in the capital of J and of the mutual fund company). If the bank decides to support management, then it will have a direct marginal benefit of lZdp (the fees received for the services provided) and a marginal cost of qBdp (since it is a shareholder of J).8 Additionally, the fund has a marginal loss of abnormal income given by Eq. (2). This loss affects the fund’s performance and, consequently, the quantity of funds entrusted to the mutual fund company for management at the start of the next (and subsequent) investment cycles. It will also decrease the mutual fund company’s expected future management fees and profits. This cost includes the effect that the dp monetary units cause on the value of (i) the future capital inflows and (ii) the mutual fund company. If the bank provides the service and supports management (that is, 0odr1), then, on date 1, the expected value for V2 is:      a  wpd a  wpd ¼ V 0 1 þ g 1 þ g2 E V 2 j Eða2 Þ ¼ P0 P0

(4)

Do Financial Conglomerates Have an Incentive to Prevent Managers?

197

If the bank decides not to offer the service and opposes management (i.e., d ¼ 0), then E(a2) ¼ a and E½V 2 jEða2 Þ ¼ a ¼ V 0 ð1 þ g1 þ g2 aÞ

(5)

Subtracting Eqs. (4) from (5) gives the expected variation of V2 resulting from the deal: E[DV2|E(a2) ¼ a wpd/P0] ¼ V0g2wpd/P0. Given Eq. (1), the deal has an impact on V2, V4, V6, V8,y, VN. Thus EðDV tþ2 jDV 2 Þ ¼ ðV 0 g2 wpd=P0 Þð1 þ g1 þ g2 aÞt=2 ;

with t ¼ 2; 4; 6; . . . ; 1 (6)

Equation (6) gives the expected impact on the values managed by the mutual fund company. Multiplying this by the profit achieved per each managed monetary unit (f) results in the expected variation in the bank’s dividends (D) at the end of each management cycle. Hence,   a  wpd V 0 g2 wpdf ¼ E DD4 j Eða2 Þ ¼ ¼ E½DV 2 f P0 P0 EðDDtþ4 jDV 2 Þ ¼

  V 0 g2 wpdf ð1 þ g1 þ g2 aÞt=2 ; P0

(7)

with t ¼ 2; 4; 6; . . . ; 1 (8)

Combining Eqs. (7) and (8) gives a perpetuity with an initial value of V0g2wpdf/P0 (the first cash flow is paid on date 4) and a periodic growth rate of g1 þ g2aU The Gordon method of discounting dividends then yields 

 a  wpd V 0 g2 wpdf E DVMC j Eða2 Þ ¼ ¼ P0 P0 ðR  g1  g2 aÞ

(9)

where R (Wg1 þ g2a) represents the appropriate discount rate9 and DVMC is the change in the value of the mutual fund company. We can now compute the bank’s result (RB) by subtracting the loss incurred as a shareholder and the loss of the value of the mutual fund company (which is 100% owned by the bank) from the fees received: RB ¼ lZpd  qB pd 

V 0 g2 wpdf P0 ðR  g1  g2 aÞ

(10)

198

CARLOS ALVES AND VICTOR MENDES

Equilibrium Conditions As seen in Eq. (3), the managers will accept the deal with the bank only if RA W 0, that is, if lo(ZqA)/ZU On the other hand, from Eq. (10) we conclude that the bank will only accept the deal if lWqB/Z þ V0g2wf/ [ZP0(Rg1g2a)]. Juggling the conditions so that the deal is advantageous for both the bank and the managers, we conclude that Proposition 1. The existence of the deal depends upon the structure of ownership of the company. In fact, the deal is only mutually advantageous provided that l(0olo1) satisfies the condition: qB =Z þ V 0 g2 wf=½ZP0 ðR  g1  g2 aÞoloðZ  qA Þ=Z

(11)

The lower (upper) threshold corresponds to the price that leads to a null result for the bank (the managers). Hence, both the bank and the managers will receive a positive profit provided that the bank fees is between these bounds. Therefore, a mutually beneficial deal depends on the structure of ownership of the listed company (i.e., it depends on qA and qB, the shares owned by managers and the bank, respectively). However, all else being equal, a higher qA means that J’s managers face a greater opportunity in deals that divert funds from the firm’s operational accounts, given that, being shareholders, they also bear the effects of such an action. Similarly, a higher stake held by the bank means a larger lower threshold and, therefore, a smaller range (or equilibrium space) for a mutually beneficial deal. These results are consistent with empirical evidence (e.g., Chaganti & Damanpur, 1991; McConnell & Servaes, 1990; Morck, Shleifer, & Vishny, 1988). The result put forth in Proposition 1 is independent of the percentage of the services provided by the bank (d). In fact, one may state that Proposition 2. If there is a deal, then agency costs will be maximized. Indeed, for each l, the result for the bank (RB) is a linear function of d with a positive slope when RBW0. Thus, d ¼ 1 maximizes RB, that is, if there is a deal then the bank fully supports management. Using Proposition 1, we can see that the lower threshold for the equilibrium space (and therefore a smaller range for a profitable deal) is higher when the performance of the fund is high. Thus, if the bank is able to attract investment inflows that react to performance then there is a smaller equilibrium space for the deal (i.e., for the development of projects outside

Do Financial Conglomerates Have an Incentive to Prevent Managers?

199

the company). In other words, the bigger the skill of the asset management component of the bank’s business, the lower the conflicts of interest between the bank shareholders and the mutual fund investors. There is actually a threshold for a (the fund’s performance) that prevents the deal from taking place at all and allows the full convergence of interests from both the bank shareholders and mutual fund investors. This threshold is obtained by equating the upper and lower thresholds in Eq. (11) and solving for a. The following can then be stated: Proposition 3. The interests of the bank shareholders and of the fund investors are more (less) aligned when the asset management business is a more (less) prominent business area for the bank. In fact, if aZ(R  g1)/g2 –V0wf/[P0(Z  qA  qB)], then no deal is possible and the bank will oppose the managers. This threshold includes two elements. First, it depends on the cost of capital (R) and the parameters that define the reaction of investors to performance (g1 and g2). Second, it depends on the profit rate of the asset management business (f), the efficiency of the supervisory mechanisms (Z), and the structure of ownership of the company (qA, qB, qF).10 In particular, the higher (lower) the number of shares held by fund F (V0w/P0), the lower (higher) the performance of the fund that is necessary to eliminate the agency costs. Besides, the higher (lower) the fund size (V0), the lower (higher) the performance needed to eliminate the agency costs. In a universal banking system, there might be banks that rely less on the asset management area (lower alphas).11 If this were the case, then there is a greater probability for conflicts of interests between the fund investors and the bank shareholders. However, this may be compensated if the bank has a relevant stake in firm J.12 In other words, the interests of the bank shareholders and of the fund investors are mostly aligned when the bank relies heavily on the asset management area. This means one would expect a more effective company involvement from banks that have strong interests in the asset management business. However, if the bank relies less heavily on asset management but has a relevant stake in a firm, then there might be a higher alignment of the interests of fund investors and bank shareholders. Three corollaries can be drawn from this discussion. From Proposition 3, in order to eliminate agency costs, the performance of the fund needs to be lower (higher) to compensate for higher (lower) investor reaction. In other words, an intense investor reaction implies a greater space of convergence between the interests of the bank shareholders and fund investors. Hence,

200

CARLOS ALVES AND VICTOR MENDES

the financial conglomerate will most likely ensure adequate oversight of firms in which the mutual fund has a stake. Thus, Corollary 1. The more intensely the fund investors react, then the greater is the converging space of the interests between the bank shareholders and the fund investors. This result has important policy implications from a regulatory standpoint. In fact, the implementation of measures aimed at transferring capital between funds without cost would allow mutual fund investors to intensify their reaction to fund performance, therefore increasing the likelihood of lower agency costs. It follows immediately from Corollary 1 that if mutual fund investors do not react to the performance exhibited by the fund, then the interests of the bank shareholders and those of the fund investors are less aligned. Under these circumstances, the conglomerate will have a lower incentive to oppose management and will most likely not ensure adequate oversight of the firm. This is put forth in Corollary 2: Corollary 2. The absence of performance reaction minimizes the incentive to oppose management. Finally, Proposition 3 implies that the higher (lower) monitoring efficiency of the supervisory authorities, then the lower (higher) fund performance that is necessary to oppose management. Or, Corollary 3. The existence of strong investor protection mechanisms decreases the alpha threshold necessary to oppose management. Thus, considering Proposition 3 and Corollary 3, we can conclude that the efforts of the supervisory authorities to protect investors should be directed toward the business relationships between banks (with reduced asset management interests) and companies in which a smaller stake is held, arguably cases in which agency costs are more likely to occur. By doing so, they would help ensure a greater convergence of interests between firm and bank shareholders in situations where this convergence is less likely to occur.

CONCLUDING REMARKS Institutional investors who actively control and monitor the governing of firms in which they have a stake is seen by professional and academic circles

Do Financial Conglomerates Have an Incentive to Prevent Managers?

201

as a possible means of reducing agency costs. However, this possibility is treated skeptically by some who question the motivation, availability, and interest on the part of the institutional investors to play such a role. In this chapter, we develop a model to analyze the activism of a financial conglomerate in monitoring the managers of a listed company. We conclude that if the fees charged by a bank are within a certain range, then undertaking a project outside a company will be attractive to both the bank and the company’s managers, thereby maximizing agency costs. Thus, financial conglomerates will not oppose managers from pursuing their own interests at the expense of the shareholders. However, if the bank has a high capacity to obtain capital inflows that react to mutual fund performance, then there will be larger space for converging interests between the bank shareholders and fund investors. This leads us to the conclusion that, all else being equal, opposition to bad governance is most often expected from the bank with the bigger asset management business. However, holding stake in a firm also has a decisive role. For example, if one bank concentrates more on asset management while the other has a higher stake in the firm, then one can not say beforehand which bank will show a greater probability of ensuring adequate oversight of the firm’s managers. We also conclude that an increase in efficient monitoring by firm shareholders and supervisory authorities results in the decrease of the critical performance threshold (alpha) that prevents the existence of agency costs. Therefore, firm shareholders and supervisory authorities should pay special attention to the banking relationships of firms and banks where asset management is secondary (i.e., fees earned by the mutual fund company represent a small source of income for the conglomerate) and that have smaller direct stakes in the firm.

NOTES 1. Financial conglomerates or universal banks are institutions that can perform any banking activity as well as other financial and securities business (e.g., mutual fund management), within the legal entity of the bank or via a separate subsidiary. In opposition, nonuniversal (i.e., specialized) banks are banks that only carry out some specialized banking or financial activities. 2. Holderness (2003) provides an excellent survey of blockholders and corporate control, and concludes that small shareholders and regulators have little reason to fear large shareholders in general, especially when a large shareholder is active in firm management.

202

CARLOS ALVES AND VICTOR MENDES

3. The performance reaction assumption is consistent with the existing literature (Christoffersen, 2001; Goetzmann & Peles, 1997; Ippolito, 1992; Sirri & Tufano, 1998). 4. This is the prominent mutual fund management model in continental Europe. In general, European countries do not have any constraints to the ownership structure of the collective investment schemes (CIS) operators (‘‘the legal entity that has overall responsibility for the management and performance of the functions of the CIS,’’ in the words of IOSCO), and in general mutual fund companies are 100% owned by banks. Furthermore, funds (the CIS) are not legal entities, and the mutual fund company is the legal entity responsible for the management of the mutual fund. 5. Suppose, for example, that the manager remuneration is a function of the company’s earnings per share (EPS). The acquisition of another firm (the payment being in cash) increases J’s EPS and the managers’ remuneration (and thus is favorable to the interests of the managers). Suppose that the acquisition does not bring any synergies. If the acquisition is successful, then it will bring a loss of value for J’s shareholders (Moeller, Schlingemann, & Stulz, 2005) who would be better off if cash were distributed as dividends. B would have the role of an investment bank, assisting the acquisition (and receiving the fees). 6. In a sense, B is an arbitrator. In the example provided in Footnote no. 5, suppose that B supports management but only up to a certain point. That is, the bank supports it but only if the price paid is below a certain level. If the price paid is above this level, then the bank opposes management. Another example is the following. Suppose that Jus managers need financial support from B to develop the project outside the firm. B can loan out a part (but not all) of the requested funds. Thus, the project will be partially developed outside the firm and partially within the firm. 7. We assume that the managers’ salaries do not depend on the firm’s performance. If this is the case, then it should be added to the opportunity cost. However, profit sharing has the same effect as the ownership of stock, for which reason this possibility was not explicitly considered so as not to overload the model. 8. We assume that if the bank refuses the deal, then it will not lose a future client. This assumption permits any future costs regarding a client’s losses to be ignored. If these costs exist, then one additional parameter should be added to the model; the conclusions, however, would be unaltered. 9. We assume that the period of effective capitalization is equal to one investment cycle. 10. V0w/P0 is the number of shares of J held by F (qF). 11. For instance, these banks may not hire people with very high skills in portfolio management. 12. The higher (lower) qB, the lower (higher) the alpha threshold will be.

REFERENCES Admati, A. R., Pfeiderer, P., & Zechner, J. (1994). Large shareholder activism, risk sharing, and financial market equilibrium. Journal of Political Economy, 102, 1097–1130.

Do Financial Conglomerates Have an Incentive to Prevent Managers?

203

Agrawal, A., & Knoeber, C. R. (1996). Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, 31, 377–397. Cadbury Report. (1992). Report of the committee on the financial aspects of corporate governance. London: Gee. Carleton, W., Nelson, J., & Weisbach, M. (1998). The influence of institutions on corporate governance trough private negotiations: Evidence from TIAA-CREF. Journal of Finance, 53, 1335–1362. Chaganti, R., & Damanpur, F. (1991). Institutional ownership, capital structure and firm performance. Strategic Management Journal, 12, 479–491. Chevalier, J., & Ellison, G. (1997). Risk making by mutual funds as a response to incentives. Journal of Political Economy, 105, 1167–1201. Christoffersen, S. (2001). Why do money fund managers voluntarily waive their fees?. Journal of Finance, 56, 1117–1140. Del Guercio, D., & Hawkins, J. (1999). The motivation and impact of pension fund activism. Journal of Financial Economics, 52, 293–340. Diamond, D. W. (1984). Financial intermediation and delegated monitoring. Review of Economic Studies, 51, 393–414. Faccio, M., & Lasfer, M. (2000). Do occupational pension funds monitor companies in which they hold large stakes?. Journal of Corporate Finance, 6, 71–110. Gillan, S., & Starks, T. (2000). Corporate governance proposals and shareholder activism: The role of institutional investors. Journal of Financial Economics, 57, 275–305. Goetzmann, W., & Peles, N. (1997). Cognitive dissonance and mutual fund investors. Journal of Financial Research, 20, 145–158. Gordon, L. A., & Pound, J. (1993). Information, ownership structure, and shareholder voting: Evidence from shareholder-sponsored corporate governance proposals. Journal of Finance, 48, 697–718. Hampel Report. (1998). Committee on corporate governance: Final report. Retrieved from www.cnmv.es/index.htm. Holderness, C. G. (2003). A survey of blockholders and corporate control. Federal Reserve Bank of New York Economic Policy Review, April, pp. 51–64. IOSCO. (2003). Collective investment schemes as shareholders: Responsibilities and disclosure. Retrieved from www.iosco.org/library/index.cfm?whereami ¼ pubdocs Ippolito, R. (1992). Consumer reaction to measures of poor quality: Evidence from the mutual fund industry. Journal of Law and Economics, 35, 45–70. Karpoff, J., Malatesta, P., & Walkling, R. (1996). Corporate governance and shareholders initiatives: Empirical evidence. Journal of Financial Economics, 42, 365–395. La Porta, R., Lopez-De-Silanes, F., Shleifer, A., & Vishny, R. (1998). Law and finance. Journal of Political Economy, 106, 1113–1155. Maug, E. (1998). Large shareholders as monitors: Is there a trade-off between liquidity and control?. Journal of Finance, 53, 65–98. McConnell, J., & Servaes, H. (1990). Additional evidence on equity ownership and corporate value. Journal of Financial Economics, 27, 595–612. Moeller, S., Schlingemann, F., & Stulz, R. (2005). Wealth destruction on a massive scale? A study of acquiring-firm returns in the recent merger wave. Journal of Finance, 60, 757–782.

204

CARLOS ALVES AND VICTOR MENDES

Morck, R., Shleifer, A., & Vishny, R. (1988). Management ownership and market valuation. Journal of Financial Economics, 20, 293–315. Murphy, K., & van Nuys, K. (1994). Governance, behavior, and performance or state and corporate pension plans. Unpublished manuscript. Payne, T. H., Millar, J. A., & Glezen, G. W. (1996). Fiduciary responsibility and bank-firm relationships: An analysis of shareholder voting by banks. Journal of Corporate Finance, 3, 75–87. Porter, M. (1992). Capital disadvantage: America’s failing capital investment system. Harvard Business Review, 70, 65–82. Prevost, A., & Rao, R. (2000). Of what value are shareholder proposals sponsored by public pension funds?. Journal of Business, 73, 177–204. Romano, R. (1993). Public pension fund activism in corporate governance reconsidered. Columbia Law Review, 93, 795–853. Short, H., & Keasey, K. (1997). Institutional shareholders and corporate governance in the United Kingdom. In: K. Hopt, H. Kanda, M. Roe, E. Wymeersch & S. Prigge (Eds.), Comparative corporate governance – The state of the art and emerging research (pp. 18–53). New York, NY: Oxford University Press. Sirri, E., & Tufano, P. (1998). Costly search and mutual fund flows. Journal of Finance, 53, 1589–1622. Shleifer, A., & Vishny, R. (1997). A survey of corporate governance. Journal of Finance, 52, 737–783. Smith, M. (1996). Shareholder activism by institutional investors: Evidence from CalPERS. Journal of Finance, 51, 227–252. Suto, M., & Toshino, M. (2005). Behavioural biases of Japanese institutional investors: Fund management and corporate governance. Corporate Governance: An International Review, 13, 466–477. Wahal, S. (1996). Pension fund activism and firm performance. Journal of Financial and Quantitative Analysis, 31, 1–23. Yuan, R., Xiao, J., & Zoub, H. (2008). Mutual funds’ ownership and firm performance: Evidence from China. Journal of Banking and Finance, 32, 1552–1565.

THE IMPACT OF FOREIGN GOVERNMENT INVESTMENTS: SOVEREIGN WEALTH FUND INVESTMENTS IN THE UNITED STATES Elvira Sojli and Wing Wah Tham STRUCTURED ABSTRACT Purpose – Study the role of sovereign wealth funds (SWFs) as an example of foreign and politically connected large shareholders, and their impact on firm value. Methodology/approach – Use a sample of SWF large U.S. investments where SWFs intend to actively engage with management to analyze not only whether but also why SWF investments outperform the market in both the short- and long term from the perspective of internationalization, political connections, and corporate governance. Findings – Foreign and politically connected large investors, like SWFs, improve firm value through the provision of SWF domestic market access and government-related contracts. In the short run, the market welcomes SWF investments in expectation of potential monitoring and internationalization benefits. In the long run, the target firms’ degree of Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 207–243 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012011

207

208

ELVIRA SOJLI AND WING WAH THAM

internationalization and Tobin’s q increase substantially after SWF investments. The increase in q is directly related to the number of government-related contracts granted by SWF countries. Social implications – SWF investment benefits appear to outweigh the costs for firm value and shareholders. The results point to the benefits of large and foreign investors for shareholders. Originality/value of paper – This is the first work to provide evidence on how foreign government-related shareholders can affect firm value. Keywords: Foreign political connections; government-related deals; internationalization; large shareholders; sovereign wealth Funds JEL classifications: F23; G15; G23; G32; G34

INTRODUCTION Institutional investors have been the focus of much research in the last decade, given their increasing portfolio size and the role that they play in the global financial markets as large shareholders. A new class of institutional investors, sovereign wealth funds (SWFs), has recently received increasing media, political, and corporate attention. SWFs are different from other institutional investors (i.e., mutual, pension, and hedge funds), because they are not only large, but also foreign and politically connected. International Financial Services London (2009) describes SWFs as: independent, increasingly active, and having higher risk tolerance and longer investment horizons than other institutional investors. In addition, SWFs are likely to continue to have a large and important impact as institutional players in financial markets with their increasing assets under management, future potential for growth, and desire to diversify into more mature global markets. Despite SWFs’ importance and unique characteristics, we know very little about the effects such investors have as shareholders in foreign financial markets. In this chapter, we study the role of SWFs as an example of foreign and politically connected large shareholders, and their impact on firm value. The role of SWFs as investors in global capital markets is unclear, as they can provide benefits as well as impose costs on the companies they invest in. On the one hand, SWFs as foreign investors can be considered as providers

SWF Investments in the United States

209

of foreign direct investments (FDI). Theories of corporate multinationalism and internationalization suggest that FDI from SWF countries might increase the value of a firm through intangible assets like providing access to previously inaccessible markets, patents, or consumer goodwill (e.g., Go¨rg & Strobl, 2001; Lipsey, 2002). On the other hand, SWFs can use their investments to extract technological know-how and close the gap with the developed countries, in terms of comparative advantage. In addition, SWFs have political connections to the governments of the countries where they originate from, i.e. they represent a foreign government connection for the target firm. Empirical evidence shows that firms politically connected to domestic politicians generate higher market returns than their peers, as they benefit from preferential treatment (e.g., Bunkanwanicha & Wiwattanakantang, 2009; Fisman, 2001; Faccio, 2006; Faccio & Parsley, 2009; Goldman, Rocholl, & So, 2009a, 2009b). Firms with foreign government connections might also benefit from such treatment. However, foreign government-related shareholders can force the firm to take suboptimal investment decisions that benefit the economy of the investing countries.1 Finally, as large shareholders, SWFs might improve corporate governance and add value to firms through monitoring (Shleifer & Vishny, 1986). However, they might also extract rents from firms and enjoy the private benefits of control (Shleifer & Vishny, 1997; Burkart, Gromb, & Panunzi, 2000). Using a sample of SWF large U.S. investments where SWFs intend to actively engage with management, we analyze not only whether but also why SWF investments outperform the market in both the short- and long term from the perspective of internationalization, political connections, and corporate governance. We find that the market reacts positively to SWF investments in the short-run in the expectation of increased corporate monitoring and future increases in business internationalization. This finding provides evidence that independent foreign institutions like SWFs, which do not have an incentive to be loyal to the management (i.e., ‘‘pressure resistant’’), have the potential to enhance firm value through both direct and indirect monitoring, consistent with the findings of Ferreira and Matos (2008). This result also suggests that the possibility of future market access in the investors’ countries is important to market participants in evaluating SWF investments. However, it is the government affiliation of the investors that plays an important role in increasing firm value in the long run. The SWF target firms generally outperform hedge fund target firms and catch up with matched firms in terms of Tobin’s q two years after the investment. The increase in q is directly related to the provision of

210

ELVIRA SOJLI AND WING WAH THAM

government-related contracts. This result constitutes first hand evidence on the effects of foreign political connections on firm value and on one of the mechanisms via which political connections can affect firm value. Shareholder benefits ensuing from foreign political connections appear to outweigh the costs of private benefit extraction from SWF investors. However, we find a decrease in the technological and competitive gap between the United States and the SWF countries that provide the government-related contracts. Overall, SWFs provide clear benefits for shareholders, but their impact on the economy of the recipient country, the United States, remains unclear. We use Securities Exchange Commission (SEC) Schedule 13 filings of ownership that are larger than 5% to identify SWF targets from 1997 to 2008. Using these SWF block investments ensures that they have high enough stakes in a company to have incentives to engage with the management. In the short-run, SWF investments are perceived positively by the market, generating an average abnormal return of 12.5% during the event window. The average short-term reaction is comparable to that of hedge fund investments (Brav, Jiang, Partnoy, & Thomas, 2008; Klein & Zur, 2009) and much higher than that of other institutional investors. We find that the cross-sectional variation of abnormal returns is related to the market expectation for increased monitoring and for benefits from increased internationalization. Our findings support the literature on effective monitoring by institutional investors that are not bound by reporting regulations, like hedge funds and private equity firms, in comparison to other heavily regulated institutional investors such, as mutual and pension funds (i.e., Romano, 2001; Gillan & Starks, 2007; Cronqvist & Fahlenbrach, 2009, among others). As one of the world largest institutional investors and with high stakes in companies, SWFs often have specialized teams to assert their rights and interests (International Financial Services London, 2009). Furthermore, in our sample, they often hold long-term concentrated positions, which promotes and extends their effectiveness as informed monitors, consistent with Chen, Harford, and Li (2007). As foreign investors, SWFs can provide investment avenues and expertise in markets where target firms do not have experience, and thus generate value via foreign market penetration. Our results on internationalization support Gozzi, Levine, and Schmukler (2008), who provide evidence that newly internationalized firms experience an increase in market value in the year before and the year of internationalization.

SWF Investments in the United States

211

The positive abnormal returns are not transient and continue throughout the holding period, where an average buy-and-hold abnormal return of 16% per annum is generated for the shareholders. The levels of Tobin’s q and degree of internationalization (DOI) increase substantially in the post-SWF investment period. In the long run, we find a statistically significant 50% ex post increase in the DOI of SWF target firms and a 31% increase in their Tobin’s q. Through a detailed analysis of strings of corporate events involving the target firms, we find that these firms experience an increase in foreign sales and government-related contracts after SWF investments. The target firms’ q’s increase significantly with the provision of governmentrelated contracts. Furthermore, using an event study, we find that the stock price of firms that are given SWF government-related contracts increases on average by 3% above the market portfolio upon the announcement of the contract. Our results provide first hand evidence on how government-related shareholders can affect firm value. Although the literature has only focused on whether political connections affect market value in the short-run (e.g. Fisman, 2001; Faccio, 2006; Faccio & Parsley, 2009; Goldman et al., 2009a), we find that government-related contracts are one of the mechanisms via which government connectedness can result in firm value increases in the long run. Furthermore, we are the first chapter to provide evidence that not only domestic but also foreign government-related entities can create longterm value for the firm and the shareholders. Finally, we investigate and provide evidence on the underlying motives for foreign government involvement and procurement provision from economic theories. The findings also add to the literature on multinationalism that shows positive spillover effects from the investments of companies from developed to less developed countries (e.g. Doukas & Travlos, 1988; Errunza & Senbet, 1981, 1984; Fatemi, 1984; Morck & Yeung, 1991, 2001). Investments from less developed countries, like the SWF ones, also increase firm value through positive spillovers of knowledge, productivity gains between firms, and encouraging foreign corporate expansions. Recent work finds that, generally, SWFs invest minority stakes in small, nonlisted, mainly domestic companies (Balding, 2008; Bernstein, Lerner, & Schoar, 2009; Bortolotti, Fotak, Megginson, & Miracky, 2009), and they diversify across asset classes and geographic regions, but without taking excessive risk (Balding, 2008). Also, SWFs are more likely to invest at home when local politicians are involved than when foreign managers are in control (Bernstein et al., 2009). Several studies conclude that SWFs behave more like mutual funds (Beck & Fidora, 2008; Caner & Grennes, 2008) or

212

ELVIRA SOJLI AND WING WAH THAM

hedge funds (Balding, 2008; Kotter & Lel, 2011). In the short run, the market reacts positively to SWF investments (e.g., Dewenter, Han, & Malatesta, 2010) and negatively to SWF divestments (Kotter & Lel, 2011; Dewenter et al., 2010). The results in the long-run are mixed. Bortolotti et al. (2009) find a negative two year matched firm adjusted return, and firms where SWFs have higher stakes yield significantly more negative returns, whereas returns increase with better SWF governance. However, Fernandes (2009) finds that firms that have higher SWF ownership have higher valuations (15–20%) and better operating performance. Dewenter et al. (2010) investigate the effect of SWF investment on firm value using data for small and large SWF investments in home and foreign countries, mainly in Asia. They find that often SWFs are active investors and their domestic target firms are subject to favorable government decisions, consistent with previous literature (e.g., Fisman, 2001; Faccio, 2006; Faccio & Parsley, 2009; Goldman et al., 2009a). Differently from these papers, we focus on the impact of internationalization and foreign government connections on firm value in developed markets which is a novel issue that has not been investigated before. We use SWFs as an example of investors with these characteristics. In addition, we study a sample of U.S. listed firms where SWFs are large investors (hold more than 5% of a firm’s outstanding shares) and actively engage with management. Previous papers have focused on mainly small or portfolio related investments. It is difficult to address issues related to performance influence when the SWF owns 0.5% stake in a firm. In addition, it is difficult to investigate investment intentions when the sample includes portfolio investments that are only held to collect dividends. Using data of large and active investments allows us to clearly address the impact of internationalization, foreign political connections, and monitoring on firm value.2 In related a study of government-controlled acquirers and mergers and acquisitions (M&A), Karolyi and Liao (2009) find that the share-price reactions to the announcements of such acquisitions are not different from other M&A announcements. However, they find important differences between SWFs and other government-controlled acquirers. SWF-led acquisitions are less likely to fail, they are more likely to pursue targets that are larger in total assets and with fewer financial constraints. Furthermore, they find that the market reactions to SWF-led acquisitions, while positive, are statistically and economically much smaller than those of other government-controlled acquirers. The chapter proceeds as follows. The next section presents the data. The second section describes the characteristics of the SWF investments. The

SWF Investments in the United States

213

third section discusses the short- and long-term results and their implications. The fourth section concludes the chapter.

DATA DESCRIPTION Throughout this chapter, we use the term SWFs to refer to both sovereign wealth funds and sovereign-owned enterprizes (also known as state-owned enterprizes), as they are controlled by the government. International Financial Services London (2009) classifies sovereign investment vehicles in two broad categories: sovereign wealth funds and other sovereign investment vehicles. Sovereign wealth funds can be divided into two groups: stabilization and savings funds. Savings funds have mandates to invest in internationally diversified portfolios, while stabilization funds provide support for the budget deficit. The other investment vehicles comprise: pension reserve funds, investment funds (normally counted as reserves), development funds (targeted towards domestic socioeconomic projects), and sovereign-owned enterprizes. The latter are companies over which the state has significant control, where the minister of finance or managers from the savings fund generally sit on the board of directors. These enterprizes often undertake foreign investments, and in several cases they do so in conjunction with savings funds. For example, Singapore Technologies Telemedia Pte Ltd is 100% owned by Temasek Holding, the investment arm of the government of Singapore, and as such it promotes the investment strategies of the government. Thus, sovereign-owned enterprizes represent just another vehicle of government investment, and any analysis of foreign government investment should consider not only SWFs, but also sovereignowned enterprizes. We group sovereign-owned enterprizes together with government savings funds under the term SWF. There is no central data provider that collects information on SWF investments, instead there are several, partial, privately held databases that try to track SWF investments, but none of them is complete. Thus, we construct an independent sample of SWFs as large shareholders in U.S. publicly listed companies.3 The sample is based on the self-reporting of SWFs under the 1934 Exchange Act. Rule 13d-1 of the Act imposes on investors that cross the 5% investment threshold to declare their investment intentions to the SEC. If investors have an interest in influencing the management of the company, they file a Schedule 13D report. In particular, Item 4 of Schedule 13D requires the filing investor to declare the reasons for acquiring the shares. Investors filing Schedule 13D have to do so within ten days of acquiring over 5% of the

214

ELVIRA SOJLI AND WING WAH THAM

outstanding shares of the company. Also, ‘‘any material change’’ in the position or intention of the investor should be promptly reported via an amendment to the 13D filing (Schedule 13D/A). If investors have not acquired the securities with the purpose of changing or influencing the control of the issuer, i.e. acquiring stock as part of the normal course of business, they file a Schedule 13G report. Investors filing Schedule 13G should report no later than February 14 following the calendar year of the investment, i.e. one year after the investment has occurred. Filing a 13G form implies that it is illegal for the investor to engage with management in any form, thus they can not even do monitoring. Our data collection comprises several steps. First, we search the web and create a database of all possible sovereign wealth funds and enterprizes, their branches, and major investments. We cross-check and amend this list with information provided from SWF Radar, SWF Institute, and Amadan International. The first two are nonprofit organizations that follow SWF investments, the latter is a for-profit research and consulting firm that specializes on gathering and analyzing information on SWFs.4 Next, we identify SWF investments using Item 2 (‘‘Identity and Background’’) in the SEC filing reports. We search all Schedule 13D and Schedule 13G filings, separately, for SWF investors in Item 2. Table A1 in the appendix provides a list of all the Schedule 13 (D and G) investors identified in Item 2. In the third step, we gather and analyze all the Schedule 13D filings and their amendments. Filers provide information on the filing date, transaction date, ownership stake and its changes, number of shares and voting rights, and purpose of investment. From the information in Item 4 ‘‘Purpose of Transaction,’’ we exclude events related to bankruptcy reorganization or the financing of a distressed firm. We perform the same search for investments filed under Schedule 13G. We find 58 Schedule 13D filings and 35 Schedule 13G filings (hereafter referred to as 13D and 13G, respectively). We follow a similar procedure to identify a comparison group of hedge fund investments. We search for any 13D filings that include the phrase ‘‘hedge fund’’ in Item 2 ‘‘Identity and Background.’’ Furthermore, we search for filings from the largest hedge funds like: DE Shaw, Soros, Man Group, and funds recorded in Hedge Fund Research. The search yields 122 such investments for the period 1997–2008, and we can find information on the Center for Research in Security Prices (CRSP) and Compustat for 98 firms. There are no overlaps in the samples of SWF and hedge fund target firms. We collect data on foreign sales and foreign assets of the SWF target companies (13G and 13D) from the annual reports and the SEC 10 K filings, to investigate DOI. We use this information to calculate the ratio of the

SWF Investments in the United States

215

foreign sales to total sales and foreign assets to total assets. The DOI is then measured as the average of the two, Sullivan (1994).5 No database collects information about granted government-related contracts outside the United States, to the best of our knowledge. Thus, we search Factiva for news articles of contracts signed by the target company in the country where the SWF investment originates from. For each firm in our sample, we use five key words for the search: deal, lease, contract, permit, and patent. We study all the contracts and deals before and after the SWF investment date to extract the news articles that are related to government sanctioned contracts. We find 88 such deals all pertaining to 13D investments, 27 before and 61 after the SWF investment. We conduct a similar search for the matched firms of SWF targets. Two examples might help the reader to understand the nature of these government-related deals. The first is the link between Dubai and Nasdaq OMX. Investment Corporation of Dubai and Borse Dubai Limited purchased 30.3% of the outstanding shares of Nasdaq OMX on February 27, 2008. On July 2, 2008, the Organization of Asia-Pacific News Agencies reports that Dubai International Financial Exchange (DIFX) will start working with a new software supplied by Nasdaq OMX Group Inc. In November 2008, DIFX was renamed Nasdaq Dubai. Another example is the recent purchase of 8.1% of the outstanding shares of Advanced Micro Devices (AMD) by Mubadala Development Co. on October 8, 2008. On January 2, 2009, ArabianBusiness.com reports: ‘‘The Committee on Foreign Investment in the United States (CFIUS) has cleared a joint venture between Advanced Micro Devices and the Advanced Technology Investment Company [ATIC], owned by the government of Abu Dhabi. The new company will be headquartered in Silicon Valley, with its R&D and manufacturing teams in New York, Dresden, and Texas. As part of the agreement, ATIC will own the controlling interest with a 55.6% share in the venture. Mubadala Development Co., another Abu Dhabi investment company, is increasing its investment in AMD to 19.3% from its present 8.1% by purchasing shares and warrants worth $314 million.’’

SWF INVESTMENT CHARACTERISTICS Stated Objectives and Investment Characteristics Table A2 Panel A in the appendix summarizes the stated objectives of SWFs for their investments in U.S. listed companies. These objectives can be

216

ELVIRA SOJLI AND WING WAH THAM

classified into five broad categories (Brav et al., 2008). The first investment objective category, general undervaluation/maximize shareholder value, includes all the events in which the investor thinks that the target firm is undervalued and/or that the investor can help to increase shareholder value. This is the largest category, comprising over 51% of the sample. It is not surprising that SWFs main stated objective is a general one that involves nonaggressive talks with the management, given their politically controversial role. SWFs’ intentions are more likely to be linked with monitoring and engaging with the management to improve firm value in a nonconfrontational manner, rather than act as pure activist investors and undertake proxy fights.6 The second objective, business strategy, comprises 24% of the filings. The third category, sale of target company, includes 19% of sample. The fourth group of events, governance, relates to changes which include board independence and fair representation, more information disclosure, rescinding takeover defences, etc. Generally, SWF investors in this category request and achieve the inclusion of a representative on the board of directors. All the events in which the investor aims to affect and change the payout policy and/or the capital structure of the target company are grouped in the last category, capital structure, which is also the smallest. The objectives, apart from maximizing shareholder value, are not mutually exclusive because an investor can adopt a combination of tactics to achieve the aim of improving the value of the target company. Table A2 Panel B provides an overview of the target industries for 13D and 13G investments. The majority of 13D targets are in the IT & Telecom and manufacturing sectors. However, SWF 13G investments focus on business services (prepackaged software and computer programming and data processing), which is also where entrepreneurial hedge funds invest heavily (Klein and Zur, 2009). The distribution of target industries is considerably different from other studies, which report that SWFs largely invest in the finance and real estate industries (20% in Fernandes, 2009 and 43% in Bortolotti et al., 2009). The different findings imply that SWFs target different companies as large and active shareholders. The 13D and 13G investments also differ in terms of investment shares and sizes, presented in Table A2 Panel C. The 13D investments reach up to a 70.6% holding of the target company and the average investment size is twice as large as that of 13G investments. The average holding is about 22% for 13D investments and 8% for 13G investments.7 Panels D and E in Table A2 show the country of origin of the investors and the investment year. 13D investments are evenly distributed among the different countries in the sample and they occur more often after 2002, as the size of SWFs

SWF Investments in the United States

217

has grown. Thus, our analysis is not biased towards the investment decisions and the policy of one individual country over a certain time period.

What Attracts SWF Investments? Because SWFs file for both active and passive investments, we are interested in understanding what types of companies they become actively involved in. The first two columns of Table 1 provide an overview of the characteristics of SWF target firms in the year before they are targeted. We compare 13D targets with three groups of peers: matched group, 13G investments, and hedge fund 13D targets. Comparing 13D and 13G target characteristics sheds light on the circumstances in which SWFs choose to actively engage with management versus being passive shareholders. The comparison with hedge funds helps to assess whether SWFs have the same objectives and target the same firms as other active investors. Initially, we compare the 13D target firms with a group of industry/size/ book-to-market matched companies, taken from the universe of companies reported in Compustat. The matched firms are from the same year, share the same two-digit SIC code, and pertain to the same size/book-to-market 5  5 sorted portfolio. For the comparison of market capitalization, the matched firms are matched based only on industry and book-to-market. For the book-to-market and Tobin’s q comparison we only match according to industry and size. We calculate the difference in difference between the foreign government target firms (treatment group), their matched firms and 13G investments. The 13D investment targets are deemed to display different characteristics from peer groups if the Wilcoxon and Student t-tests are associated with a p-value lower than 10%. Column (3) of Table 1 shows the average difference between 13D target firms and the matched group, and column (4) presents the t-statistic associated with the difference in characteristics. SWF targets are similar to matched firms in terms of capital structure and profitability, and they do not appear to be poorly governed. They only differ in terms of firm valuation, in comparison to matched firms. 13D targets have 46% lower Tobin’s q (defined as (book value of total assets þ market value of equity-book value of equity)/book value of total assets) and an 11% higher book-to-market (BM) than their matched group.8 The comparison with 13G targets yields similar results to the matched group, in terms of firm size and capital structure. 13D targets exhibit higher

218

Table 1. Firm Characteristic

Mean (1) 11,201.22 0.60 1.70 0.15 0.09 0.37 0.18 0.01 0.11

Difference with Matched

Difference with 13G

Difference with HF

Standard Deviation Average Difference t-Statistics Average Difference t-Statistics Average Difference t-Statistics (2) (3) (4) (5) (6) (7) (8) 26,805.08 0.31 1.32 0.21 0.18 0.28 0.23 0.02 0.23

3,079.70 0.12 0.46 0.01 0.01 0.01 0.04 0.00 0.06

1.01 2.43 2.71 0.38 0.27 0.16 1.35 0.10 0.98

1,506.00 0.01 0.39 0.06 0.03 0.04 0.02 0.00 0.13

0.39 0.29 2.08 1.98 1.07 0.88 0.61 0.53 3.91

10,915.00 0.18 0.34 0.18 0.16 0.03 0.01 0.05 0.06

2.82 4.04 1.80 5.68 5.63 0.87 0.17 15.03 1.63

Notes: This table presents the summary statistics of SWF target firms and comparisons with a set of matched firms, 13G, and hedge fund targets for the year before the SWF investment. Columns (1) and (2) present the mean and the standard deviation of the target company characteristics. Columns (3) and (4) present the average difference with an industry/size/book-to-market matched group and the associated t-statistic. The industry match is from the Compustat universe of firms that share the same two-digit SIC code. Matching is based on industry and book-to-market for the MV comparison, and on industry and size for the BM and Tobin’s q comparison. Columns (5) and (6) show the difference with 13G investments and the associated t-statistic. Columns (7) and (8) show the difference with hedge fund investments in the same period and the associated t-statistic. MV is the market capitalization in million U.S. dollars; BM the book to market ratio defined as (book value of equity/market value of equity); Tobin’s q defined as (book value of total assets þ market value of equity-book value of equity)/ book value of total assets; ROA the return on assets defined as EBITDA/lag(assets); CF the cash flow defined as (net income þ depreciation and amortization)/lag (assets); LEV the book leverage ratio defined as debt/(debt þ book value of equity); CASH defined as (cash þ cash equivalents)/assets; DIVYLD the dividend yield defined as (common þ preferred dividends)/(market value of common stocks þ book value of preferred); and PAYOUT the payout ratio, defined as total dividend payments divided by net income before extraordinary items. All the data is obtained from CRSP/Compustat. , , and  represent significance at the 10%, 5%, and 1% level, respectively.

ELVIRA SOJLI AND WING WAH THAM

MV BM Tobin’s q ROA CF LEV CASH DIVYLD PAYOUT

Summary Statistics

Firm Characteristics.

SWF Investments in the United States

219

returns for the assets used in production (ROA, the ratio of EBITDA to lagged assets). Furthermore, 13D targets have higher payout ratios (PAYOUT, the ratio between total dividend payments and net income before extraordinary items) than 13G targets. The most striking difference in characteristics is between 13D and hedge fund targets, as shown in columns (7) and (8). The market value of 13D firms is much higher than that of hedge funds. SWF investments are 10 times larger than the ones reported for activist hedge funds (Brav et al., 2008; Klein & Zur, 2009). This is not surprising as the total amount of investment resources available to each SWF is much larger than that available to each hedge fund individually. Thus, SWFs might be in a unique position to affect companies where hedge funds do not have enough capital to invest. In terms of firm valuation, 13D targets have significantly higher Tobin’s q but lower book-to-market than hedge fund targets. SWF investments appear to be financially more sound than hedge fund investments, if one considers ROA and cash flows (CF). In terms of capital structure, 13D investments have lower dividend yields (DIVYLD) than hedge fund targets. These statistics clearly reject any hypothesis that SWFs behave similarly to hedge funds in their choice of targets. In summary, SWFs appear to invest in technology and manufacturing companies whose value can be improved, but they do not target the worst companies or the ones in need of most change like hedge funds do.

THE EFFECT OF SWF INVESTMENTS ON FIRM VALUE Univariate Analysis We first analyze some statistics for the main variables of interest before and after the SWF investment summarized in Table 2 for SWF targets and their matched firms.9 Industry Tobin’s q (Industry q) is the average Tobin’s q of firms that share the same first two digits of the SIC codes with SWF targets, and relative Tobin’s q (Relative q) is the difference between firm Tobin’s q and industry Tobin’s q. The pre-investment period covers three years before the investment, whereas the post-investment period lasts until divestment. The average Tobin’s q of the target firms significantly increases in the postinvestment period by 31%, from 1.60 to 2.14. The average q of the SWF target companies is lower than the Industry q before and 18% larger than the Industry q after the SWF investment. The change relative to the industry

220

ELVIRA SOJLI AND WING WAH THAM

Table 2.

Pre- and Post-Investment Characteristics. Pre

DOI Tobin q Industry q Relative q Gov. Cont. % Gov. Cont. %, Matched Group Diff. Gov. Cont. (13D-Matched) (in %) Gov. Cont. %, G13 Firms Gov. Cont. Number Total Presence (in %) Total Presence, Matched Group (in %) Diff. Total Presence (13D-Matched) (in %) New Blockholders 1 m (in %) New Blockholders 1y (in %)

0.16 1.60 1.94 0.34 15.52 19.57 4.05 0.00 1.17 34.48 39.13 4.65 1.72 13.79

Post

Diff.

0.24 2.14 1.81 0.33 36.21 19.57 16.64 0.00 2.65 58.62 45.65 12.97 8.62 32.76

0.08 0.54 0.13 0.67 20.69 0.00 21.37 0.00 1.48 24.14 6.52 17.62 6.60 18.97

Notes: This table presents preliminary statistics for 13D target firm characteristics, pre- and postSWF investment. Pre refers to the average of the three years before the SWF investment. Post covers the period after the SWF investment, until divestment or the end of the sample period. Diff. is the difference between the pre- and post period characteristics. DOI is the degree of internationalization calculated as the average of the foreign share of sales and assets. Tobin’s q is defined as (book value of total assets þ market value of equity-book value of equity)/book value of total assets as reported in Compustat. Industry q is the average Tobin’s q of all the firms in Compustat that share the same two-digit SIC code with the target firm. Relative q is the difference between Tobin’s q and Industry q. Govt. Cont. % is the percentage of SWF target firms that receive government-related deals as collected from the Factiva search. Govt. Cont. %, Matched Group is the percentage of matched firms that receive government deals as collected from the Factiva search. Diff. Gov. Cont. (13D-Matched) is the difference between the average number of SWF target firms and matched firms that receive government-related contracts. Govt. Cont. %, G13 Firms is the number of government-related contracts that 13G target firms receive. Govt. Cont. Number is the average number of deals for the SWF target firms that did obtain such contracts. Total Presence is the percentage of firms that have a presence in the investor’s country as either reported in Segments Compustat or obtained from the Factiva search. Total Presence, Matched Group is the matched firms presence in the investor’s country as either declared in Segments or obtained from the Factiva search. Diff. Presence (13D-Matched) is the difference between the average total presence of SWF target firms and matched firms in the investor country. New Blockholders represent the percentage of SWF target firms where there are new shareholders with shares of more than 5% that filed a 13D form a month or an year before and after the SWF investment. , , and  represent significance at the 10%, 5%, and 1% level, respectively.

is statistically significant. DOI also increases by 50% in the post-SWFinvestment period. In the period before the SWF investment, only 15% of the firms in our sample receives government-related deals (Govt. Cont. %) from the investor’s

SWF Investments in the United States

221

country, whereas this share increases significantly to 36% in the period after SWF investments. In contrast, 19% of matched firms receive a contract in the SWF country before the SWF investment in the target firm and the same after. The gap in government-related deals between the SWF target firms and the matched group widens substantially after the SWF investment in the target firm, as SWF targets receive substantially more government-related deals. It is also worth noting that 13G target firms do not receive such contracts. The average number of deals given per SWF target (Govt. Cont. Number) doubles in the post SWF investment period, and the increase is significantly different from zero. In the sample, China provides contracts to 31% of the SWF target companies that receive at least one, United Arab Emirates to 27%, Saudi Arabia to 18%, and Singapore to 18%. For robustness, we record the number of target companies and matched firms that have sales in the SWF investor country prior to the investment, as reported in the Segments database in Compustat. Unfortunately, the data reported in Segments is very often aggregated at the continent level, making it very difficult to distinguish in which specific country companies are present. Furthermore, our Factiva and company annual report search indicates that some companies already had a presence in the SWF investor country, which the Segments database does not report. Thus, we merge the information from the Factiva search to that from the Segments database in Total Presence. From merging the two datasets, we find that 34% of our SWF target sample and 39% of the matched firms already had a presence in the investor’s country before the SWF investment, an indication that SWFs target companies are behind their peers in accessing these particular markets. The presence of target firms increases significantly, by 24%, while that of matched firms by 6% after SWF investments occur.10 The presence of matched firms is substantially lower than that of SWF target firms in the post SWF investment period. Only in 5% of the cases, SWF target companies that have a prior presence in the country receive government-related deals. Table 2 also provides information on new 13D blockholders investing in the target preceding and following SWFs investments. We find that for only 2% of the firms there are new investors in the month before the SWF investment and for 9% of the firms in the month after the investment. In the year surrounding the SWF investment, 14% of the firms experience new investments from blockholders before and 33% after the SWF investment. The post-SWF investment increase in blockholders is statistically significant. It appears that other investors perceive SWFs as informed/value creating investors. These preliminary results indicate that SWF targets experience significant increases in q, DOI, and government-related deals after the investment. It is

222

ELVIRA SOJLI AND WING WAH THAM

unlikely that SWFs have access to a category of managers that is superior to that of hedge funds or asset management firms in choosing target firms, but they seem to generate value by providing access to a pool of investment opportunities previously unavailable to the target firm.

Market reaction around the 13D filing We investigate the short-run performance of SWF investments by constructing the cumulative abnormal returns (CARs) starting from 20 trading days before the 13D filing and ending 20 days after the filing. CARs are calculated above the expected market returns estimated for the period (255,20) measured using the value-weighted NYSE/Amex/NASDAQ index from CRSP.11 Fig. 1 shows the evolution of the average CAR through the event window. Positive abnormal returns start to accumulate about 10 days before the filing. The run-up can be explained by the fact that transactions occur sometime between 10 and 1 day before the 13D filing. The filing is important in revealing whether the SWF has any intention to actively engage with the company. The average CAR reaches 11.5% just before the 13D filing date and an upward trend continues for the 20 days after the filing. The total average abnormal return for the event window is 12.5%, which is statistically different from 0. There is a 2% increase in CAR on the day of the filing, which might be related to the active engagement intentions. Sixty percent of the targeted firms experience positive abnormal returns: the 25th, 50th and 75th percentile values are 8%, 7%, and 24%, respectively. The abnormal returns remain significantly different from zero for smaller event windows, but they decrease in size to 3% for (1, þ 1) and 8% for (10, þ 10). The abnormal returns are higher than those reported in Dewenter et al. (2010) most probably due to our focus only on deals where SWFs acquire large stakes and intend to actively engage with management. We compare the average CAR for 13D filings with those of 13G filings, to evaluate whether there is a difference on how the market perceives these investments. For this comparison, we use the investment date (i.e., the date in which the transaction occurs) instead of the filing date as the event date. This is done because the 13G report can be filed up to one year after the investment. The share price impact of the SWF investment would have completely disappeared by the 13G filing date. Furthermore, we only include investments where SWFs acquire similar holdings in the target 13D firms as in 13G targets for comparison purposes.

223

SWF Investments in the United States 600.0%

23.0%

18.0% Abnormal Volume

400.0% 13.0% 300.0%

8.0%

200.0%

Cumulative Abnormal Return

500.0%

3.0%

100.0%

−2.0% 20

18

t+

16

t+

14

t+

12

t+

t+

8

10 t+

6

t+

4

t+

2

t+

13

t+

4 D t−2 Fi lin g

6

t−

8

t−

10

t−

12

t−

14

t−

16

t−

18

t−

t−

t−

20

0.0%

Volume

CAR

Fig. 1. Average Cumulative Abnormal Return Around 13D Filing. Notes: This figure presents the average cumulative abnormal return for 13D investments (right axis), in excess of the market returns (value weighted CRSP index) for the period (20, þ 20) days around 13D filings. The dotted lines represent the 95% confidence bounds for the average CAR. The bars (left axis) show the increase (in percentage points) in trading volume during the same (20, þ 20) window compared to the average volume during the preceding (100, 40) event window.

Fig. 2 presents the average CAR around the investment date. First, we notice that 13D investments generate much larger abnormal returns, which remain high throughout the event window. The magnitude of the average CAR for 13D firms is similar to the one reported in Fig. 1, although we only include the smaller 13D investments. The market expects higher future cash flows from 13D targets than 13G ones, as a result of the SWF investment. The CAR for 13G investments is 4.7% and not statistically different from zero, while that of 13D investments is 10.6% and statistically different from zero. Second, there is a run-up for both types of investments two days before the investment, when the 5% threshold is crossed. This is not surprising for 13G deals, since some of the recent 13G transactions have been widely discussed in the media as they involved nonvoting stakes in high profile companies, i.e., Temasek’s investment in Merrill Lynch. The run-up for 13D events is more

224

ELVIRA SOJLI AND WING WAH THAM

14.0% 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% −2.0%

t−20 t−18 t−16 t−14 t−12 t−10 t−8 t−6 t−4 t−2 Event t+2 t+4 t+6 t+8 t+10 t+12 t+14 t+16 t+18 t+20 Data

−4.0% 13D

13G

Fig. 2. Cumulative Abnormal Returns Around 13D and 13G Investment Dates. Notes: This figure presents cumulative abnormal returns for the 13D and 13G investments, in excess of the market returns (value weighted CRSP index) for the period (20, þ 20) for the event date, i.e. when the benchmark of 5% is actually breached, as reported in the filings. Only 13D firms that acquire similar proportion of shares outstanding in a company to 13G firms are included.

difficult to explain, as we cannot find any discussion of the deals before the investment date via a Factiva search. However, one possibility is that the SWFs are being front-run, as they build up their position in the target companies and do not acquire all the shares in one transaction. The statistically and economically large cumulative abnormal returns after the SWF investment provide initial evidence that the market expects the benefits deriving from such foreign and politically connected investors to outweigh the costs.

Explaining the cross-sectional differences in market reaction SWFs are large investors that originate in a foreign country where they are politically connected. Large investors have the potential to generate positive returns for the firm by increasing monitoring of the management, as well as

SWF Investments in the United States

225

to extract personal benefits. Politicians and media heavily scrutinize SWFs as investors, therefore SWFs try to minimize the publicity related to the extraction of personal benefits from target companies and focus on interacting with management to create value. As foreign investors, SWFs can provide investment opportunities and expertise in markets where the target firm does not have experience, and thus generate value via foreign market access. As politically connected, SWFs can create value by giving government-related contracts to target firms. We try to separate the effect of each of these characteristics on market reaction and test three hypotheses: i.

There is a positive relation between abnormal returns and the benefits of monitoring. ii. There is a positive relation between abnormal returns and the expectation of future internationalization. iii. There is a positive relation between abnormal returns and the expectation of future government-related deals. To investigate the first hypothesis, we follow (Brav et al., 2008) and use dummy variables derived from 13D filings, as presented in Table A2 Panel A. Each of the five explanatory variables is a dummy variable that takes the value of one if the intended objective is the measured one, and zero otherwise. For the second hypothesis, we use the difference between the DOI in the year of the investment and the average DOI for the years after the investment, DDOI. To investigate the third hypothesis, we include a dummy variable taking the value of one if the company receives contracts after the SWF investment and zero otherwise, Govt. Cont.12 We consider this to be a lower bound estimate of the number of contracts received per company, as the Factiva search might not cover all the contracts awarded. There is a strong empirical link between secondary market liquidity and shareholder dispersion. Liquidity measures worsen with more blockholders (Mukherji, Kim, & Walker, 1997) and the probability of trading against an informed shareholder increases (Heflin & Shaw, 2000). We control for liquidity using the difference in Amihud illiquidity measures three months before the announcement (100,40) and during the announcement (20, þ 20), Illiquidity Ann. In our estimation, we also control for: the size effect measured using the log of market capitalization, ln(MktCap), on the day before the first date of the event window (20); the average abnormal return for the window (20, þ 20) obtained from the same investor in previous 13D investments as an indication of previous success, Avg. Pre-return; firm distress due to debt problems using Long Term Debt, a dummy variable that takes the value of one if the firm has outstanding long-

226

ELVIRA SOJLI AND WING WAH THAM

term debt in the investment year, and zero otherwise; the initial investment share in the target firm, Initial Investment; and Diversification, the number of markets a firm is present in taken from the Segments database in Compustat. All the nondummy variables are deviations from the sample mean, thus, they can be interpreted as partial effects of the explanatory variable on the CAR. No intercept is included in the regression since the investment intention dummies span the array of the constant Suits57. In Table 7, we present the results of an OLS regression (correcting for autocorrelation and heteroscedasticity). We find that all monitoring dummies have positive coefficients that lead to 7% to 20% higher abnormal returns, but not all of them are statistically significant. Intentions to change the business strategy and capital structure are statistically significant and have the highest impact. This is an indication that the market expects SWFs to engage with management, mostly by influencing the business strategy. In addition, DDOI has a large impact on returns, implying an increase of 1.22% in the returns for a 1% change in DOI. The market prices potential future benefits that derive from expansion in new markets that are related to the investor. This is consistent with the findings of Gozzi et al. (2008) that share prices reflect the potential benefits of internationalization quickly. Government contracts, on the contrary, do not appear to have a statistically significant impact. This result might be due to the difficulty in understanding the link between SWF investors and the potential to obtain governmentrelated contracts (Table 3). From the control variables, it is worth noting that a decrease in liquidity of the asset by 1% leads to higher CAR of 0.12%. Also, the market expects 0.98% higher CAR whenever the previous investments abnormal returns increase by 1%. As expected, the more diversified firms exhibit lower abnormal returns, but the effect is statistically insignificant. The size of the initial investment and firm size do not affect short-term returns. Long-term debt also does not have a significant impact on CAR, implying that the market does not expect debtholder expropriation from SWFs.13

Temporary Price Impact An alternative explanation for the high abnormal return observed around and after the 13D filing is that the price increase is only due to price pressure from excessive demand, and the change in price is temporary. Then, we would expect a temporary increase in prices around the 13D filing and a subsequent downward correction. The abnormally high trading volume

227

SWF Investments in the United States

Table 3.

General Business strategy Sale Governance Capital structure D DOI Govt. Cont. ln(MktCap) Illiquidity Ann Avg. Pre-return Initial investment LT Debt Diversification Adj. R2

Determinants of Short-Run Abnormal Returns. Coefficient

t-Statistics

0.18 0.20 0.20 0.07 0.18 1.22 0.04 0.01 0.12 0.98 0.11 0.11 0.01

1.33 2.35 1.61 0.64 2.07 1.70 0.66 0.06 8.86 2.10 0.76 1.09 0.82

0.20

Notes: This table presents the relation between the 13D filing cumulative abnormal returns and market variables. The dependent variable is the cumulative abnormal return during the period (20, þ 20) days around the Schedule 13D filing. The explanatory variables are: five measures of active engagement as presented in Table A2, Panel A; a measure of the change in the degree of internalization, DDOI, calculated as the difference between DOI (average of foreign sales and assets to total sales and assets) the year before the investment and the average DOI subsequent to the SWF investment; Govt. Cont. a dummy variable that takes the value of one if the company is awarded any government-related contracts after the investment, or zero; ln(MktCap) the log of market capitalization; Illiquidity, the difference in Amihud measures three months before the announcement (100, 40) and during the event window (20, þ 20); the average return obtained from the same investor in previous 13D filing related investments Avg. Pre-return; Initial Investment accounts for the size of the initial investment; LT Debt a dummy variable for the existence of long-term debt in the firm in the investment year; and Diversification a variable that contains the total number of geographical segments where the company is present. t-statistics (t-stat) are heteroscedasticity consistent. All nondummy variables are expressed as deviations from the sample mean. There is no intercept in the regression. , , and  represent significance at the 10%, 5%, and 1% level, respectively.

around the announcement time, averaging 143% above normal as shown in Fig. 1, reinforces the overreaction explanation. However, the average CAR in Fig. 1 remains stable in the month after the 13D filing and does not revert towards zero. This is an initial indication that the overreaction explanation may not be valid. We investigate the temporary price impact alternative more formally using event-time portfolios for the years before and after the 13D filing. We

228

ELVIRA SOJLI AND WING WAH THAM

track all SWF target and matched stocks for periods of three months, for the year before and after the SWF investments. Every three months, we form a daily portfolio of all SWF target firms and of the matched firms, separately. The event period portfolio is held only for two months, the same as the event window used in the previous analysis. We estimate the portfolio excess return ‘‘Alpha’’ over the Fama-French factors: market excess return, small-minus-big (SMB), and high-minus-low (HML), and the Carhart (1997) momentum factor (MOM). Daily data are obtained from the FamaFrench directory in Wharton Research Data Services (WRDS). The ex ante portfolios could not have been constructed at the time of the investment, but they are informative of the development of the investment. Table 4 shows the regression results using the equally and value-weighted CRSP indexes as the benchmark market return.14 If the observed abnormal return is temporary, then we should observe a positive ‘‘Alpha’’ on the event period and negative ‘‘Alpha’’ after the event for the SWF targets portfolio.15 The results show that the only significant ‘‘Alpha’’ coefficient for SWF target firms is the one of the event period, i.e. (20, þ 20) days around the 13D filing and it is positive. There is no other statistically significant ‘‘Alpha’’ for any of the periods in the quarters after the SWF investment in the target firms. Therefore no reversion has occurred, and the observed event window CAR is unlikely to be due to market overreaction or temporary price pressure. In addition, there is no significant ‘‘Alpha’’ for the SWF targets in the year before the SWF investment. This finding can be considered as evidence against market timing. However, there are statistically significant ‘‘Alphas’’ for the matched firms in the year before the SWF investment for equally weighted portfolios. This effect disappears when value weighted portfolios are used. The fit of the four factor model is much better for matched firms than for the SWF targets.

Long-Run Impact of SWF Investments The short-run positive abnormal returns reflect the market expectation of firm future profits. This expectation might not materialize in the future, which makes the study of the long-term performance of the target companies necessary. To understand whether SWF investments generate long-term value for the company, we calculate the buy-and-hold abnormal returns (BHAR) for all the investments. We use the CRSP value-weighted index as the benchmark, starting from 20 days prior to the SWF investment

229

SWF Investments in the United States

Table 4.

Event Time Portfolios.

13D Firms

Matched Firms

Panel A: Equally weighted Window (Months)

Alpha Coefficient

t-Statistics

R2

Alpha Coefficient

t-Statistics

R2

(12, 10) (9, 7) (6, 4) (3, 1) Event (1,3) (4,6) (7,9) (10,12)

0.0006 0.0002 0.0011 0.0026 0.0040 0.0008 0.0002 0.0018 0.0003

0.75 0.31 1.39 1.85 2.17 0.98 0.20 1.57 0.32

0.07 0.06 0.09 0.05 0.10 0.12 0.10 0.15 0.21

0.0007 0.0007 0.0004 0.0001 0.0005 0.0002 0.0003 0.0006 0.0005

1.88 2.12 1.00 0.28 0.84 0.57 0.71 1.36 1.14

0.31 0.36 0.36 0.42 0.47 0.41 0.40 0.37 0.45

Panel B: Value weighted Window (Months)

Alpha Coefficient

t-Statistics

R2

Alpha Coefficient

t-Statistic

R2

(12, 10) (9, 7) (6, 4) (3, 1) Event (1,3) (4,6) (7,9) (10,12)

0.0009 0.0002 0.0012 0.0021 0.0033 0.0010 0.0006 0.0014 0.0002

1.01 0.20 1.43 1.52 2.02 1.19 0.54 1.07 0.21

0.07 0.07 0.09 0.05 0.12 0.12 0.11 0.13 0.21

0.0005 0.0007 0.0006 0.0003 0.0005 0.0003 0.0002 0.0001 0.0002

1.16 1.87 1.42 0.71 0.76 0.84 0.44 0.25 0.47

0.31 0.34 0.34 0.40 0.44 0.38 0.37 0.31 0.41

Notes: This table presents abnormal returns from the investment in SWF targets and the matched firms before and after their investment. These are regression estimates and t-stats from equal- and value-weighted portfolio regressions. Window refers to the investment period in months. Alpha is the intercept of the factor model regression, and is calculated using the fourfactor model with the value- and equally weighted NYSE/Amex/NASDAQ CRSP portfolio as benchmark, the Fama-French size and book-to-market factors, and the Carhart momentum factor. Panels A and B present the results using the equally and value weighted CRSP portfolio portfolio, respectively. t-statistics (t-stat) are heteroscedasticity consistent. All variables are taken from the WRDS Fama-French Factors database. , , and  represent significance at the 10%, 5%, and 1% level, respectively.

until the investment drops below the 5% threshold, as reported in a Schedule 13D/A statement.16 In cases where there is no exit report, we assume that the investment continues until the end of the investigated period, December 31, 2008. SWFs divest from a company only in 19% of

230

ELVIRA SOJLI AND WING WAH THAM

our sample. The average BHAR is 16% per year.17 However, BHAR is sensitive to risk adjustment and may result in overstating the abnormal performance (see Franks, Harris, & Titman, 1991; Lang & Stulz, 1994, and references therein). Thus, we focus on firm fundamentals to better understand the implications of SWF investments for target firms, instead of analyzing the determinant of BHAR.

Firm Performance We investigate changes in firm value, profitability, and efficiency, in the two years prior and subsequent to the investment, for an initial analysis of changes in firm fundamentals. We show these changes in comparison to the matched group, 13G, and hedge fund investments. We use Tobin’s q as a measure of firm value, and EBITDA/assets and EBITDA/sales as efficiency and profitability measures, respectively. All the data are taken from Compustat. Table 5 Panel A shows that SWF targets exhibit the same level of profitability and efficiency as their matched firms, before and after the SWF investment. The q of SWF targets is significantly lower, by 46%, than that of the matched firms the year before the SWF investment. During the first two years of SWF investment, q increases from being lower to not being statistically different from the matched firms. The q of SWF target firms increases significantly, by 28%, in comparison to the matched group from the year before the investment to two years after the investment. The difference with 13G targets is stronger than that with the matched group, as presented in Table 5 Panel B. The 13D target firms start with no difference in efficiency and profitability and 39% lower Tobin’s q than the 13G investments. In the two years subsequent to the SWF investment, 13D firms show 15% higher EBITDA/assets and 37% higher EBITDA/sales than the 13G firms. This is a significant increase in efficiency of 13% and in profitability of 29% for 13D targets in comparison to 13G targets, in a span of three years. Furthermore, q increases from 39% significantly worse than 13G targets to 10% higher than that of 13G targets, a statistically significant increase of 49%. These results suggest that SWF investors engage with 13D targets in a different way compared to 13G targets. One important difference is that none of the 13G targets in our sample receives governmentrelated contracts before or after the SWF investment. Comparing the long-term performance of 13D investments and hedge fund investments sheds light on some important differences between the two groups of investors, as presented in Table 5 Panel C. Generally, SWF target

231

SWF Investments in the United States

Table 5. Ebitda/Asset

Target Firm Performance. t-Statistics

Tobin’s q

t-Statistics

0.21 0.05 0.08 0.20 0.11 0.06

1.00 1.18 0.45 0.75 1.29 0.67

0.27 0.46 0.40 0.30 0.18 0.28

1.97 2.71 2.50 1.91 1.18 3.45

0.27 0.09 0.07 0.15 0.37 0.29

1.10 1.56 0.36 0.55 7.33 3.20

0.17 0.39 0.77 0.16 0.10 0.49

1.44 2.08 5.35 0.65 0.60 2.63

Panel C. Difference with hedge funds investments 5.63 0.04 t2 0.14 4.75 0.41 t1 0.13  6.39 0.03 Event 0.11 3.27 0.17 tþ1 0.07 3.76 0.22 tþ2 0.11 (t þ 2)  (t1) 0.02 0.49 0.19

0.17 7.55 0.18 0.64 4.37 2.16

0.03 0.34 0.15 0.14 0.37 0.02

0.22 1.80 1.02 0.59 2.26 0.10

t-Statistics

Panel A: Difference with matched t2 0.01 t1 0.03 Event 0.00 tþ1 0.02 tþ2 0.01 (t þ 2)  (t1) 0.03

firms 0.6 1.51 0.15 0.78 0.28 0.67

Panel B: Difference with 13G investments t2 0.04 1.67 t1 0.01 0.53 3.59 Event 0.06 3.08 tþ1 0.07 5.05 tþ2 0.15 2.82 (t þ 2)  (t1) 0.13

Ebitda/Sale

Notes: This table presents accounting measures of target company performance in the period of two years before and after SWF investments. Panel A presents the difference with an industry/ size/book-to-market matched group and the t-statistic for the difference is based on unequal variance groups. Matching is based on industry and book-to-market for the MV comparison, and on industry and size for the BM and Tobin’s q comparison. Panel B presents the difference with 13G investments and the t-statistics (t-stat) on the difference based on unequal variance groups. Panel C presents the difference with hedge fund investments in the same period and the t-statistic on the difference based on unequal variance groups. , , and  represent significance at the 10%, 5%, and 1% level, respectively.

firms are significantly more profitable and more efficient than hedge fund investments. The difference in profitability, EBITDA/sales, decreases significantly in the post-hedge fund investment period by 19%, as does the difference in efficiency, but SWF targets still perform better on these two accounts. The Tobin’s q of hedge fund targets is 37% lower than those of SWFs’ two years after the investment, 2% higher than the pre-investment difference. This result is consistent with the findings of Brav et al. (2008) that hedge fund target performance improves significantly two years after the

232

ELVIRA SOJLI AND WING WAH THAM

investment in terms of efficiency and profitability. Overall, SWF investments lead to higher Tobin’s q and an increase in the efficiency of the firm as compared to matched firms, passive investments, and hedge funds.

SWFs Investments’ Effect on Tobin’s q We investigate the determinants of Tobin’s q for the SWF target firms following the methodology of Villalonga and Amit (2006). Tobin’s q has been widely used as a valuation measure in corporate-governance studies (e.g. Demsetz & Lehn, 1985; Morck, Shleifer, & Vishny, 1988; Lang & Stulz, 1994; Gompers, Ishii, & Metrick, 2003), among many others. Our dependent variables are Tobin’s q and relative Tobin’s q (Relative q). Relative q is used to alleviate the problem of estimation error for Tobin’s q (Gompers et al., 2003). In the regression, we control for changes in the dividend and leverage policy, total assets, one-year sales growth, the market beta over the five years before the investment, and diversification, the number of geographical segments a company is present in. The group of the control variables employed encompasses those used by Chen et al. (2007), Gozzi et al. (2008), and Ferreira and Matos (2008). We also include year and firm fixed effects in our regression. Table 6 shows the results of panel regressions for Tobin’s q and Relative q after the SWF investment, using robust standard errors adjusted for time and firm dependence (see Petersen, 2009; Thompson, 2006). Columns (1) and (3) present the regression results only for the control variables. Both Tobin’s q and Relative q increase when there are increases in the dividend yield, but total asset size, sales growth, and leverage do not have a significant impact on firm value. The effect of Industry q on the level of firms’ Tobin’s q is large and highly statistically significant. The control variables explain 76% of the panel variation in Tobin’s q. The explanatory power of the control variables is 18% for Relative q. Columns (2) and (4) include the effects of DOI and of governmentrelated deals (Govt. Cont.). Govt. Cont. is a dummy variable that takes the value of one for each year where a company receives at least a government-related contract, otherwise it is zero. We consider this to be a lower bound estimate of the number of contracts received per year, as the Factiva search might not cover all the received contracts. DOI has a statistically insignificant impact on Tobin’s q. This result is congruent with Gozzi et al. (2008) who find that change in internationalization does not affect Tobin’s q after the investment year. We find that the post-

233

SWF Investments in the United States

Table 6.

Explaining Tobin’s q After Investment. Tobin’s q

Dividend/book value of equity Debt/mkt value of equity ln(Assets) Growth Beta Diversification Industry Tobin q

0.32 [1.32] 0.01 [0.33] 0.01 [0.20] 0.31 [1.06] 0.04 [0.49] 0.01 [0.64] 0.98 [3.61]

DOI Gov. Cont. Year dummies Firm-fixed effects Adj. R2

Yes Yes 0.76

0.42 [1.79] 0.01 [0.04] 0.01 [0.08] 0.30 [1.05] 0.04 [0.51] 0.02 [0.85] 0.87 [3.29] 0.24 [0.45] 0.73 [3.20] Yes Yes 0.78

Relative q 0.32 [1.34] 0.01 [0.33] 0.01 [0.20] 0.31 [1.07] 0.04 [0.49] 0.01 [0.64]

0.43 [1.85] 0.01 [0.08] 0.01 [0.11] 0.28 [0.99] 0.04 [0.51] 0.02 [0.84]

Yes Yes 0.18

0.21 [0.40] 0.71 [3.18] Yes Yes 0.23

Notes: This table presents a panel regression of the Tobin’s q level of the target companies starting from the SWF investment until their divestment or the end of the sample. Relative q is the industry adjusted Tobin’s q Growth is the one-year sales growth of the company. Beta is the average market beta over five years, where the market benchmark is the S&P500 index. Diversification includes the total number of geographical segments where the company is present as reported in the Segments database in Compustat. DOI is average of foreign sales and assets to total sales and assets. Gov. Cont. is a dummy variable that takes the value of one if the company is awarded any governmentrelated contracts in a given year, and zero otherwise. , , and  represent significance at the 10%, 5%, and 1% level, respectively. t-statistics using robust standard errors adjusted for time and firm dependence are presented in square brackets.

investment increase in Tobin’s q is significantly related to the provision of government-related deals from the SWF acquiring countries. The Tobin’s q for a firm that receives government-related deals in a given year is a 0.73 higher than for a firm that does not receive a contract, and the impact is highly economically significant. This is not surprising as 36% of the sample receive some form of government-related deals in the postinvestment period. The adjusted R2 for the Tobin’s q regression increases by 2% and that of the Relative q increases by 5%. The control variables have the same sign across the different specifications and the results are

234

ELVIRA SOJLI AND WING WAH THAM

Table 7. Period

Market Value Impact of Government-Related Contracts. Raw

p-Value

Abnormal

p-Value

Panel A: Pre-SWF investment contracts Event 1.52 (1, þ 1) 1.24 (1, þ 5) 1.08 (1, þ 10) 0.64

0.13 0.36 0.48 0.48

1.56 1.15 1.15 1.16

0.08 0.35 0.40 0.44

Panel B: Post-SWF Investment contracts Event 2.31 (1, þ 1) 3.37 (1, þ 5) 2.17 (1, þ 10) 2.39

0.00 0.00 0.04 0.04

2.48 3.10 2.06 2.23

0.00 0.00 0.04 0.04

Notes: This table presents the cumulative abnormal returns around the announcement of a government-related contract for the 13D target firms in the SWF country. The benchmark model is the value-weighted CRSP index. Raw represents the raw abnormal returns and Abnormal presents the returns above the market model. Event is the day the contract is reported in the media, as found from Factiva. Panels A and B show the results for the contracts awarded before and after the SWF invested in the firm, respectively.

consistent to using Tobin’s q or Relative q. We find no country specific or industry effects.18 The above analysis indicates that one conduit for the value creation (increase in Tobin’s q) of SWFs as investors is the provision of government-related contracts. To further elaborate on the effect of government-related contracts, we perform an event study around the time of the reporting of the contract in Factiva. We analyze CARs with respect to the market model for all the contracts that we recorded in the Factiva search, divided in two groups: pre- and post-SWF investment. The event date is the date in which the first article on the contract appears in Factiva. We estimate the market model for the period (255,10) using the valueweighted NYSE/Amex/NASDAQ index from CRSP. We then calculate the abnormal returns for different time periods over the expected market returns. We find that the stock price increases substantially on the day of the announcement of such contracts, but the effects around the announcement date are different between pre- and post-SWF investment contracts. The results in Panel A of Table 7 show that there is only a temporary increase in stock price of 1.52% on the day of the announcement for contracts issued before the SWF investment. The CARs for any other period are not significantly different from zero. The impact of government-related

SWF Investments in the United States

235

contracts is much higher for deals reported after the SWF investment. Panel B in Table 7 shows that the announcement day impact is 2.48%, double that of pre-SWF investment contracts, and the ( þ 1,1) day return is 3.10% above the expected return. These returns remain high and significant for two weeks after the announcement. Thus, the government-related contracts contribute substantially to a nontemporary increase in the firm’s market value. SWF investment benefits appear to outweigh the costs for firm value and shareholders, as SWF target firms exhibit large BHAR and increases in Tobin’s q. Our results suggest that one of the mechanisms by which politically connected investors contribute to firm value are the provisions of contracts and permits. Furthermore, results from the Tobin’s q analysis in Table 6, the abnormal increase in share price due to the provision of government-related contracts in Table 7, and the univariate analysis of market access post-SWF investment in Table 2 show that foreign politically connected investors can also generate value, as they provide access to previously inaccessible markets.

CONCLUSIONS In this chapter, we investigate the role of a new class of institutional investors, namely Sovereign Wealth Funds, in financial markets. SWFs are different from any other investor analyzed so far, as they are not only large, but also foreign and politically connected. We find that SWFs invest in growth firms with low exposure to their countries, where they can provide diversification opportunities. The market reacts positively to SWF investments in the short-run, in the expectation of increased monitoring and future increases in internationalization. In the long-run, SWF target firms outperform hedge fund investments and catch up with the matched firms in terms of Tobin’s q. We find that the increase in q is due to the provision of government-related contracts. Our results point to the benefits of large and foreign investors for shareholders. In addition, we provide evidence that government-related contracts are one of the mechanisms by which government connections can affect firm value in the long run.

NOTES 1. Policy makers are especially interested in this point. Two recent examples are: an interview of UK political figures in the program Money Programme by BBC One

236

ELVIRA SOJLI AND WING WAH THAM

‘‘Who is buying up Britain?’’ where it is claimed that the investment by Abu Dhabi Investment Authority in Barclays Bank might lead to more investments of the company in Abu Dhabi, e.g., establishment of call centers in Abu Dhabi. Second, during the special theme discussion at the World Economic Forum 2008 in Davos, Larry Summers argued very strongly that foreign government-related shareholders might use their political leverage to extract political and nonpolitical benefits from target firms. 2. A further advantage of our dataset is that we know the exact event date and the number of shares acquired, instead of relying on the media reports of the event. 3. Data on SWFs as large shareholders is publicly available and we have collected it for France, the United Kingdom, and the Netherlands. Unfortunately, the intention of the investment and the divestment date is not included in this data, thus we cannot use it for most of our analysis. 4. It should be noted that our definition of SWFs is quite close to the definition of government controlled acquirers in Karolyi and Liao (2009). Karolyi and Liao (2009) use the acquirer’s ultimate parent as classified in Thomson Reuters Security Data Corporations (SDC) Platinum Mergers and Corporate Transactions dataset. We construct sovereign-owned enterprizes using information provided by different governmental institutions and savings funds and their portfolio holdings. We have cross-checked our data with SDC and find a 50% overlap in companies classified as government owned. 5. Ideally, we would also need to calculate the ratio of overseas subsidiaries as a percentage of total subsidiaries and top managers’ international experience, but this information is not available to us. 6. We do not find any proxy filings where SWFs rally shareholders to push for a policy change. 7. The total investment size into 13D and 13G companies, at the time of the investment, is 1.6% of the current size of the total SWF portfolio. This might seem a small fraction, but one has to bear in mind that a large fraction of the SWF portfolios is invested either domestically or in foreign bonds. Unfortunately, the exact size/proportion of their assets that is allocated to foreign equity is not publicly available. 8. There are different ways to calculate Tobin’s q, but Chung and Pruitt (1994) find that the above measure explains more than 95% of the variation in more complicated measures. For robustness, we replicate our results using different measures of q throughout the analysis. 9. The matched firm government-related deals and foreign country presence is calculated as follows. Instead of using the average presence for the group of matched firms, we set match firm government-related deals and presence equal to one if at least one of the matched firms receives a government-related deal or is present in the SWF country. We do this in order to mitigate the effects coming from the large number of matched firms that we use. Using the average for the matched firms as in Eq. (1), we find a substantially lower percentage of firms that receive governmentrelated deals and have lower presence in the SWF countries. 10. Given our choice of asymmetric pre- and post-investment period, one might argue that the difference in the statistics occurs because of the longer post-investment period. We do not believe this is the case for two reasons. First, the pre-investment

SWF Investments in the United States

237

period is longer than the post-investment period for firms that receive an SWF investment after 2006. Second, we calculate the differences in the reported variables for equal pre- and post-investment windows (i.e., take a pre-investment period that is as long as the post-investment), and the results remain qualitatively unchanged. However, some firms drop out of the analysis because there is no information for years further out in the pre-investment period, especially for foreign presence and sales. 11. We also use the Fama-French three factor model as the benchmark and the results remain qualitatively the same. 12. DDOI and Govt. Cont. are ex post measures. It is not common to use forward looking variables to explain current returns, but, in the case of internationalization, Gozzi et al. (2008) show that the benefits are immediately reflected in prices, i.e., the share price reacts upon announcement. Thus, it seems justifiable to use such a measure, especially given the observation that the DOI impact might not be identifiable after the event window (Gozzi et al., 2008). Using ex post data to infer ex ante values, we implicitly assume that investors are rational and their ex ante rational expectations are efficient. 13. Other variables that were used to explain the abnormal returns but are not significant are: the Truman transparency index, SWF Institute governance and transparency index, average of previous years’ GDP growth, World Bank ease of doing business variables, blockholders, industry and country dummies, and shortterm debt variables. We also used the current and previous year leverage as an explanatory variable instead of the long-term debt dummy, and the results do not change quantitatively. 14. We only report the ‘‘Alpha’’ and the R2 for the model to conserve space. The full results can be obtained from the authors upon demand. 15. A positive ‘‘Alpha’’ implies positive abnormal returns over the benchmark portfolio, outperformed the benchmark, and a negative ‘‘Alpha’’ implies a negative abnormal return compared to the benchmark portfolio, underperformed compared to the benchmark. 16. It is not possible to track investments less than 5% as SWFs are not financial institutions and do not have to file 13F forms. 17. We also calculated the cumulative market-adjusted return and the results are qualitatively similar. 18. We have estimated the effect of government-related contracts on the change in Tobin’s q and using different definitions of Tobin’s q and the result remain qualitatively similar.

ACKNOWLEDGMENTS We are grateful for constructive comments to Ajay Adhikari, Lauren Cohen, Ingolf Dittmann, Mara Faccio, Stefan Frey, Ulrich Hege, Jim Hodder, Harrison Hong, Abe de Jong, Markku Kaustia, April Klein, Josh Lerner, Arvind Mahajan, Robert Marquez, Marieke van der Poel, Jo¨rg

238

ELVIRA SOJLI AND WING WAH THAM

Rocholl, Peter Roosenboom, Lukas Roth, Michael Schill, Rob Weiner, participants at the 2010 European Winter Finance Summit, 2010 European Finance Association Meeting, 2010 Econometric Society World Congress, 2010 Asian Finance Association International Conference, 2010 Northern Finance Association Meeting, 2010 NTU International Conference on Economics, Finance and Accounting, AFFI 2010, Infinity 2010, 6th Portugese Finance Network Conference, and seminar participants at Erasmus University, the George Washington University (International Business Department), International Monetary Fund, Kogod School of Business, Norges Bank, University of Bristol, and Warwick Business School. Sojli gratefully acknowledges the financial support of the European Commission Grant PIEF-GA-2008-236948 and Erasmus Trustfonds. Tham gratefully acknowledges the financial support of the European Commission Grant PIEF-GA-2009-255330 and Erasmus Trustfonds.

REFERENCES Balding, C. (2008). A portfolio analysis of Sovereign Wealth Funds. Working Paper, University of California, Irvine. Beck, R., & Fidora, M. (2008). The impact of Sovereign Wealth Funds on global financial markets. Intereconomics, 43, 349–358. Bernstein, S., Lerner, J., & Schoar, A. (2009). The investment strategies of Sovereign Wealth Funds. Working Paper 14861, NBER. Bortolotti, B., Fotak, V., Megginson, W., & Miracky, W. (2009). Sovereign Wealth Fund investment patterns and performance. Working Paper, University of Oklahoma, Oklahoma. Brav, A., Jiang, W., Partnoy, F., & Thomas, R. (2008). Hedge fund activism, corporate governance, and firm performance. Journal of Finance, 63, 1729–1775. Bunkanwanicha, P., & Wiwattanakantang, Y. (2009). Big business owners and politics. Review of Financial Studies, 22, 2133–2168. Burkart, M., Gromb, D., & Panunzi, F. (2000). Agency conflicts in public and negotiated transfers of corporate control. Journal of Finance, 55, 647–677. Caner, M., & Grennes, T. J. (2008). Sovereign Wealth Funds: The Norwegian experience. Working Paper 1295364, SSRN. Carhart, M. M. (1997). On persistence in mutual fund performance. Journal of Finance, 52, 57–82. Chen, X., Harford, J., & Li, K. (2007). Monitoring: Which institutions matter? Journal of Financial Economics, 86, 279–305. Chung, K. H., & Pruitt, S. W. (1994). A simple approximation of Tobin’s q. Financial Management, 23, 70–74. Cronqvist, H., & Fahlenbrach, R. (2009). Large shareholders and corporate policies. Review of Financial Studies, 22, 3941–3976.

SWF Investments in the United States

239

Demsetz, H., & Lehn, K. (1985). The structure of corporate ownership: Causes and consequences. Journal of Political Economy, 93, 1155–1177. Dewenter, K. L., Han, X., & Malatesta, P. H. (2010). Firm values and Sovereign Wealth Fund investments. Journal of Financial Economics, 98, 256–278. Doukas, J., & Travlos, N. G. (1988). The effect of corporate multinationalism on shareholders’ wealth: Evidence from international acquisitions. Journal of Finance, 43, 1161–1175. Errunza, V. R., & Senbet, L. W. (1981). The effects of international operations on market value of the firm: Theory and evidence. Journal of Finance, 36, 401–417. Errunza, V. R., & Senbet, L. W. (1984). International corporate diversification, market valuation, and size-adjusted evidence. Journal of Finance, 34, 727–745. Faccio, M. (2006). Politically connected firms. American Economic Review, 96, 369–386. Faccio, M., & Parsley, D. (2009). Sudden deaths: Taking stock of geographic ties. Journal of Financial and Quantitative Analysis, 33, 683–718. Fatemi, A. M. (1984). Shareholder benefits from corporate international diversification. Journal of Finance, 39, 1325–1344. Fernandes, N. G. (2009). Sovereign Wealth Funds: Investment choices and implications around the world. Working Paper, European Corporate Governance Institute. Ferreira, M. A., & Matos, P. (2008). The colors of investors money: The role of institutional investors around the world. Journal of Financial Economics, 88, 499–533. Fisman, R. (2001). Estimating the value of political connections. American Economic Review, 91, 1095–1102. Franks, J. R., Harris, R. S., & Titman, S. (1991). The postmerger share-price performance of acquiring firms. Journal of Financial Economics, 29, 81–96. Gillan, S. L., & Starks, L. T. (2007). The evolution of shareholder activism in the United States. Journal of Applied Corporate Finance, 19, 55–73. Goldman, E., Rocholl, J., & So, J. (2009a). Do politically connected boards affect firm value? Review of Financial Studies, 22, 2331–2360. Goldman, E., Rocholl, J., & So, J. (2009b). Political connections and the allocation of procurement contracts. Working Paper, University of North Carolina at Chapel Hill. Gompers, P., Ishii, J., & Metrick, A. (2003). Corporate governance and equity prices. Quarterly Journal of Economics, 118, 107–155. Go¨rg, H., & Strobl, E. (2001). Multinational companies and productivity spillovers: A metaanalysis. Economic Journal, 111, 723–739. Gozzi, J. C., Levine, R., & Schmukler, S. L. (2008). Internationalization and the evolution of corporate valuation. Journal of Financial Economics, 88, 607–632. Heflin, F., & Shaw, K. W. (2000). Blockholder ownership and market liquidity. Journal of Financial and Quantitative Analysis, 35, 621–633. Karolyi, A. G., & Liao, R. C. (2009). What is different about government-controlled acquirers in cross-border acquisitions? Working Paper, Rutgers University. Klein, A., & Zur, E. (2009). Entrepreneurial shareholder activism: Hedge funds and other private investors. Journal of Finance, 64, 187–229. Kotter, J., & Lel, U. (2011). Friends or foes? The stock price impact of Sovereign Wealth Fund investments and the price of keeping secrets. Journal of Financial Economics, forthcoming. Lang, L. H. P., & Stulz, R. M. (1994). Tobin’s q, corporate diversification, and firm performance. Journal of Political Economy, 102, 1248–1280. Lipsey, R. E. (2002). Home and host country effects of FDI. Working Paper 9293, NBER.

240

ELVIRA SOJLI AND WING WAH THAM

London, I. F. S. (2009). Sovereign Wealth Funds 2009. Report, IFSL. Morck, R., Shleifer, A., & Vishny, R. W. (1988). Management ownership and market valuation: an empirical analysis. Journal of Financial Economics, 20, 293–315. Morck, R., & Yeung, B. (1991). Why investors value multinationality? Journal of Business, 64, 165–187. Morck, R., & Yeung, B. (2001). Why firms diversify: Internalization vs agency behaviour. In: B. Lev (Ed.), Intangibles. Oxford: Oxford University Press. Mukherji, S., Kim, Y. H., & Walker, M. C. (1997). The effect of stock splits on the ownership structure of firms. Journal of Corporate Finance, Contracting, Governance and Organization, 3, 167–188. Petersen, M. A. (2009). Estimating standard errors in finance panel data sets: Comparing approaches. Review of Financial Studies, 22, 435–480. Romano, R. (2001). Less is more: Making institutional investor activism a valuable mechanism of corporate governance. Yale Journal on Regulation, 18, 174–251. Shleifer, A., & Vishny, R. W. (1986). Large shareholders and corporate control. Journal of Political Economy, 94, 461–488. Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. Journal of Finance, 52, 737–783. Suits, D. B. (1957). Use of dummy variables in regression equations. Journal of the American Statistical Association, 52, 548–551. Sullivan, D. (1994). Measuring the degree of internationalization of a firm. Journal of International Business Studies, 25, 325–342. Thompson, S. (2006). Simple formulas for standard errors that cluster by both firm and time. Working Paper, Harvard University. Villalonga, B., & Amit, R. (2006). How do family ownership, control and management affect firm value? Journal of Financial Economics, 80, 385–417.

SWF Investments in the United States

241

APPENDIX

Table A1.

Sample of Investors.

Acquirer

Status

Abu Dhabi Investment Authority Aluminum Corporation of China Bank of China Beijing Holdings Limited Chia Tai International Telecommunication Company Limited China Aerospace International Holdings Ltd China Biotech Holdings Ltd. China Minmetals Non-Ferrous Metals Co. Ltd. China Network Communications Group Corporation China Satellite Launch & Tracking Control General China United Telecommunications Corporation China Netcom Group Corporation (Hong Kong) Limited CITIC Group Datang Telecom Technology & Industry Holdings Limited DIC SAHIR Limited Dubai Investment Group L.L.C. Dubai World Emirates International Capital Advisory GIC Infrastructure Pte Ltd. GIC Real Estate, Inc., GIC Special Investments, Pte. Ltd. GIC Investment Corporation, Pte. Ltd. Huachen Automotive Group Holdings Company Limited Huizhou Municipal Government Investcorp S.A. Investment Corporation of Dubai Istithmar PJSC Jumeirah Assets LLC Kingdom of Norway, Ministry of Petroleum & Energy Kingdom Holding Lenovo Group Limited Moskovskaya Telecommunikatsionnaya Corporatsiya Mubadala Development Company PJSC Norges Bank Norsk Hydro ASA Nye Telenor East Invest AS Qatar Investment Authority Qatar Telecom

SWF SOE SOE SWF SOE SOE SOE SOE SOE SOE SOE SOE SWF SOE SWF SWF SWF SWF SWF SWF SWF SWF SOE SWF SOE SWF SWF SWF SWF SWF SOE SOE SWF SWF SOE SOE SWF SOE

242

ELVIRA SOJLI AND WING WAH THAM

Table A1. (Continued ) Acquirer

Status

Singapore Technologies Engineering Ltd Singapore Technologies Pte Ltd Singapore Technologies Semiconductors Pte Ltd Singapore Technologies Telemedia Pte Ltd Singapore Telecommunications Ltd Sing Sat Pte Ltd. Telenor East Invest AS Temasek Holdings (Private) Limited The Government of the Bolivarian Republic of Venezuela The Government of the Republic of Indonesia

SOE SOE SOE SOE SOE SOE SOE SWF SWF SWF

Notes: The table presents the sample of government-related investors identified in the Schedule 13 filings.

Table A2.

Investment Objectives and Characteristics. Events

Panel A:. Objective categories 1. General undervaluation/Maximize shareholder value 2. Business strategy 3. Sale of target company 4. Governance 5. Capital structure

Percentage of Sample

30

51.7

14 11 6 5

24.1 19.0 10.3 8.6

13D

13G

– 11.4 21.4 42.9 1.4 1.4 5.7 12.9

2.6 2.6 18.4 7.9 – 10.5 13.2 42.1

Panel B: Industry groups (in %)

Agriculture Mining Manufacturing IT & Telecom Wholesale trade Retail trade Finance Business services

243

SWF Investments in the United States

Table A2. (Continued ) Panel C: Invested capital 13D Percentile 5% 25% 50% 75% 95%

13G

Share (%)

Size (in million USD)

Share (%)

Size (in million USD)

5.0 7.6 22.6 42.2 70.6

5.0 18.4 101.2 298.0 1913.7

5.1 5.6 7.6 10.4 30.5

3.1 29.3 64.1 125.1 1746.2

Panel D: Country investment shares 13D (in %) Norway Singapore China Saudi Arabia Qatar Indonesia United Arab Emirates Bahrain

13.8 19.0 17.2 13.8 1.7 1.7 20.7 12.1

13G (in %) 8.6 57.1 0.0 17.1 2.9 0.0 2.9 11.4

Panel E: Year distribution Year

13D

13G

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

6 2 2 1 0 8 4 4 6 4 12 9

3 3 2 8 1 0 3 3 5 1 3 6

Notes: The sample of U.S.-listed companies includes 58 13D and 35 13G events. Panel A summarizes the objectives of SWF investors as stated in 13D filings, where the categories are defined as in Brav et al. (2008). Categorization is not mutually exclusive. Panel B presents the industry focus of SWF investments reported under Schedule 13D and 13G. Panel C provides the distribution of the initial investment share in the target company and the size of these investments on the filing day. Panel D presents the distribution of the country of origin of the investor. Panel E presents the distribution of the investments across the sample period.

WHAT DO SOVEREIGN WEALTH FUNDS IMPLY FOR FINANCIAL STABILITY? Tao Sun and Heiko Hesse STRUCTURED ABSTRACT Purpose – Study the potential implications of sovereign wealth funds (SWFs) on financial stability. Methodology/approach – By assessing whether and how stock markets react to the announcements of investments and divestments to firms by SWFs, this chapter takes advantage of a hand-collected database on investments and divestments by major SWFs to evaluate the short-term financial impact of SWFs on selected public equity markets in which they invest. Findings – Results show that there was no significant destabilizing effect of SWFs on equity markets, which is consistent with anecdotal evidence. Social implications – SWFs could promote financial stability and should be given more development space. Originality/value of the chapter – This study contributes to the emerging academic literature that seeks to analyze the behavior of SWFs in financial markets.

Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 245–262 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012012

245

246

TAO SUN AND HEIKO HESSE

Keywords: Sovereign wealth funds; financial stability; event study JEL classifications: G14; G15; G18; G30

INTRODUCTION Since the beginning of the financial crisis in the summer of 2007, financial stability has been at the forefront of policy discussions. At the same time, sovereign wealth funds (SWFs) have become dominant players, as they have injected significant capital in major financial institutions. There also has been a large focus on the objectives and governance principles of SWFs including the consequences of their investment decisions, which culminated into a multilateral effort and the Santiago Principles in 2008, which has enhanced the credibility of SWFs. Recently in some countries, SWFs were instructed by their governments to invest into domestic financial institutions (e.g., bank recapitalizations) and the overall stock market to support battered stock prices or have supported domestic deposit insurance schemes. Some SWFs with a clear stabilization purpose have, following their mandates, financed fiscal deficits or even bolstered stimulus packages. Research on the financial stability implications of these funds has been slowly emerging, hampered by lack of data on their asset allocations. Although SWFs have become significant institutional investors in global financial markets in recent years with an estimated $3.7 trillion of assets under management in 2010 according to Kern (2010), the overall SWF size is still small with one-sixth of the investment-fund industry and 4 percent of bank assets. For instance, pension funds with $32 trillion, reserve excl. gold with $9.6 trillion, or insurance companies with $21 trillion dwarf the size of SWFs. But nevertheless, SWFs are expected to significantly grow in the forthcoming years spurred by favorable commodity prices that, for instance, benefit SWFs of oil-exporting countries as well as due to capacity constraints in these countries to absorb such large incomes and hence the need to diversify with investments abroad. There have been many arguments put forth regarding the potential positive and negative effects of SWFs on global financial markets. For example, some argue that SWFs can play a stabilizing role in global financial markets. First, many commentators point out that as long-term investors with no imminent call on their assets, and with mainly unleveraged positions, SWFs are able to sit out longer during market downturns or even trade against market trends. In addition, SWFs in some countries, particularly in the Middle East, have

What Do Sovereign Wealth Funds Imply for Financial Stability?

247

recently supported domestic equity markets and financial institutions to boost investor confidence. Second, large SWFs may have an interest in pursuing portfolio reallocations gradually so as to limit adverse price effects of their transactions. Third, SWFs could, as long-term investors and by adding diversity to the global investor base, contribute to greater market efficiency, lower volatility, and increased depth of markets. Fourth, SWF investments may enhance the depth and breadth of markets. Although SWFs appear to have been a stabilizing force thus far, given their size, there are circumstances in which they could cause volatility in markets. Having large and often intransparent positions in financial markets, SWFs – like other large institutional investors – have the potential to cause a market disturbance. For instance, actual or rumored transactions may affect relative valuations in particular sectors and result in herding behavior, adding to volatility. Deeper markets, such as currency markets, can also be affected, at least temporarily, by rumors or announcements about changes in currency allocations by central banks or SWFs. To the extent that SWFs invest through hedge funds that rely on leverage or are subject to margin requirements, such investments may inadvertently magnify market changes. For markets to absorb flows from any major investor class without large price fluctuations, it helps if they can anticipate the broad allocation and risk-preference trends of such investor classes. Opacity about such trends can lead to inaccurate pricing and volatility. As regards these financial stability implications of SWFs, both theoretical and empirical research has begun to be implemented. Recent capital injections by SWFs in financial institutions have intensified the debate on the impact on financial stability. SWFs from East Asia and the Middle East were frequently in the news, as major mature market financial institutions required additional capital. In total, SWFs have reportedly contributed more than $50 billion of such capital since November 2007. The capital injections by SWFs have augmented the recipient financial institutions’ capital buffers and have been helpful in reducing various firm-specific risk premia, at least in the short term, as the injection curtailed the need to reduce bank assets to preserve capital. The announcements of capital injections from SWFs have assisted in stabilizing share prices and the elevated Credit Default Swap (CDS) spreads, at least over the short run (Global Financial Stability Report, April 2008). In most cases, after the announcement of new capital injections, the initial share price reaction to the SWF investments was positive, with announcements of asset write-downs offset by hand-in-hand capital injections from investor groups in which the SWF had a significant role. Although other factors are not taken into account, this initial evidence supports the view that SWFs could have a volatility-reducing impact on markets.1

248

TAO SUN AND HEIKO HESSE

The global financial crisis has hit many of the SWFs quite hard. Portfolio losses have been higher for SWFs that have a higher share of equities in their investment portfolio or large illiquid positions in private equity or hedge funds. Given that SWFs typically have a fairly long investment horizon, they are likely to sit out these unrealized losses. According to Kunzel, Lu, Petrova, and Pihlmann (2011), portfolio losses were as high as 30 percent for some SWFs in 2008 (Fig. 1). The rebound in stock markets since early 2009 has provided SWFs with a buffer and those SWFs that have bought countercyclically have even benefited. In particular, Fig. 1 shows that for savings, pension reserve, and reserve investment funds, 2009 has been a significant rebound in SWF returns from the heavy losses in 2008. Because of the unrealized and realized portfolio losses during 2008, together with some SWF rescuing domestic companies in troubles and the general market volatility, SWFs have become more cautious in their investment approach. In addition, a stabilizing impact of SWFs can be seen from the fact that despite the wide ramifications of the global financial crisis, assets under managements have actually recovered and at times even increased for the different types of SWF except stabilization funds, as Fig. 2 illustrates. This chapter, using an event study approach and based on a handcollected database, endeavors to deepen the analysis of SWFs’ impact on financial stability by differentiating scenarios, including investments and Stabilization/Savings Funds

8 6 4 2 0

Chile-ESSF Timer-Leste Trinidad and Azerbaijan Tobago 2007 30

2008

Pension Reserve Funds

10 0 Chile-PRF

Australia New Zealand

Savings Funds

Norway

Canada

2007

2009

20

-10

50 40 30 20 10 0 –10 –20 –30 –40

Ireland

2008

2009

Reserve Investment Funds

20 15 10 5 0 –5

United States Temasek

China

Korea

–10 –15 –20

-20 -30 2007

Fig. 1.

2008

2009

2007

2008

2009

SWF Returns, 2007–2009. Source: Kunzel et al. (2011).

249

700 Index, December 2007=100

180

140

100 Stabilization Funds 60

Russia-RF Chile-ESSF

Azerbaijan

400

Botswana

300 200 100

07 M ar -0 8 Ju n08 Se p08 D ec -0 8 M ar -0 9 Ju n09 Se p09 D ec -0 9

ec -

07 M ar -0 8 Ju n0 Se 8 p08 D ec -0 8 M ar -0 9 Ju n09 Se p09 D ec -0 9

D

ec D

340 Index, December 2007=100

140

100

Savings Funds 60

Canada USA Norway

20

Pension Reserve Funds Australia Chile-PRF Russia-NWF New Zealand Ireland

300 260 220 180 140 100 60

Fig. 2.

9 -0 ec

D

9

09 p-

Se

9

-0 Ju n

8

-0 ar

M

08

-0 ec

D

8

8

Se

p-

-0

-0

-0 ar

M

Ju n

7

9 -0 ec

D

9

09 p-

Se

9

-0 Ju n

8

-0

-0

ar

ec

M

D

8

08

-0

pSe

Ju n

-0 ar

M

ec

-0

8

7

20 ec

Index, December 2007=100

Timor Leste

500

0

20

D

Stabilization and Savings Funds

600

D

Index, December 2007=100

What Do Sovereign Wealth Funds Imply for Financial Stability?

SWF Assets under Management, December 2007–December 2009. Source: Kunzel et al. (2011).

divestments in advanced and emerging economies, financial and nonfinancial sectors, and higher and lower level of corporate governance. The overall findings suggest that there is no significant destabilizing effect of SWFs on equity markets. This empirical study contributes to the emerging academic literature that seeks to analyze the behavior of SWFs in financial markets. The chapter proceeds as follows: the second section briefly reviews the literature and some conceptual issues. The third section outlines an event study approach and describes data. The fourth section presents empirical results. The fifth section concludes.

LITERATURE REVIEW SWFs are defined as special-purpose investment funds or arrangements owned by the general government. They are often established out of balance of

250

TAO SUN AND HEIKO HESSE

payments surpluses, official foreign currency operations, proceeds of privatizations, fiscal surpluses, or receipts resulting from commodity exports. Their total size has been estimated at $2 trillion–$4 trillion.2 The edited book Economics of Sovereign Wealth Funds by Das, Mazarei, and van der Hoorn (2010) provides a comprehensive overview of the issues by pulling together the leading academics, policy-makers, and practitioners in this area. There has been some empirical research, using equity market indicators and an event study approach to examine the role of SWFs as major institutional investors. Bortolotti, Fotak, Megginson, and Miracky (2009) based on the Monitor Group database of SWF transactions find a positive short-run announcement effect of SWF investments and negative long-run abnormal returns. Dewenter, Han, and Malatesta (2010) and Knill, Lee, and Mauck (2009) obtain similar results. In particular, Dewenter et al. (2010) find abnormal returns of 1.5 percent for share purchases and 1.4 percent for divestments. Interestingly, these magnitudes are related to the transaction size in a nonlinear fashion. For instance, larger investment transactions are characterized by higher abnormal returns to a certain point when the effect reverses. Knill et al. (2009) also analyze the risk/return relations of the target firm and the relevant market as well as their volatility. Their findings suggest that after a SWF investment, the returns of the target firm and the market decline while their volatility also falls. In another in an event study, Chhaochharia and Laeven (2008) find that the announcement effect of SWF investments is positive. They report that share prices of firms respond favorably when SWFs announce investments, in part because these investments happen when their targets are in financial distress. But the long-run performance of equity investments by SWFs tends to be poor (see Fotak, Bortolotti, & Megginson, 2008, for similar results). Another event study analysis by Kotter and Lel (2008) show that the cumulative abnormal return of SWF investments has an announcement effect similar to that of investments by hedge funds and institutional investors, such as CalPERS on stock returns. In addition, investments by more transparent SWFs have a larger cumulative abnormal return by an order of 3.5, suggesting that voluntary SWF disclosure might serve as a signaling device to investors. In addition, Kotter and Lel (2008) also obtain a significant negative but small announcement impact from SWFs’ divestures. Beck and Fidora (2008) conduct a country case study of Norway’s SWF and ask whether its exclusion of companies that violate the ethical guidelines of the Ministry of Finance exhibit price pressures on those companies. Their findings suggest no significant negative abnormal returns following the divesture of these companies.

What Do Sovereign Wealth Funds Imply for Financial Stability?

251

Given the lack of publicly available data on SWF asset allocations, another strand of research has been on the theory side. Lam and Rossi (2010) develop a theoretical model that aims to examine the impact of SWFs on global financial stability during periods of stress. Their findings indicate that SWFs have a risk-sharing role in financial markets. As part of the International Monetary Fund (IMF)-coordinated process of the Santiago Principles that provide generally accepted principles and practices for SWFs, Hammer, Kunzel, and Petrova (2008) examine the asset allocation and risk management frameworks of SWFs based on a detailed survey. The results show that SWFs have specific investment objectives in place, adopt an asset approach (mean-variance style) in determining their asset allocation strategy, utilize common risk measures (e.g., credit ratings, value-at-risk models, tracking errors, duration, and currency weights) for their risk management, and have explicit limits in their investment classes and instruments. Simulations of SWFs’ asset allocations have been undertaken by Kozack, Laxton, and Srinivasan (forthcoming). Specifically, they create two stylized diversified portfolios, one mimicking Norway’s SWF and the other representing some well-established SWFs, and they conduct a scenario analysis of the impact from a further diversification of sovereign assets. Although the calibrations are highly sensitive to the underlying model assumptions, the findings indicate that advanced economies will see lower capital inflows, while emerging market countries will be the primary beneficiaries. Their quantitative results are consistent with the back-of-the envelope calculations of Beck and Fidora (2008), which imply a net capital outflow from the United States and the euro area and net inflows to emerging market countries over the medium term. In the same vein, Jen and Miles (2007) and Hoguet (2008) points out that there is scope for the global equity risk premium to fall and for real bond yields to rise if SWFs allocate their assets to equities. In addition, as SWFs increasingly diversify into global portfolios, their activities may place some downward pressure on the dollar as they exit dollar-denominated assets. To summarize, existing research on SWFs suggests that they can be a stabilizing force in global financial markets. Event studies do not find a destabilizing impact from SWF investments and divestments in equity markets, whereas simulations of SWF asset allocations only imply a gradual shift with modest economic effects. With SWFs improving their transparency and disclosure over time, the availability of historical SWF transactions would provide researchers with the necessary data to further examine their implications for financial stability.

252

TAO SUN AND HEIKO HESSE

DATA AND METHODOLOGY This empirical research assesses whether stock markets react to the announcements of investments and divestments to firms by SWFs using an event study approach. The objective is to investigate the information content of these announcements. On the basis of 166 publicly traceable hand-collected events of investments and divestments by major SWFs during 1990–2009, this section evaluates the short-term financial impact of SWFs on selected public equity markets in which they invest. Moreover, the impact will be further analyzed on different sectors (financial and nonfinancial), actions (buy and sell), market types (developed and emerging markets), as well as level of corporate governance (higher and lower levels). The results are expected to give some hints on how stock markets react to the capital investments and divestments by SWFs and present some implications on SWFs’ stabilizing role in global financial market. Investigating divestments is of particular interest because if stock price reactions are abnormally high (relative to the market) there may be destabilizing effects to the degree that others ‘‘front run,’’ ‘‘herd,’’ or otherwise mimic SWFs’ investment behavior. This might be particularly problematic if prices slip below predefined target levels of other investors and thus prompting their forced sales.

Data Several SWFs that have bought or sold shares of firms in the advanced and emerging stock markets are included in the study. Among them are Abu Dhabi Investment Authority, China Investment Corporation, Government of Singapore Investment Corporation, Kuwait Investment Authority, Korea Investment Corporation, Libyan Investment Authority, Mubadala, Qatar Investment Authority, and Temasek. The source of information on the events is SWFs’ web sites and various financial news and reports such as Factiva. Target firm actual total returns (and price indices) and country stock market returns (and price indices) are obtained from Datastream International database.3 This search results in a total of 166 investment/ divestment events in 115 unique firms, with some firms receiving multiple SWF investments through time between 1990 and 2009. This sample is then combined with firm-level and country-level data collected from Bloomberg and SWF-specific data from various sources including Truman (2008a, 2008b).4

What Do Sovereign Wealth Funds Imply for Financial Stability?

Table 1. Country Australia Austria China Egypt France Germany Iceland India Indonesia Italy Japan Malaysia Pakistan Philippines Portugal Singapore South Korea Spain Sweden Switzerland Taiwan Province of China Thailand UK US Vietnam Total

253

Country of Target Firms. Number 6 1 17 2 8 7 1 13 5 6 2 7 4 1 2 22 3 3 2 2 1 2 31 17 1 166

Table 1 describes the number of SWF investments and divestments across the country of target firms, whereas Table 2 displays the distribution of the sample by the identity of the acquiring SWF. Given public availability of individual buy and sell transactions, observation numbers by Abu Dhabi Investment Authority and the two Singapore SWFs GIC and Temasek are dominating the sample. In addition, the United Kingdom ranks the first as the target in the advanced economies that have seen the most SWFs transactions, with China and India being among the top in the emerging economies. Fig. 3 shows the ratios on SWFs’ investments/divestments in full sample as well as in subsamples – in financial and nonfinancial sectors, in developed and emerging markets, and by SWFs with different levels of corporate governance. Regarding the detailed classifications of the

254

TAO SUN AND HEIKO HESSE

Table 2. SWF

Acquiring SWFs. Number of Observations

Country

26 11 38

United Arab Emirates China Singapore

14 1 2 2 23 49 166

Kuwait Korea Libya United Arab Emirates Qatar Singapore

Abu Dhabi Investment Authority (ADIA) China Investment Corporation (CIC) Government of Singapore Investment Corporation (GIC) Kuwait Investment Authority (KIA) Korea Investment Corporation (KIC) Libyan Investment Authority (LIA) Mubadala Qatar Investment Authority (QIA) Temasek Total

investments/divestments, several observations stand out: first, investments account for over 80 percent of SWFs transactions; second, investments/ divestments mostly took place in the nonfinancial sector; third, the transactions in the emerging economies are almost the same as those in the advanced economies; fourth, the SWFs with higher level of corporate governance have implemented most of the transactions.

Methodology If markets are rational, the effects of an event should be reflected immediately in stock returns and prices. Thus a measure of the event’s impact can be constructed using stock prices and returns observed over a relatively short time period. To benchmark the returns of the stock relative to the event, the overall stock market returns, in percentage changes, for the corresponding country are used. Specifically, the following steps are taken for implementing the event study:  Determination of the selection criteria for the inclusion of given SWFs. The sample contains several SWFs, which have bought or sold stakes in financial firms and nonfinancial firms.  Collection of a number of such events and compilation of a list of firms and dates by searching publicly available databases to find news announcements on SWFs’ actions.

255

What Do Sovereign Wealth Funds Imply for Financial Stability? 90 80 70 60 50 40 30 20 10

Sell by low level

Buy by low level

Sell by high level

Buy by high level

Buy and Sell in Emerging

Buy and Sell in Developed

Buy and Sell in Non-financial

Buy and Sell in Financial

Sell

Buy

0

Fig. 3. Ratios of SWF Investments and Divestments. Notes: (1) The SWFs with high-level corporate governance refer to those whose total score is higher than 40, whereas low level refer to lower than 40 (Truman, 2008a, 2008b); (2) the ratios are calculated separately on the following four subgroups: (i) buy and sell; (ii) buy and sell in financial sector, buy and sell in nonfinancial sector; (iii) buy and sell in developed economies, buy and sell in emerging economies; and (iv) buy by high-level governance, sell by high-level governance, buy by low-level governance, and sell by low-level governance. Source: IMF staff estimates.

 Identification of the events of SWFs’ investments/divestments. Since the event date can be determined with precision, as regards to the short-term analysis, we employ a five-day (seven-day) event window, composed of two (three) pre-event days, the event day, and two (three) post-event days. In this way, rumors that precede the formal announcement can enter the assessment. And as well, in illiquid markets, prices may take a couple of days to adjust to new information. As robustness tests, we vary the event window to four pre-event days, the event day and four post-event days.  Definition of the ‘‘estimation window’’. Following Peterson’s framework (1989), we will estimate the market model on the 200 trading days ending 30 days before the announcement of the investments/divestments. Ending the sample before the event assures that the ‘‘normal’’ behavior of returns

256

TAO SUN AND HEIKO HESSE

is not contaminated by the event itself. For robustness tests, we vary the estimation periods (100 days and 300 days) and using price indices instead of total returns of each firms and economy. Prediction of a ‘‘normal’’ return during the event window in the absence of the event, using a one-factor OLS regression equation:5 rit ¼ ai þ bi rmt þ eit where rit is the percentage change of returns of the stock relative to the event, rmt is the percentage change of overall stock market returns, ai and bi are regression coefficients, and the eit is an error term.  Calculation of the abnormal return within the event window. Having calculated estimates of a´ i and aˆ i with the data from the estimation period, we calculate the abnormal returns by differencing the actual and estimated returns, ARit ¼ Rit  Rit ¼ Rit  ða þ bi Rmt Þ where Rit is the estimated return. Specifically, the abnormal return observations must be aggregated to draw overall inferences for the event of interest. The aggregation can be along two dimensions – through time and across securities. The individual securities’ abnormal returns, in the case of five days, can be aggregated for each event day, t ¼ t2, t1, t, t þ 1, t þ 2 during the event window. Given N events (a total of 166 in the entire sample), the sample average aggregated abnormal returns (AAR) for period t is AARt

N 1X ARit N i¼1

The average abnormal returns can then be aggregated over the event window to calculate the cumulative average abnormal return (CAAR) for each firm i. CAARt ¼

2 X

AARt

t¼2

 Testing whether the abnormal return is statistically different from zero. As the numbers of observation in the event window are limited (five or seven days), we use t-tests rather than the Z-score, the latter usually requiring at least 50 observations to get a statistically robust results.6

What Do Sovereign Wealth Funds Imply for Financial Stability?

257

EMPIRICAL RESULTS Table 3 presents the AAR and CAAR for the (2, þ 2), and (3, þ 3) windows using price indices. In general, the AAR is positively associated with SWFs’ buy actions and not significantly negatively with SWFs’ sell actions in the full sample. Moreover, overall, the results suggest that the share price’s combined responses to SWFs’ investments and divestments in developed economies are significant (panel C), whereas those in emerging economies are not (panel D). In addition, SWF investments in the financial sector have a larger impact on share prices than in the nonfinancial sector. These differences in responses may be due to the relatively more transparent equity markets in developed economies as well as in the financial sector with potentially higher signaling and information flow. Different scenarios are tested using these events. Specifically, panel A of Table 3 reports the AAR and CAAR around the announcements of SWF investments for the entire sample of 134 observations during the period between 1990 and 2009. The AAR is 0.26 percent and 0.22 percent for windows of (2, þ 2), and (3, þ 3) around the announcement date, and the CAAR is 0.75 percent and 0.98 percent, respectively. The sign teststatistics for the AAR are also highly significant for the two windows. Panel B reports the AAR and CAAR around the announcements of SWF divestments for the entire sample of 32 observations during the period between 1990 and 2009. The AAR is 0.07 percent and 0.03 percent for the windows (2, þ 2), and (3, þ 3) around the announcement date, and the CAAR is 0.26 percent and 0.24 percent, respectively. The sign test-statistics for the AAR and the CAAR are insignificant for the two windows. Panel C reports the AAR and CAAR around the announcements of SWF investments and divestments for the developed economy sample of 87 observations during the period between 1990 and 2009. The AAR is 0.25 percent and 0.20 percent for the windows (2, þ 2), and (3, þ 3) around the announcement date, and the CAAR is 0.87 percent and 1.10 percent, respectively. The sign test-statistics for the AAR and the CAAR are highly significant for the two windows. Panel D of Table 3 reports the AAR and CAAR around the announcements of SWF investments and divestments for the emerging economy sample of 79 observations during the period between 1990 and 2009. The AAR is 0.14 percent and 0.10 percent for the windows (2, þ 2), and (3, þ 3) around the announcement date, and the CAAR is 0.38 percent and 0.20 percent, respectively. The sign test-statistics for the AAR and CAAR are insignificant for the two windows.

258

TAO SUN AND HEIKO HESSE

Table 3. Event Window

Stock Market Reactions to Announcements of SWF Investments and Divestments (Price Indices).

t-Statistic of AAR

Mean of AAR

t-Statistic of CAAR

Mean of CAAR

Panel A: Buy only, 134 events from 101 firms 0.26 (2, þ 2) 4.09 0.22 (3, þ 3) 3.84

3.46 4.71

0.75 0.98

Panel B: Sell only, 32 events from 23 firms (2, þ 2) 0.24 0.07 (3, þ 3) 0.16 0.03

1.32 1.67

0.26 0.24

Panel C: Buy and sell in developed economies only, 87 events from 55 firms 0.25 5.45 (2, þ 2) 5.05 0.20 6.37 (3, þ 3) 2.83

0.87 1.10

Panel D: Buy and sell in emerging economies only, 79 events from 60 firms (2, þ 2) 0.98 0.14 1.58 (3, þ 3) 0.94 0.10 1.33

0.28 0.20

Panel E: Buy in developed economies only, 72 events from 51 firms 0.31 4.44 (2, þ 2) 5.5 0.25 5.66 (3, þ 3) 3.22

0.95 1.23

Panel F: Sell in developed economies only, 15 events from 9 firms (2, þ 2) 0.12 0.04 1.02 (3, þ 3) 0.20 0.05 0.67

0.28 0.14

Panel G: Buy in emerging economies only, 62 events from 50 firms (2, þ 2) 1.94 0.21 2.37 0.19 3.41 (3, þ 3) 2.58

0.53 0.69

Panel H: Sell in emerging economies only, 17 events from 14 firms (2, þ 2) 0.47 0.17 1.05 (3, þ 3) 0.40 0.10 1.80

0.24 0.32

Panel I: Buy in financial sector only, 41 events from 24 firms 0.66 (2, þ 2) 2.46 0.65 (3, þ 3) 2.91

5.45 6.82

2.45 3.42

Panel J: Sell in financial sector only, 5 events from 3 firms (2, þ 2) – – (3, þ 3) – –

– –

– –

Panel K: Buy in nonfinancial sector only, 93 events from 77 firms (2, þ 2) 0.35 0.04 1.56 (3, þ 3) 0.10 0.01 3.83 Panel L: Sell in nonfinancial sector only, 27 events from 20 firms (2, þ 2) 0.24 0.07 1.32 (3, þ 3) 0.16 0.03 1.67

0.17 0.35 0.26 0.24

259

What Do Sovereign Wealth Funds Imply for Financial Stability?

Table 3. (Continued ) Event Window

t-Statistic of AAR

Mean of AAR

t-Statistic of CAAR

Panel M: Buy by high level in governance only, 76 events from 59 firms (2, þ 2) 0.22 0.04 0.50 (3, þ 3) 0.11 0.02 0.68 Panel N: Sell by high level in governance only, 26 events from 19 firms (2, þ 2) 0.38 0.12 0.92 (3, þ 3) 0.27 0.06 0.91

Mean of CAAR

0.07 0.06 0.18 0.14

Panel O: Buy by low level in governance only, 58 events from 45 firms 0.69 3.04 (2, þ 2) 2.72 0.51 4.07 (3, þ 3) 2.26

1.91 2.26

Panel P: Sell by low level in governance only, 6 events from 4 firms (2, þ 2) 0.40 0.23 1.21 (3, þ 3) 0.32 0.13 1.64

0.75 0.88

Source: IMF staff estimates. Note: As there are no qualified observations before/after the corresponding event dates, there are no results for the group of ‘‘sell in financial sector only (Panel J).’’

The impact is further analyzed on the investments/divestments separately in different market types (developed and emerging markets), different sectors (financial and nonfinancial), and level of corporate governance (high and low). In general, according to the AAR, investments in developed economies (panel E) and emerging economies (panel G) are statistically significant, whereas divestments in developed economies (panel F) and emerging economies (panel H) are generally statistically insignificant. These demonstrate that SWF investments produce positive impact in both developed and emerging economies, whereas their divestments led to little negative impact.7 In addition, the positive impact of ARR and CAAR for the investments by low-level governance SWFs is significantly larger than those by high-level governance SWFs because the investment/divestment behaviors of low-level governance SWFs may be more speculative and unexpected, thus triggering larger market impact upon the announcement of their actions. This is in line with the idea that transparency matters. This could also indicate that the improvement of corporate governance in SWFs would be helpful in reducing the impact on market volatility.8 As a robustness check, we use the event window of (4, þ 4) to test the impact of SWFs’ actions. In addition, we vary the estimation periods (100 and 300 days). The results are robust to different event windows and the estimation periods.

260

TAO SUN AND HEIKO HESSE

CONCLUSION This chapter assesses whether and how stock markets react to the announcements of investments and divestments to firms by SWFs using an event study approach. On the basis of 166 publicly traceable events collected on investments and divestments by major SWFs during the period of 1990–2009, we evaluate the short-term financial impact of SWFs on selected public equity markets in which they invest. The impact is further analyzed on different sectors (financial and nonfinancial), actions (buy and sell), market types (developed and emerging markets), countries, and level of corporate governance (high and low). Overall, this event study does not find any significant destabilizing effect of SWFs on equity markets as measured by abnormal return behavior, which is consistent with the emerging academic literature that uses the event study methodology. This study contributes to the slowly emerging field of empirical studies of SWF behavior in financial markets. However, it should be noted that the longer-term impact and the potentially stabilizing role of SWFs as major institutional investors will require a broader set of data and a more rigorous empirical assessment. The long-run impact of SWF investments could be subject to the macroeconomic and financial conditions. In the case of recent investments in some U.S. and European financial institutions under conditions of distress, SWFs’ action could not buffer those institutions from further large losses. Therefore, it will be hard to draw conclusions for overall global and regional financial stability only from these 166 events. Other methods to examine the empirical impact of SWFs would require more detailed knowledge of SWF investments and their timing and amount – data that is presently not available. Some progress may be possible with hypothetical scenarios, but hypothetical market responses to SWF investments require a thorough understanding of how asset allocations are constructed and the size, depth, and breadth of the corresponding markets.

NOTES 1. With the continuing increase in banks’ losses and write-downs during the subprime crisis, the rescue of Bear Stearns, collapse of Lehman Brothers and U.S. government intervention into major financial institutions, the longer-term share price development of banks that obtained initial capital injections from various SWFs, has been obviously very negative. But the short-term reaction of SWFs financial support has been perceived as very supportive by the financial market in most cases. 2. A new report by International Financial Services London has revealed that sovereign wealth funds total assets increased 18 percent to $3.9 trillion in 2008 from

What Do Sovereign Wealth Funds Imply for Financial Stability?

261

$3.3 trillion in 2007. Total assets are now contracted to reach $8 trillion by 2015, down from their $10 trillion estimated in 2008. 3. Datastream is the only data vendor that provides total return stock market indices for all the relevant countries, correcting index returns for the implications of dividend payments, stock splits, and other such changes. 4. The score of each SWF is from the ‘‘total’’ score of Truman (2008a, 2008b). We take those higher than 40 as ‘‘high,’’ whereas those lower than 40 as ‘‘low’’ in the econometric analysis. 5. As the ‘‘market model’’ is most commonly used to generate expected returns and no better alternative has yet been found despite the weak relationship between beta and actual returns (Armitage, 1995), we use the market model to predict ‘‘normal’’ return. To test for robustness, a three-factor model could also be employed. 6. The t-test is of interest because it can accommodate the differences of the abnormal returns over time and especially across types of markets. The event study approach shows the explicit impact of SWF actions, because the methodology is based on individual purchases and sales of publicly available equities. 7. Although the combined impact of investments and divestments in emerging economies (panel D) is insignificant, the impact of investment in emerging economies is significant (panel G). The reason could be the individual impact of investments was offset by the divestments when both actions are jointly tested. 8. This is in line with the positive market responses to the investments in the entire sample. The reason is that SWFs with low level of corporate governance accounts for the majority sample of SWF investments.

ACKNOWLEDGMENTS Many thanks are due to Laura Kodres for her guidance and suggestions, Udaibir Das, and to participants at an IMF Seminar. Oksana Khadarina provided research support.

REFERENCES Armitage, S. (1995). Event study methods and evidence on their performance. Journal of Economics Surveys, 8, 25–52. Beck, R., & Fidora, M. (2008). The impact of sovereign wealth funds on global financial markets. ECB Occasional Paper Series No. 91. European Central Bank, Frankfurt. Bortolotti, B., Fotak, V., Megginson, W. L., & Miracky, W.F. (2009). Sovereign wealth fund investment patterns and performance. Unpublished. University of Oklahoma. Chhaochharia, V., & Laeven, L. (2008). Sovereign wealth funds: Their investment strategies and performance. CEPR Discussion Paper No. 6959. Center for Economic Policy Research, London. Das, U. S., Mazarei, A., & van der Hoorn, H. (2010). Economics of sovereign wealth funds: Issue for policymakers. Washington, DC: International Monetary Fund.

262

TAO SUN AND HEIKO HESSE

Dewenter, K. L., Han, X., & Malatesta, P. H. (2010). Firm value and sovereign wealth fund investments. Journal of Financial Economics, 98, 256–278. Fotak, V., Bortolotti, B., & Megginson, W. (2008). The financial impact of sovereign wealth fund investments in listed companies. Unpublished. University of Oklahoma. Hoguet, G. R. (2008). The potential impact of sovereign wealth funds on global asset prices. Vision, 3(2), 23–30. Hammer, C., Kunzel, P., & Petrova, I. (2008). Sovereign wealth funds: Current institutional and operational practices. IMF Working Paper no. 254. International Monetary Fund, Washington. International Monetary Fund. (2008). Do sovereign wealth funds have a volatility-absorbing market impact? Global Financial Stability Report, April. International Monetary Fund, Washington, DC. Jen, S., & Miles, D. K. (2007). Sovereign wealth funds and bond and equity prices. Morgan Stanley Research, 31 May. Kern, S. (2010). Sovereign wealth funds: New economic realities and the political responses. In: U. S. Das, A. Mazarei & H. van der Hoorn (Eds.), Economics of sovereign wealth funds: Issue for policymakers. Washington, DC: International Monetary Fund. Knill, A., Lee, B.-S., & Mauck, N. (2009). ‘Sleeping with the enemy’ or ‘an ounce of prevention: Sovereign wealth fund investments and market instability. Working Paper. Florida State University, Tallahassee, FL. Kotter, J., & Lel, U. (2008). Friends or foes? The stock price impact of sovereign wealth fund investments and the price of keeping secrets. International Finance Discussion Papers no. 940. Board of Governors of the Federal Reserve System, Washington, DC. Kozack, J., Laxton, D., & Srinivasan, K. (forthcoming). The macroeconomic implications of sovereign wealth funds. IMF Working Paper; Chapter 14 in Economics of sovereign wealth funds: Issues for policymakers, 2011. International Monetary Fund, Washington. Kunzel, P., Lu, Y., Petrova, I., & Pihlmann, J. (2011). Investment objectives of sovereign wealth funds – A shifting paradigm. IMF Working Paper WP/11/19. International Monetary Fund, Washington. Lam, R. W., & Rossi, M. (2010). Sovereign wealth funds – Investment strategies and financial stress. Journal of Derivatives and Hedge Funds, 15, 304–322. Peterson, P. D. (1989). Event studies: A review of issues and methodology. Quarterly Journal of Business and Economics, 28(Summer), 36–66. Truman, E. M. (2008a). A blueprint for sovereign wealth fund best practices. Peterson Institute Policy Brief, Number PB08-3, April. Retrieved from http://www.ciaonet.org/pbei/iie/ 0001182/index.html. Truman, E. M. (2008b). The management of China’s international reserves and its sovereign wealth funds. Paper prepared for the Chinese Academy of Social Sciences Conference Marking the 30th Anniversary of the Reform and Opening-up, Beijing, China, December 16–17, Peterson Institute for International Economics.

AFRICA’S QUEST FOR DEVELOPMENT: CAN SOVEREIGN WEALTH FUNDS HELP? Thouraya Triki and Issa Faye STRUCTURED ABSTRACT Purpose – This chapter discusses the potential role that Sovereign Wealth Funds (SWFs) could play to enhance development in African economies, both as recipient and home countries. Methodology – We use hand collected data on the universe of Africa’s SWFs, their sizes and transparency, and reporting scores to provide a landscape of these funds. We also focus on a sample of investments in Africa made both by African and foreign SWFs to describe the type of interventions these vehicles have been making on the continent. Findings – Our analysis shows that African SWFs are small, suffer from poor governance, and are mainly focused on stabilizing local economies. This suggests that their potential role as long-term institutional investors to foster economic growth is likely to be limited if current practices are maintained. On the other hand, foreign SWFs are increasingly interested in Africa and are poised to play a bigger role in supporting the continent’s growth if the right strategies are implemented.

Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 263–290 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012013

263

264

THOURAYA TRIKI AND ISSA FAYE

Social implications – The chapter identifies opportunities that Africa offers to SWFs as well as the challenges that need to be addressed in order to enhance SWFs’ role in supporting the continent’s development. Originality/value of paper – This chapter provides the first comprehensive landscape of African SWFs while also describing their interventions. It also uses an original data set to describe the geographic and sector distributions of foreign SWFs investments in Africa. Keywords: Sovereign Wealth Funds (SWFs); Africa; institutional investors Jel Classifications: G28; O16; O55

At a time when people are fearful of Sovereign Wealth funds, I’m saying let’s look at this as an opportunity. — (R. Zoellick, World Bank, 2008)

Sovereign Wealth Funds (SWFs) have emerged as potential solutions to actively manage foreign reserves accumulated from commodity sales or strong exports. They correspond to government-owned investment vehicles managed by a state-controlled entity or external managers, on behalf of a nation, to serve primarily medium to long-term economic and financial objectives. Their existence could be traced back to the 1950s when Kuwait established in 1953 a SWF to manage its foreign reserves. Impressive growth in the size of SWFs assets and the recent eye-popping cash infusions they made into high-profile Western financial institutions, like Morgan Stanley, Citigroup, UBS, and the Blackstone group, to mitigate the negative effects of the financial crisis, helped spur the phenomenal increase in their popularity. Latest statistics published by Preqin show that SWFs managed USD 4 trillion in assets as of December 2010, 11% more than in 2009, reflecting the start of a global economic recovery.1 OECD expects assets under SWFs management to reach USD 10 trillion by 2015 while Fig. 1 depicts the strong positive association between the value of total assets managed by SWFs and commodities prices over the period 1999–2009. Preqin (2010) estimates that exports of hydrocarbon and other commodities provide respectively 60% and 8% of resources managed by SWFs. Significant revenues from commodities over the last decades had led to the inception of a number of SWFs in Africa, notably in oil-exporting countries (e.g., Libya, Nigeria, and Chad). Botswana (Pula Fund) and Ghana (Minerals Development Fund) pioneered the establishment of

265

Africa’s Quest for Development: Can Sovereign Wealth Funds help? $ Billion 4500

$USD 400

4000

350

3500

300

3000 250 2500 200 2000 150 1500 100

1000

50

500 0

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Year

0

Assets under management Brent Crude Oil Price Annual average DJUBSTR

Fig. 1. Evolution of SWFs Assets as Compared to Oil and Commodity Prices. Source: Authors’ calculation using the 2009 Monitor Group SWF annual report, International Financial Statistics published by the IMF and Dow Jones official website. The line with triangles describes return on the Dow Jones-UBS Commodity Index (DJUBSTR).

African SWFs in 1993. According to our research, the continent counts at least 15 SWFs (Appendix A1). With the notable exceptions of the Libyan Investment Authority (LIA) and the Algerian Revenue Regulation Fund, which rank among the largest 15 SWFs worldwide in terms of size, African funds are dwarfed by their peers from other regions of the world (mainly Asia and the Middle East).

266

THOURAYA TRIKI AND ISSA FAYE

SWFs are often created either to stabilize government fiscal and/or foreign exchange revenues and macroeconomic aggregates by smoothing out fluctuations in export prices and demand, or to save for future generations a fraction of the revenues accruing from the sale of nonrenewable natural resources. There is considerable controversy about the relative merits of SWFs and their value added. Proponents of SWFs argue that these funds can help foster economic growth and prosperity for current and future generations by showcasing successful experiences such as Norway. They also point out that these vehicles can help stabilize the global financial system by providing cross-border liquidity in times of financial turmoil. Opponents, on the other hand, are expressing serious concerns that SWFs would endow governments with too much power, which could move the global economy away from liberalism and impede market forces and competition. A second reservation concerns the possibility that SWFs may threaten national security in the recipient countries if the investments are made for political rather than economic purposes. Such a scenario would trigger a protectionist backlash that could have disastrous effects on the world economy. Where does Africa stand in this debate? To what extent, if at all, SWFs can benefit African economies? Can the controversy discussed above be resolved in the case of Africa? Unfortunately, the literature does not provide clear answers to these questions, as research about SWFs potential support to Africa’s development is rather scant. This largely reflects the strong opacity surrounding SWFs existence, holdings, and institutional arrangements. The objective of this chapter is to improve understanding of SWFs activities in Africa and to discuss the potential role that SWFs could play in African economies, both as recipient countries and home countries. The remainder of the chapter is structured as follow. The first section draws a detailed portrait of African SWFs, providing what we believe is the first comprehensive list of those funds, putting them in perspective, and describing their characteristics and activities. The second section investigates the interventions of foreign SWFs in Africa. The third section discusses the potential benefits of SWFs for African states given the socioeconomic context of those economies. The fourth section discusses issues that may arise from SWFs operations in Africa. The last section concludes by providing some recommendations. The chapter makes several contributions to the debate on the role that institutional investors are poised to play in global capital markets. First, it analyzes how African economies can benefit from SWFs and use them as a

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

267

channel to tap into international financial markets. Second, the chapter documents the size of assets managed by SWFs and describes how and to what extent they can contribute into broadening and deepening African financial systems. This includes discussions on their capacity to mobilize sizable amount of long-term financing and to diversify African financial systems out of the banking sector, through investments in a various set of nonbank financial assets (equity, fixed income securities, real estate, etc.) and institutions (insurance, leasing companies and private equity funds). Last but not least, the chapter examines the very important role that SWFs could play to stabilize the global financial system, through large injections of funds into the global economy. This is documented with reference to investments made by the LIA in selected European (e.g., Italy, United Kingdom, Netherlands–Belgium, and Spain) financial institutions to prevent some of the deleterious effects of the recent global financial crisis.

WHAT DO AFRICAN SWFs LOOK LIKE? According to our research, Africa counts 15 SWFs (Appendix A1). Among the five largest African SWFs, four are sourced from oil and gas revenues, the last being sourced from diamonds, minerals, and other natural resources. These funds were established on a voluntary basis, with the notable exception of Chad’s future generation fund that resulted from the World Bank’s requirement to establish a petroleum revenue management law in Chad as a condition to disburse a loan aimed at funding the Duba oil fields and the Chad–Cameroon pipeline. Strong opacity surrounding their existence, holdings, and institutional arrangements makes tracking of African SWFs a challenging task. A plausible explanation for this limited attention is the relatively small size of African SWFs compared to their counterparts from other regions of the world as well as their passive management strategies. African SWFs Motives It comes out fairly clearly from Appendix A1 that African SWFs are predominantly driven by stabilization motives and to a lesser extent by the need to generate higher returns on domestic resources in order to accumulate wealth for future generations. For most African countries, stabilization needs are twofold. On the short term, African countries need to smooth their expenditures in a context of volatile commodity prices to avoid

268

THOURAYA TRIKI AND ISSA FAYE

challenges in macroeconomic planning resulting from revenue instability (Asfaha, 2007). On the long term, African countries need to protect themselves against decline in revenues resulting from depletion of nonrenewable resources. Moreover, nonrenewable commodities are often the single most important source of foreign currency revenues in these countries which makes them haunted by the paradox of plenty or the so called resource curse. Auty (1993) first introduced this term to describe the potential devastating effects that natural resources could have on economic growth in developing countries, therefore transforming resources from a desirable asset to a curse. The resource curse thesis is based on observations that countries richly endowed with natural resources tend to have lower rates of economic growth and development than countries with fewer natural resources. The resource curse can originate from different causes, including government mismanagement of revenues, weak governance, or the Dutch Disease.2 As shown in Appendix A1, African SWFs are commodity-based and derive their funding from foreign currency resulting from commodity sale. This makes them useful to absorb large surpluses of foreign exchange, avoid inflationary pressures as well as the need for sterilization interventions that could be costly for African countries with prevailing high interest rates. In theory, foreign reserves accumulation through SWFs represents also a ‘‘selfinsurance’’ against capital flight that should favor autonomy in macroeconomic policy (Griffith-Jones & Ocampo, 2010). Available information suggests that African SWFs have been subject to regular capital withdrawals to balance governments’ budgets and repay external debt. For instance, the balance of Nigeria Excess Crude Account (ECA) decreased from USD 20 billion in 2008 to less than USD 3 billion in 2010 while Sudan almost wiped out its Oil Revenue Stabilization Fund (ORSF) (Medani, 2010). Similarly, Algeria has been using its Fonds de Regularisation des Recettes (FRR) to repay public debt and fund fiscal deficits while Mauritania withdrew USD 45 million from its Fonds National des Revenus des Hydrocarbures leaving a balance of USD 34.25 million as of March 2009.3 Such statistics suggest that African governments kept spending while also accumulating resources in their stabilization funds, which may have potentially resulted in zero net savings. This raises concerns about intergenerational equity and long-term fiscal and macroeconomic sustainability, especially in a context of external negative shocks. With few notable exceptions, most African countries have no limitations on the amount that could be used to close budget deficits from commodity-linked revenues.

269

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

Such features have been identified by Asfaha (2007) as common design problems in SWFs. Yet, one might argue that reducing external debt decreases the financial burden on future generations, which is only true if the reduction in debt is permanent and leads to improved economic growth. In the African context, this still needs to be proved.

African SWFs Size The regional distribution of SWFs (Fig. 2) displays a predominance of the Middle East (43%) followed by Asia (36%), and Europe (18%).4 Africabased SWFs have a market share that presumably does not exceed 2%. As of December 2009, African SWFs had USD 114.27 billion in assets under management, much less than their Middle East peers, which held assets amounting to USD 1.41 trillion. Interestingly, African SWFs have experienced a surge from 2008 to 2009 despite decreasing oil prices. Potential explanations for this growth include an increase in the volume of commodity exports, a raise in the share of foreign reserves received by SWFs, or the establishment of new SWFs on the continent. While statistics describing global assets under SWFs management are publicly available, very little information exists on their individual characteristics. According to our estimates, Africa counts 15 SWFs $ billion 2000 1755 1800 1600 1400

2008

1447.42 1287

1409.33

2009

1200 1000 685.62 624

800 600 400 200

117 76.18

0 Middle East

Americas

Europe

Asia

114.27 39

78 76.18

Africa

Other

Fig. 2. SWFs’ Assets under Management by Region. Source: Authors calculations using International Financial Services London Research (2009), and SWFs Institute website.

270

THOURAYA TRIKI AND ISSA FAYE

(Appendix A1), the LIA being the largest with assets amounting to USD 70 billion. Additional SWFs will presumably be launched in African countries including Zimbabwe and Mauritius.5 Similarly, several countries that have already stabilization funds are now considering the establishment of SWFs with savings and development mandates. For instance, Nigeria, which established in 2004 the ECA to insulate the Nigerian economy from boom and bust cycles in oil prices, is expected to launch soon the Nigeria Posterity Fund to stabilize macroeconomic fundamentals, accumulate savings for future generations, and develop critical infrastructure. Thus, the growing number of African funds is likely to increase the share of African SWFs in global SWF assets.

African SWFs Governance Structures So far, public disclosure about assets, strategies, rationale, and structure of African SWFs remains extremely heterogeneous and scarce. This makes governance a main issue to be addressed for African SWFs. Governance encompasses institutional arrangements to report on investments composition and performance, and accountability and transparency measures to ensure prudent management of sovereign resources and independent decision making. Table 1 documents the low level of transparency of African SWFs as measured by the Linaburg-Maduell Transparency Index.6 Moreover, as can be seen from Table 1, out of the 15 African SWFs we were able to identify, only 3 (from Libya, Botswana, and Equatorial Guinea) have signed the Santiago Principles.7 Nevertheless, African signatory countries barely disclose information about their SWFs activities or structure. The transparency of each fund is usually related to the openness of the country’s political system. Thus, setting up corruption-free SWFs in several African countries, known to have opaque political regimes, is very challenging. Governance problems in African SWFs may arise from lack of institutional arrangements. For example, Nigerian finance minister recently announced that the ECA is not backed by a law and that ‘‘the process of accessing the ECA is not as transparent and clear to the Nigerian people, therefore there is a general perception that there is some level of mismanagement.’’8 Governance issues may also arise from poor enforcement of existing institutional arrangements. For instance, Chad amended in 2005 its well-developed national revenue management law in order to increase the share of oil revenues that goes into the budget revenue from 15% to 30%. Later it included defense in the discretionary expenses and canceled the fund

271

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

Table 1.

Transparency Status of African SWFs.

SWF Name

Country

Fonds de Re´gulation des Recettes Reserve Fund for Oil Pula Fund Fund for Future Generations Fonds de Stabilisation des Recettes Budge´taires Fund for Future Generations Fonds de Stabilisation des Recettes Budge´taires Fund for Future Generations Minerals Development Fund Libyan Investment Authority Fonds National des Revenus des Hydrocarbures Minerals Development Fund Excess Crude Account National Oil Account Oil Revenue Stabilization Fund

Santiago Principles Signatory

L-M Transparency Index

Algeria Angola Botswana Chad Congo

No No Yes No No

1 NA 3 NA NA

Equatorial Guinea Equatorial Guinea

Yes No

NA NA

Gabon Ghana Libya Mauritania

No No Yes No

NA NA 2 1

Namibia Nigeria Sa˜o Tome´ and Principe Sudan

No No No

NA 1 NA

No

NA

Source: Authors’ calculation using data from Monitor Group and the SWFs Institute. Data reported for the 1st Quarter 2010. Unfortunately, the value of this index is not available for African SWFs that were not previously identified in the literature.

for future generation (Asfaha, 2007). This type of behavior casts doubt about the quality of governance in African SWFs.

African SWFs Investments Unfortunately, it is very difficult to find accurate information about how and where African SWFs invest their resources. Available data suggest that African SWFs have been adopting prudent investment strategies with an emphasis on liquidity, reflecting mainly their stabilization mandates. For example, a recent IMF report shows that Nigeria’ ECA is mainly invested in short-term, liquid government securities and money market instruments, while research published by JP Morgan (2008) shows that the Pula Fund has invested 59% of its assets in bonds and 13% in cash and restricts its

272

THOURAYA TRIKI AND ISSA FAYE

investments to rated assets. Data also show that African SWFs are actively investing outside Africa. Asfaha (2007) reports that Chad invest its proceeds from natural resources sales abroad while Sao Tome and Principe oil revenue management law prohibits investments in companies controlled by locals (Albin-Lackey et al., 2004). Similarly, Belaicha, Bouzidim, and Labaronne (2009) estimate that half of Algeria’s foreign currency reserves have been invested in US sovereign bonds and deposits and Tier-one banks. It therefore appears that African SWFs are mainly seeking ‘‘safe investments’’ in stable economies leaving limited resources for their local economies, and even less for their neighboring countries. The LIA remains the only SWF that has an active and aggressive investment strategy. LIA was created in December 2006 by a decree of the Comite´ Populaire Ge´ne´ral, with the purpose of consolidating existing investment vehicles, namely the Libyan Arab Foreign Investment Company, the Libyan African Investment Portfolio (LAP), the Long-Term Investment Portfolio and the Oil Investment Company, which have become subsidiaries. Appendix A3 provides a description of LIA’s subsidiaries. Most of LIA’s investments in Africa are undertaken by the LAP. Its subsidiary, the Libyan African Investment Company (LAICO) is present in 30 African countries. We were able to track 114 investments made by LIA over the last 3 decades, out of which 24 are located outside Africa. While this sample describes only part of LIA’s activities, it still provides insightful information about the region and sector distribution of its investments. Fig. 3 shows that West Africa is the main target of LIA investments, followed by East and Central Africa while North Africa and Southern Africa rank at the bottom. However, the value clustering shows a different picture with North Africa capturing USD 9 billion, the highest amount of investments. This probably reflects the stable and business friendly environment offered by North African countries compared to sub-Saharan Africa. The sector distribution of LIA investments in Africa shows a large number of small-scale deals in the real estate, hotels and restaurants, and agriculture sectors as well as a small number of large deals in infrastructure and oil and gas sectors. LIA deals outside Africa targeted mainly companies from Italy and the United Kingdom. Oil and gas and manufacturing captured the largest number of these investments while the financial sector benefited from the highest share of deal values. African SWFs Reputation Cash infusions made by Africa-based SWFs have not always been greeted with alloyed gratitude. A survey conducted by Hill & Knowlton and Penn

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

273

Outside Africa

South Africa

West Africa

East Africa

Central Africa

North Africa 0

Fig. 3.

10

20

30

40

Regional Distribution of Selected LIA Investments. Source: Authors’ calculation.

Schoen Berland (2010) on national officials’ attitudes toward SWFs shows that African SWFs in Libya, Algeria, and Nigeria were ranked less favorably than their Middle Eastern counterparts (Fig. 4). According to this survey, even African host countries like Egypt share this view. The negative perception that African SWFs are suffering from mainly reflects the negative image of African countries and not necessarily wrong doing by these funds, especially that some of them do not invest abroad. This is corroborated by the results of the above-mentioned survey which documents that home country reputation is a major determinant of the image, transparency, and governance structure of a SWF. It should also be noted that most of the non-African SWFs that received better ranking do not necessarily disclose more information about their asset allocation or returns than African funds. Prominent concerns that were expressed by recipient countries include fear from the increased role of the states in financial and economic systems, and the possibility that some African SWFs’ may pursue ‘‘non commercial’’ objectives. Such concerns led the Pentagon to cancel in 1986 a USD 7.9 million contract between the US marine and Fiat because of the LIA

274

THOURAYA TRIKI AND ISSA FAYE

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1

Si

N or w ng ay H ap on or g e Ko n Ku g w ai t Ab Dub u ai D M hab al i ay Ba sia hr ai n Q at R ar us s M ia ex ic Br C o un hi na ei D Om ar u an Ka ssa za lam kh st an Li by Al a Bo ger st ia w a N na ig er ia

0

Fig. 4. Extent of SWFs Investment Approval by Home Country. Source: Sovereign brands survey 2010. The figure summarizes responses to the question: To what extent do you approve or disapprove of SWFs from the following countries investing in your country? (Strongly/somewhat disapprove).

shareholding in the company.9 Africa remains also portrayed by the mass media as a charity case suffering from political violence, corruption, and famine. This cast doubts about the capacity of African SWFs to have value added as investors and the potential negative effects that their presence could have on the transparency and governance structure of beneficiary investees. This negative perception most likely translates into additional barriers to African SWFs activities. Recent turmoil in Libya and allegations about control of LIA resources by political elite are likely to further cast doubts about the legitimacy of African SWFs and showcase the importance of strong governance structures. Nonetheless, the negative perception does not mean that Africa’s SWFs money is not welcome in other regions of the world. Headlines from the business press have reported investments by LIA in European (e.g., Italy and Spain) financial institutions to prevent some of the deleterious effects of the crisis despite allegations about LIA weak governance. In July, 2008, LIA bought a share in the Dutch–Belgian bank of Fortis, which needed additional funds to maintain solvability. More

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

275

recently, the LIA drew public attention when it backed a new London hedge fund (FM Capital Partners) with hundreds of millions of dollars.

WHAT ARE FOREIGN SWFs DOING IN AFRICA? African countries made headlines in the business press as targets for investments by SWFs. Some governments are even creating development funds (China–Africa development funds) or Investment companies (Dubai World Africa) entirely dedicated to Africa. These vehicles are designed to take advantage of the substantial and diverse opportunities the continent is offering given its 900 million young population, its emerging economies and growing middle-income class. The continent natural resources remain also untapped, offering a wealth of opportunities for commodity investors. In a 2008 speech, the World Bank president, Robert Zoellick, called on SWFs from the Middle East and Asia to invest 1% of their assets in Africa. This could potentially channel up to USD 29.7 billion in foreign investment for Africa, almost one-third of Africa’s needs for infrastructure funding. Nevertheless, Africa’s share in foreign SWFs investments remains negligible. According to recent research published by IFLS (IFSL, 2010), Africa receives less than 5% of SWFs investment flows. For example, as of December 2009, out of 8,300 companies in which the Norway SWF held equity investments, only 144 (corresponding to 1.74%) were Africans. These companies are concentrated in 3 countries, namely South Africa (104 companies), Egypt (32 companies), and Morocco (8 companies). Tracking investments made by foreign SWFs in Africa is challenging given limited public disclosure. We were able to track a sample of 69 direct investments undertaken by foreign SWFs, including 17 investments made by the China– Africa Investment Fund. This sample is by no means exhaustive. Nonetheless, it provides useful information and stylized facts on the sector and regional allocation of foreign SWFs’ investments. Fig. 5 shows that real estate and hospitality sectors attracted the largest number of deals on the continent with North Africa attracting a smaller number of deals than sub-Saharan Africa. Interestingly, while North Africa attracted deals in the banking and financial sector, foreign SWFs invested in the industrial sector, and extractive industries in sub-Saharan Africa. Conversely, the value clustering shows that North Africa received a larger share of foreign SWFs resources, mainly to fund large real estate and infrastructure projects.

276

THOURAYA TRIKI AND ISSA FAYE

Real Estate, Hotels and Hospitality Banking Financials Infrastructure Other Industrials Extractive industries Construction/engineering services 0

5

10

Sub Saharan Africa

Fig. 5.

15

20

25

North Africa

Sector and Regional Distribution of Selected Foreign SWFs Direct Investments in Africa. Source: Authors’ calculation.

WHAT ARE THE BENEFITS OF SWFs FOR AFRICAN ECONOMIES? The landscape of African SWFs drawn earlier suggests that African SWFs are relatively small compared to their peers from other regions like the Middle East or Asia. They also suffer from governance and reputation problems that limit their ability to invest outside their home countries and to achieve good financial performance. Given their cyclical role, most African SWFs (which have a stabilization mandate) have also limited capacity to invest in long-term illiquid assets. Thus, one might argue that African SWFs have very limited value added for African economies that is linked to shortterm stabilization. However, home grown SWFs can be beneficial for African nations if they are used and structured properly in order to take advantage of their full potential. This implies that African SWFs, at least most of them, would have to go beyond their stabilization and macroeconomic stability motives to position themselves as instruments geared toward achieving economic growth, intergenerational resource transfers, infrastructure

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

277

financing, financial sector stabilization, deepening and broadening, and regional integration. Similarly, we also believe that foreign SWFs can provide a sizable source of FDIs to African countries that will lead to human and physical capital formation and ultimately growth (RiosMorales and Brennan, 2009). The benefits of creating or attracting SWFs in Africa can be appreciated from different perspectives as discussed below. SWFs as instruments to maximize investments’ risk-adjusted returns and save resources for future generations. Unlike reserves management by central banks that is usually limited to investments in US and European sovereign fixed income securities, SWFs’ holdings are more diversified and could be structured to maximize risk-adjusted returns that are not necessarily pegged to the dollar value. A business week article published in 2008, reported that the Abu Dhabi Investment Authority has returned about 10% a year since its inception,10 while the 2009 annual report for the Norway government pension fund documents an impressive 25.6% in return during 2009. These rates exceed by far any return that African central banks could potentially earn from fixed income securities. Given Africa’s demographics and important financing gaps observed in almost all sectors, accumulating resources is very important to meet the increasing needs that may arise from existing and future generations. SWFs as channels for economic diversification and development. SWFs could be useful to support economic diversification because they often invest in a wide range of asset classes. They also have long-term investment horizons and exhibit higher risk tolerance than central banks in managing foreign currency reserves. Thus, Africa-based SWFs can play an important role in supporting their local economies by directly providing capital, or by encouraging their international investees to invest in African companies. Countries like China and Saudi Arabia have been successful in using their SWFs to support their economies. According to Monitor (2008), 26% of SWFs reported investments were made in home countries. The share of SWFs resources dedicated to local investments should result from a tradeoff between the local economic needs and the amount of foreign assets required to ensure macroeconomic stability and revenue diversification. African SWFs’ investments can also be made strategically to secure inputs needed by local economies. For example, in 2007, the Abu Dhabi Mubadala took an 8.3% stake in Guinea Alumina Corporation, a USD 3 billion joint venture aimed at transforming the bauxite of Guinea into alumina. This venture will provide the alumina plant that the government of Abu Dhabi is

278

THOURAYA TRIKI AND ISSA FAYE

planning to set up with a life-long access to cost-effective alumina. African SWFs can facilitate technology transfer to African industries through their investments in multinationals as well, and by encouraging these companies to set up Research and Development (R&D) facilities in Africa. Balin (2008) argues that SWFs can play an active role in shaping up patent laws for technologies created from these R&D facilities to favor dissemination to domestic firms. Similarly, foreign SWFs resources could be channeled to Africa to develop new sectors or supporting existing ones. This could have striking effects on the amount of direct investment received by African recipient economies. Africa’s performance during the last decade shows that the continent has favorable investment prospects that fit well with the longterm, high-return perspective of SWFs. Since, foreign SWFs are looking for good investment opportunities outside the United States and other developed countries, this can turn out to be good news for Africa. SWFs as channels to bridge the infrastructure financing gap. According to estimates from Infrastructure Consortium for Africa, a little under USD 93 billion of annual investments are required to address Africa’s infrastructure needs, about one third of which is to upkeep existing networks. Infrastructure encompasses energy plants, roads, ports, water, and sanitation facilities but also information and telecommunication networks. As Africa grows at 5% per year, one can expect additional demand for more reliable and efficient infrastructure to emerge. So far, Africa’s infrastructure has been mainly funded by local governments with donors’ support and to a limited extent by private investors. According to Preqin (2011) the proportion of SWFs investing in infrastructure has increased from 47% in 2010 to 61% at the beginning of this year. This suggests that SWFs could play a bigger role to bolster infrastructure investments in Africa. Such long-term high-yielding investments meet the time and risk profile of SWF needs (OECD, 2008). First, given the monopolistic structure of many infrastructure projects in Africa, the demand for the asset tends to be inelastic and price adjustments for inflation are unlikely to be affected (JP Morgan, 2007). Therefore, infrastructure investments could provide a hedge against inflation. Second, infrastructure projects offer long-term cash flow streams that align well with SWFs investment time horizon. As a matter of fact, revenues from infrastructure projects are mainly generated through income rather than investment appreciation, which should provide more predictable, and reliable long-term cash flow streams and returns (JP Morgan, 2007). Besides, infrastructure projects historically delivered high returns

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

Table 2.

279

Risk and Return Performance of Infrastructure Projects in Percent.

Macquarie Global Infrastructure Macquarie USA Infrastructure Russell 3000 MSCI World ex US Lehman Aggregate

One Year Return

One Year Standard Deviations

Two Year Return

Two Year Standard Deviation

Three Year Return

Three Year Standard Deviation

32.28

10.23

30.80

8.78

28.50

9.23

21.05

11.05

19.68

9.75

21.48

9.98

14.53 26.33 5.38

8.12 7.95 2.62

15.45 26.90 5.29

7.41 8.75 2.48

13.81 24.02 3.88

8.02 9.67 2.78

Source: JP Morgan (2007).

(see Table 2) with low correlations with traditional asset classes thus serving as a risk reduction tool. Third, the scarcity of long-term finance on the continent and the low liquidity of African financial markets offer SWFs a good opportunity to negotiate attractive terms on their long-term funding in Africa. SWFs as channels for regional integration. African SWFs could place some of their resources in banks throughout the continent to shore up their long term deposits. Given the long time investment horizon of SWFs, this should help address the scarcity of long-term finance at the continent level. The LIA has been actively investing in hotels in Africa through LAICO. Most of these acquisitions correspond to three- to five-star hotels and are managed by international operators like the Accor or Intercontinental chain. In 2008, LAICO established a joint venture, called LAICO Hotels Management Company, with Tunisia Travel Service (TTS), a Tunisia-based company involved in the hospitality sector through hotels management, airlines and ground transportation. This illustrates how an African SWF could develop business within Africa while leveraging on another African country expertise. Given the relatively small size of most African SWFs, the latter could pool part of their resources to create regional development banks or a fund of funds that will significantly scale up their individual financing capacities. This would foster regional financial cooperation among African economies (Griffith-Jones & Ocampo, 2010). SWFs as stabilizing instruments for financial systems. Given their longterm investment horizon and low leverage, SWFs can have a stabilizing

280

THOURAYA TRIKI AND ISSA FAYE

effect on the volatility of African financial markets, especially during periods of financial turmoil. As indicated by Monitor (2009) SWFs have been instrumental in stabilizing the global financial system during the recent financial crisis while providing a total of USD 128 billion into the global economy to the substantial benefit of European and American financial institutions (USD 57.9 billion). African SWFs have contributed to stabilizing the global financial system as well. This can be seen through the intervention of the LIA to help dampen the deleterious effects of the crisis faced by some European (e.g., Italy and Spain) banks. For instance, it has been reported that in July, 2008, the LIA rescued Dutch–Belgian bank of Fortis, while acquiring some shares, to ensure its solvability. Similarly, a large number of African financial institutions – especially non banks – are not adequately capitalized. Foreign and African SWFs can strengthen the capital base of these financial institutions which should foster the continuation and expansion of their activities. As institutional investors, SWFs can also provide comfort to other investors and help improve governance and business structures of Africa’s financial institutions. Ultimately, this should lead to a more resilient financial system. In the wake of the financial crisis, China and Algeria resorted to their SWFs’ assets to recapitalize their domestic banks, which is a clear example of SWFs as means to stabilize the financial system. SWFs as channels for enhancing financial systems’ depth and breadth. In the wake of the financial turmoil, SWFs are worried that dollardenominated assets are no longer reliable stores of value. This leaves African financial markets with a window of opportunity to attract foreign investors especially that African markets have been performing quite well. On average, SWFs allocate 35% to 49% of their resources to fixed income securities, 50% to 55% to listed corporations and the remaining to alternative investments (real estate, private equity, etc.) (OECD, 2008). Using this asset allocation and a 5% share for Africa in SWFs’ portfolio, SWFs could invest up to USD 125 billion in African listed stocks, almost two-third of the combined 2009 market capitalization of the Lusaka, Nairobi, Botswana, Nigeria, and Egypt stock exchanges.11 SWFs can also enhance financial systems’ breadth by supporting nonbank financial institutions such as insurance and leasing companies and private equity funds. This will diversify financial systems in most African countries that are currently mainly bank dominated. For instance, the Norway government pension fund relies on nine external fund management companies in South Africa.

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

281

WHAT ARE THE CHALLENGES FACING SWFs INVESTMENTS IN AFRICA? SWFs offer several benefits that were discussed earlier. Yet, their establishment and management in the African context could entail the following issues. Lack of coordination with fiscal and monetary policies. In theory, SWFs are no different from any other fiscal resources at the disposal of a given government, and as such their establishment and management need to be carefully coordinated with monetary policy in the originating countries. In several African countries where inflation is an issue, a sudden increase in liquidity resulting from repatriation of returns on foreign assets can lead to unwanted inflationary pressure, which would force the monetary authority to raise interest rates above desired levels, thus slowing the economy and reducing growth. In countries with a fixed exchange rate regime, such an inflow of liquidity would lead to an undesired change in the stock of foreign reserves in order to sustain the peg. SWFs home countries also face the fundamental and longstanding issue of how to allocate their resources between SWFs and public spending on education, health, and infrastructure. In Africa, where a nonnegligible fraction of the population is plagued by poverty, hunger, and health problems, such a tradeoff remains a challenge. Finally, in countries where SWFs are funded via taxation of nonrenewable resources, the government ought to maintain the tax rates at levels that do not hamper economic activity. Furthermore, while SWFs could be used as a tool to support sound fiscal policies, they should not be viewed as a replacement solution to fiscal reforms (Le Borgne & Medas, 2007). Potential disruptive effects on markets. Potential destabilizing effects of foreign SWFs investments on recipient African countries can happen through three channels. First, large investments in recipient assets might trigger speculation bubbles leading to higher market volatility in host countries. Indeed, capital inflows are likely to affect the capital and financial account, and relative prices, and thus may affect external stability. Strong opacity characterizing most SWFs prevents proper market expectation which is likely to amplify market volatility. Small African economies and those with nascent markets are more likely to suffer from this destabilizing effect. Moreover, African stock markets are often poorly regulated when it comes to insider trading and other market manipulation, and are therefore more prone to high volatility. Second, destabilization can result from SWFs

282

THOURAYA TRIKI AND ISSA FAYE

involvement in the banking sector with SWFs distorting credit allocation process to favor their home country businesses (Heyward, 2008). Third, large reverse in SWFs flows resulting from profit repatriation or asset reallocations involve also currency transactions that might affect African countries exchange rates. Protectionist behavior against FDIs. One corollary to hostility towards SWFs investments is that recipient countries may implement protectionist regulation that adversely affects FDI flows. This is the case when SWFs are perceived as a threat to national security in the recipient countries. For instance, the Foreign Investment and National Security Act enacted in the United States in 2007 imposed more scrutiny on foreign investment made by sovereign entities. Several developed countries have also special agencies that oversee and regulate foreign investments including those done by SWFs (Committee on Foreign Investment in the US-CFIUS). African recipient countries may follow this tendency which could have negative consequences on their FDI inflows. For instance, African countries that have been implementing privatization strategies may enact such regulation to reduce foreign states intervention and encourage private sector investments. To the best of our knowledge no African country implemented regulation that specifically limits foreign sovereign investments. Yet, this may come in the near future as these SWFs (especially China) become more active on the continent.

CONCLUDING REMARKS This chapter discusses the potential role that SWFs could play in African economies, both as recipient countries and home countries. We first draw a landscape of African SWFs putting them in perspective, and describing their characteristics and investment activities in Africa. We also provide some insightful patterns about foreign SWFs activities on the continent. Our analysis suggests that African SWFs are small and mainly focusing on achieving stabilization objectives. They are also characterized by poor governance structures. Thus, their role as long-term institutional investors in Africa is likely to be negligible if current practices are maintained. To fully benefit from their SWFs, there is a need for African economies to:  Clarify SWFs’ roles, objectives, and responsibilities as suggested by the fiscal transparency and reserve management guidelines established by the IMF. SWFs should have clear objectives. Lack of clarity about the

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

283

expected outcomes from SWFs could lead to failure of these vehicles. Home countries should also ensure that investment strategies are consistent with underlying objectives.  Carefully synchronize deposits and draw downs from commodity-based SWFs with the country’s revenues accruing from the sale of nonrenewable natural resources in order to ensure that revenues are set aside to stabilize the country’s fundamentals, should resources be exhausted. For instance, countries need to establish limits on the contribution of commodity revenues to fiscal deficits and create ‘‘permanent endowment’’ that will serve long-term savings objectives only. This endowment could be used to invest in relatively illiquid assets over a longer time horizon and enhance African SWFs participation in African financial systems.  Implement strong corporate governance structures to make sure that resources are well managed and that SWFs’ investment strategies are supporting the country’s macroeconomic policies and development plans. Obviously, there is no ‘‘one-governance-structure-fits-all’’ solution given the plurality of legal forms adopted by African SWFs. These forms could also be affected by the legal origin of the fund’s home country. For instance, while Algeria, Angola, and Mauritania have French legal origins, Ghana, Namibia, and Nigeria have English ones. This could call for different governance structures. Adequate risk management systems and human resources need also to be put in place to ensure accountability and transparency. Several African SWFs are managed by local central banks. Countries should either develop internally required capacity to implement optimal investment strategies or use external managers. Conversely, foreign SWFs are expected to play a greater role in Africa. However, in order to better attract and benefit from foreign SWFs, African economies should:  Avoid overregulation of investable sectors/companies. African countries need to find a balance between protecting themselves and offering a regulatory framework conducive to SWFs involvement in their economies. This does not mean that they should enact relaxed regulation that hinders their long-term growth simply because they desperately need FDI. Relaxed FDI regulation could give foreign SWFs a high bargaining power to make acquisitions in strategic sectors and in some extent, to exert some pressure geared toward pushing the economic, financial, and regulatory reforms agenda forward in the host African countries. A potential

284

THOURAYA TRIKI AND ISSA FAYE

solution to limit SWFs influence on African economies would be to prohibit majority stakes for SWFs holdings or cancel their voting rights, should their stake exceed a threshold that needs to be determined. Another option would consist in requiring SWFs to publish a voting list on a regular basis.  Ensure a friendly business environment for foreign investors and strong protection of investors’ rights.  Make sure that the risk of SWFs controlling banks’ capital can be mitigated through the implementation of safeguards to ensure that SWFscontrolled banks are compliant with local regulation and market practices.

NOTES 1. By the end of 2009, SWFs were reported to manage 1.72% of world financial assets, twice the value of assets managed by hedge funds. 2. The term Dutch Disease was first used by the Economist magazine during the late 1970s to describe a contraction of the manufacturing sector in the Netherlands resulting from a surge in revenues from natural gas discoveries. It describes a severe decline and loss of competitiveness of the non-commodity sectors (mainly manufacturing) resulting from an appreciation of the real exchange rate as revenues from the natural resource sector enter the economy. 3. http://www.fr.for-mauritania.org/1373-0-Exclusif-For-Mauritania-publie-desdocuments-confidentiels-sur-le-pillage-du-Fond-National-des-Revenus-des-Hydrocrabures-par-la-Junte.html. 4. According to the Sovereign Wealth Funds Institute, there are currently 53 SWFs operating worldwide. In terms of market share, China, United Arab Emirates, and Norway come out as the major funds, accounting for 24%, 18%, and 12% of the global market, respectively (IFSL, 2010). The three largest SWFs in terms of asset size are commodity-based and are the following in order of importance: The Abu Dhabi Investment Authority from the UAE, the Government Pension Fund from Norway, and the SAMA Foreign Holdings from Saudi Arabia. 5. http://oxfordswfproject.com/2010/11/19/field-work-in-mauritius-please/ and http:// oxfordswfproject.com/?s ¼ zimbabwe. 6. Refer to Appendix A2 for details about the methodology used to calculate the Index. 7. The Santiago principles were launched in October 2008 by the International Working Group of Sovereign Wealth Funds (IWG) in a joint effort with the IMF to foster trust, openness, transparency, and probity in the management of SWFs. They are expected to preserve domestic SWFs and support further investments by these vehicles by addressing the fears of recipient countries. 8. http://oxfordswfproject.com/page/2/. 9. http://www.time.com/time/magazine/article/0,9171,961510,00.html.

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

285

10. http://black-capital.com/news/2010/04/inside-the-abu-dhabi-investment-authorityadia-june-2008/?lang ¼ en. 11. Based on information available on the website of the African Stock Exchanges Association.

ACKNOWLEDGMENTS The authors are very grateful to Iness Mahjoub and Aymen Dhib for excellent research assistance, Desire Vencatachellum and Hafedh Bouakez for comments provided on an earlier version of the chapter, and Victoria Barbary from the Monitor group for sharing data. The findings, interpretations, and conclusions of this chapter are entirely those of the authors. They do not represent the views of the African Development Bank, its management, or its affiliated organizations.

REFERENCES Albin-Lackey, C., Bell, J., Faria, T.M., Humphreys, H., Rosenblum, P., & Sandbu, M.E. (2004). Proposal for an Oil Revenue Management Law for Sao Tome and Principe: Explanatory Notes, Columbia University. Retrieved from http://www.columbia.edu/ Bmh2245/chapters1/enotes.pdf Asfaha, S. (2007). National Revenue Funds: Their Efficacy for Fiscal Stability and InterGenerational Equity, International Institute for Sustainable Development. Auty, R. (1993). Sustaining development in mineral economies: The resource curse thesis (p. 288). New York, NY: Routledge. Balin, B. J. (2008). Sovereign Wealth Funds: A Critical Analysis, Working Chapter Series, The Johns Hopkins University School of Advanced International Studies (SAIS). Banque des Etats de l’Afrique Centrale (BEAC). (2010). Accounting Report. Belaicha, A., Bouzidim, A., & Labaronne, D. (2009). Un fonds d’Investissement d’E´tat pour l’Alge´rie: Approche Institutionnelle et Confrontation au Mode`le Traditionnels des Fonds Souverains, Working chapter, faculte´ des sciences e´conomies, des sciences de gestion et des sciences commerciales, Universite´ Abderrahmane Mira de Bejaia (Alge´rie). Retrieved from http://www.iefpedia.com/france/wp-content/uploads/2009/12/ LABARONNE-Daniel.pdf Griffith-Jones, S., & Ocampo, J.A. (2010). Sovereign Wealth Funds, A Developing Country Perspective, Foundation for European Progressive Studies. Retrieved http://www. feps-europe.eu/fileadmin/downloads/globalprogressive/1005_SovereignWealthFunds_ SGJ_JAO.pdf Heyward, P. (2008). Are Sovereign Wealth Funds a Threat to the U.S. Banking System? Venable LLP. Hill & Knowlton and Penn Schoen Berland. (2010). Sovereign Brands Survey 2010. Retrieved from http://www.wpp.com/wpp/marketing/branding/soverign-brands-survey-2010.htm

286

THOURAYA TRIKI AND ISSA FAYE

International Financial Services London (IFSL) Research. (2009). Sovereign Wealth Funds 2009. International Financial Services London (IFSL) Research. (2010). Sovereign Wealth Funds 2010. JP Morgan. (2007). Investment Analytics and Consulting (December 2007 Edition). JP Morgan. (2008). Sovereign Wealth Funds: A Bottom-up Primer, JPMorgan Research. Le Borgne, E., & Medas, P. (2007). Sovereign Wealth Funds in the Pacific Islands Countries: Macro-Fiscal Linkages, IMF working chapter WP/07/297. The International Monetary Fund, Washington DC. Medani, A. (2010). Global Financial Crisis Discussion Series, Overseas Development Institute, Chapter 19: Sudan Phase 2. Mercer. (2010). Retrieved from http://www.mercer.com/referencecontent.htm?idContent¼ 1329380 Monitor Group-Fondazione Eni Enrico Mattei (FEEM). (2008). Weathering the Storm: Sovereign Wealth Funds in the Global Economic Crisis of 2008, SWF Annual Report 2008. Monitor Group-Fondazione Eni Enrico Mattei (FEEM). (2009). Back on Course: Sovereign Wealth Fund Activity in 2009, SWF Annual Report 2009. Norton Rose LLP. (2008). SWFs and The Global Private Equity Landscape Survey. Retrieved from http://www.nortonrose.com/knowledge/publications/2008/pub15287.aspx?page¼ all&lang¼en-gb OECD. (2008). Emerging Public and Sovereign Fund Investors in Africa’s Infrastructure: Challenges and Perspectives, official documentation at the Expert Roundtable on 11 December 2008. Preqin. (2011). 2011 Preqin Sovereign Wealth Fund Review. Retrieved from http:// www.preqin.com/item/2011-preqin-sovereign-wealth-fund-review/1/3481 Preqin. (2010). Preqin Research Report: Sovereign Wealth Funds. Retrieved from http:// www.preqin.com/docs/reports/Preqin_Sovereign_Wealth_Fund_2010_Research_Report. pdf Rios-Morales, R., & Brennan, L. (2009). The Emergence of Sovereign Wealth Funds as Contributors of Foreign Direct Investment, Oxford Business & Economics Conference Program (June 24–26, 2009). Sogge, D. (2009). Angola ‘Failed’ yet ‘Successful’, FRIDE Working chapter. Retrieved from http://www.humansecuritygateway.com/documents/FRIDE_FailedYetSuccessful_An gola.pdf Truman, E.M. (2008). A Blueprint for Sovereign Wealth Fund Best Practices, Peterson Institute for International Economics, Policy Brief N. B08–3, Washington, DC.

SWF name

Fonds de Re´gulation des Recettes Fonds de Stabilisation des Recettes Budge´taires Reserve Fund for Oil Pula Fund Fonds de Stabilisation des Recettes Budge´taires Fonds de Stabilisation des Recettes Budge´taires Fonds de Re´serves pour Ge´ne´rations Futures Fonds Souverain de la Re´publique Gabonaise Minerals Development Fund

Most Recent Estimate of Assets under Management (US$bn)

Oil

Stabilization fund

59.34

b

2009

2006c

Oil

Stabilization fund

0.003

d

2010

Angola Botswana Congo

2004e 1994g Unknown

Oil Diamonds Oil

Stabilization fund Development fund Stabilization fund

0.2 6.9 1.64

f

2008 2010 2010

Equatorial Guinea Equatorial Guinea Gabon

Unknown

Oil

Stabilization fund

1.39

d

2010

Unknown

Oil

Development fund

0.080

d

2010

1998i

Oil

Development fund

0.380

d

2010

Ghana

1994

Development fund

Libya Mauritania

2006a 2006a

Gold and other minerals Oil Oil

70 0.03425

h

2010 2009

Date of Establishment

Funding Source

Algeria

2000a

Chad

Development fund Stabilization fund

Data Source

h d

j

Year

287

Libyan Investment Authority Fonds National des Revenus des Hydrocarbures

Fund Typea

Country

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

A1. DESCRIPTION OF AFRICAN SWFs

288

A1.. (Continued ) SWF name

Minerals Development Fund Excess Crude Fund (Account) National Oil Account Oil Revenue Stabilization Fund

Country

Namibia Nigeria Sa˜o Tome´ and Prı´ ncipe Sudan

Fund Typea

Most Recent Estimate of Assets under Management (US$bn)

Minerals Oil and gas Oil

Development fund Stabilization fund Development fund

NA 3 0.010

Oil

Stabilization fund

0.15

Date of Establishment

Funding Source

1995k 2004a 2004a 2002n

Data Source

l m

o

Year

2010 2009 2009

THOURAYA TRIKI AND ISSA FAYE

Monitor group. bDirection Ge´ne´rale de la pre´vision et des politiques, Ministry of Finance, Algeria. cAsfaha (2007). dBanque des E´tats de l’Afrique Centrale (2010). eSogge (2009). fNorton Rose (2008). gBank of Botswana, http://www.bankofbotswana.bw/index.php/content/ 2009103013033-pula-fund.hMercer (2010). (i)Gabon holds since 1998 a reserve account at the level of the BEAC (Banque des Etats de l’Afrique Centrale) under the name of the Fund for Future generations. In 2010, this fund was renamed the Fonds Souverain de la Republique Gabonaise. According to BEAC report as of January 2010, the fund for future generation balance amounted to USD 0.380 billion. jMiniste`re des finances, direction ge´ne´rale du tre´sor et de la comptabilite´ publique, available at http://www.fr.for-mauritania.org/1373-0-ExclusifFor-Mauritania-publie-des-documents-confidentiels-sur-le-pillage-du-Fond-National-des-Revenus-des-Hydrocrabures-par-la-Junte.html. k MDF website available at http://www.mme.gov.na/MDF/index.htm. lInternational Monetary Fund, July 2010, Nigeria: establishing a SWF. mFinal report on assessment of public finance management in Sao Tome and Principe 2009, EC Multiple Framework Contract Beneficiaries, March 2010 available at http://ec.europa.eu/europeaid/what/economic-support/publicfinance/documents/sao_tome_e_principe_ pefa_report_2010_en.pdf. nTruman (2008). oMedani (2010). a

Africa’s Quest for Development: Can Sovereign Wealth Funds help?

289

A2. LINABURG-MADUELL TRANSPARENCY INDEX The Linaburg-Maduell Transparency Index is based on 10 essential principles that depict SWFs transparency to the public. The following principles each add one point to the index rating. The index is an ongoing project of the Sovereign Wealth Fund Institute. The minimum rating a fund can receive is 1; however, the Sovereign Wealth Fund Institute recommends a minimum rating of 8 in order to claim adequate transparency. Transparency ratings may change as funds release additional information. There are different levels of depth in regards to each principle; judgment of these principles is left to the discretion of the Sovereign Wealth Fund Institute. Point

Principles of the Linaburg-Maduell Transparency Index

þ1

Fund provides history including reason for creation, origins of wealth, and government ownership structure Fund provides up-to-date independently audited annual reports Fund provides ownership percentage of company holdings, and geographic locations of holdings Fund provides total portfolio market value, returns, and management compensation Fund provides guidelines in reference to ethical standards, investment policies, and enforcer of guidelines Fund provides clear strategies and objectives If applicable, the fund clearly identifies subsidiaries and contact information If applicable, the fund identifies external managers Fund manages its own web site Fund provides main office location address and contact information such as telephone and fax

þ1 þ1 þ1 þ1 þ1 þ1 þ1 þ1 þ1

290

THOURAYA TRIKI AND ISSA FAYE

A3. SUBSIDIARIES OF THE LIBYAN INVESTMENT AUTHORITY

Libyan Investment Authority (LIA)

Libyan African Investment Portfolio (LAP)

Libyan African Aviation Holding Company (LAAHC)

Afriqiyah Airways

Libyan Airlines

United Air Transport, …

Libya Oil Holding Ltd

Libyan Foreign Investment Company

Lap Green Holding

Long Term Investment Portfolio

Libyan African Investment Company (LAICO)

Oil Investment Company

Reserved funds: Oil Reserve Fund (ORF), (ESDF)

PORTFOLIO ALLOCATION FOR SOVEREIGN WEALTH FUNDS IN THE SHADOW OF COMMODITYBASED NATIONAL WEALTH Christopher Balding and Yao Yao STRUCTURED ABSTRACT Purpose – Study the investment and risk management approach of sovereign wealth funds when national wealth including natural resources is accounted for rather than only financial asset. Methodology/Approach – Using a range of widely used asset classes, we simulate sovereign wealth fund returns when considering only financial assets but also under varying levels of national wealth holdings in oil. We optimize two-asset financial portfolios and three-asset portfolios when including oil to maximize the risk-adjusted returns. Findings – Sovereign wealth funds by failing to invest for the national wealth portfolio are overlooking a major source of volatility. To reduce the level of volatility associated with yearly national wealth returns, allocating a higher percentage of fixed assets to high-quality fixed income and low-risk equities will maximize the risk-adjusted returns of national wealth for sovereign wealth fund states.

Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 293–312 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012014

293

294

CHRISTOPHER BALDING AND YAO YAO

Social implications – By focusing solely on the financial assets managed by sovereign wealth funds, states are exposing themselves to significant national wealth risk. Originality/Value of the paper – This is the first work to estimate the impact on national wealth of oil-dependent states by failing to account for volatile commodity prices through the investment strategies of sovereign wealth funds. Keywords: Sovereign wealth funds; risk management; national wealth; oil prices; portfolio allocation JEL classifications: G11; G15; G17

INTRODUCTION Sovereign wealth funds proclaim their similarity to other institutional investors. However, in important ways, they are not like institutional investors irrespective of the public–private interest divergence concern commonly voiced by critics. Estimates of their growth predict that by 2015, sovereign wealth funds will have assets under management of more than $12 trillion USD (Jen, 2007). Their rapid growth from the monetization of natural resource wealth is creating one of the largest classes of global institutional investors driving capital flows around the world. They are moving beyond their passive investment strategies of outsourcing asset management and increasingly relying on in house expertise. This has resulted in some well-publicized investments around the world in global brands like Citigroup and Barclays. It has also resulted in new financial centers. As one of the fastest growing institutional investors in global finance, sovereign wealth funds are driving capital flows and creating new financial centers in Abu Dhabi, Singapore, and Moscow. It is without any sense of irony that the McKinsey Global Institute referred to them as one of the ‘‘new power brokers’’ (McKinsey Global Institute, 2007). Others have noted that the rise of sovereign wealth funds signifies a shifting of the global balance of power. Sovereign wealth funds are remaking global capital markets and our understanding of institutional investors. However, sovereign wealth funds, even with prudent risk return adjustment on financial assets, may not be managing their national wealth prudently.

Portfolio Allocation for Sovereign Wealth Funds

295

The importance of commodity-dependent states prudently investing surplus capital has been noted by economists previously (Collier & Gunning, 2005). The growth of foreign exchange reserves and sovereign wealth fund assets needs to be considered jointly with their subsequent impact on national wealth and international imbalances (Alberola & Serena, 2008). The public disclosures by SWF about their returns and benchmarks yield results that broadly mirror other diversified global portfolios. Academic research, due to a relative paucity of data on the portfolios of sovereign wealth funds, has focused on different questions. Research evolved from studying the politicization of sovereign wealth fund holdings to their impact on corporate governance, firm performance, and investment returns. The results to date indicate a mixed record for sovereign wealth funds with no long-term success or failure. If sovereign wealth funds act with no political handicap or benefit from their position as public investors, then the mixed results that reflect the broader market and other institutions’ behavior appear in line with expectations. In other words, sovereign wealth funds appear to behave as profit maximizing investors. Sovereign wealth funds have followed traditional portfolio allocation and investment management strategies. These practices include establishing risk budgets, index benchmarking, asset class, and geographic diversification strategies designed to maximize their risk-adjusted returns. Their investment management strategies include utilizing basic products and services like exchange traded funds and outsourced investment managers designed to diversify risk and reduce costs. Initial concern about their activities focused on their potential private investors’ public policy role but increased scrutiny reveals their rather traditional approach to investment management (Backer, 2010; Rose, 2009). Sovereign wealth funds’ risk metrics and allocation generally produced standard portfolios designed to maximize risk-adjusted returns. Existing research, however, has overlooked the importance of portfolio risk allocation accounting for existing national wealth in the form of unmonetized natural resources (Doskeland, 2007). Most sovereign wealth funds derive their capital from oil extraction and exports resulting in large current account surpluses. Funds receive a discounted cash flow from existing national wealth via the monetization of natural resources. Due to the discounting of natural resource wealth and extractive time horizons, sovereign wealth funds’ expected growth and risk profile will remain dependent on the underlying commodity of national wealth. In other words, accounting for only the financial asset risk of a sovereign wealth fund

296

CHRISTOPHER BALDING AND YAO YAO

portfolio, ignores the underlying commodity asset forming the majority of the implied existing portfolio. If the underlying asset, in most cases oil, forming the majority of implied national wealth is volatile, then standard investment risk management may not properly account for total wealth risk. To calculate the optimal portfolio for an oil-exporting state managing a sovereign wealth fund, we created a dataset comprised of total returns from 19 major assets across equity, debt, and commodity holdings dating to 1950. We then calculate the returns on national wealth given a range of oil portfolio composition weightings and the optimal allocation between asset types across time and within decades. We find a number of interesting results. First, when the returns and volatility from oil prices are included in the risk profile of national wealth, an optimal financial asset portfolio allocates capital to lower risk equity indexes and high-quality debt. The high volatility of oil prices is balanced by lower volatility sovereign fixed income and low-risk highly liquid global equities. Second, over time as oil reserves are reduced, the sovereign wealth fund should diversify into a more balanced portfolio, though remaining over weighted to fixed income. The high-risk, high-return nature of commodity prices implies that as countries deplete their oil resources, they should allocate a higher percentage of their overall portfolio to higher risk assets. Third, the long-term growth of the sovereign wealth fund depends more on the price of oil rather than financial asset returns. In other words, while research focuses on investment management, the long-term growth of SWF’s and national wealth depends more on the future-discounted cash flow from oil exports and fiscal management. A simple implied price hedging strategy will significantly increase the risk-adjusted returns to national wealth. The portfolio allocation and investment management for sovereign wealth funds should account for national wealth risk and economic management factors beyond the basic financial return metrics that other institutional investors need not consider. National wealth and economic risk diverges from straight financial risk. Sovereign wealth funds as the guardians of national wealth need to broaden their risk purview to include the implied national wealth portfolio and fiscal management.

SOVEREIGN WEALTH FUNDS AS INVESTORS Since their discovery in 2007, there has been an extensive amount of research on the investment behavior of sovereign wealth funds. The initial concern over sovereign wealth fund investment focused on the potential conflict of

Portfolio Allocation for Sovereign Wealth Funds

297

interest between merging private investment and public financial power. Larry Summer summarized the problem writing: The logic of the capitalist system depends on shareholders causing companies to act so as to maximize the value of their shares. It is far from obvious that this will over time be the only motivation of governments as shareholders. They may want to see their national companies compete effectively, or to extract technology or to achieve influence. (Summers, 2007).

Owned primarily by nondemocratic oil-dependent exporting states, critics argued that sovereign wealth funds would allow governments to leverage investment holdings to influence foreign policy. Research however has provided evidence that sovereign wealth fund investment demonstrates minimal political influence. Early research found that sovereign wealth fund holdings resembled the portfolio construction of institutional investors when compared against asset class, geographical, and industrial diversification (Balding, 2008). Subsequent research confirmed these early findings using different datasets and techniques to analyze potential political influence over investment decision making (Avendano & Santiso, 2010). In fact, research has found that instead of leveraging existing holdings to influence policy, investment may be used as a tool to win friends. One study found that sovereign wealth funds have higher levels of investment in countries with whom they have weaker relations, implying investment is used as a method to curry favor in the recipient state (Knill, Lee, & Mauck, 2009). The importance of sound corporate governance and transparency of investment motive has been noted by critics and advocates alike (Arreaza, Castilla, & Fernandez, 2009). However, others have found, when limiting the investment to private equity and differentiating between internally and externally managed funds, sovereign wealth funds have a poor track record (Bernstein, Lerner, & Schoar, 2009). The criticism of sovereign wealth fund investment focused on the potential nefarious use of public power when coupled with private investment, but evidence appears to imply its use to buy friends. While the concern over politicizing investment decisions reducing the economic efficiency of capital remains valid in theory, there is little empirical evidence of this. Sovereign wealth funds argue that as asset managers they act like other investors, benchmarking their returns against indexes and releasing increased amounts of data about their performance. The research about the track record of sovereign wealth funds presents a mixed picture. According to one study, sovereign wealth funds demonstrate a greater cultural bias in their holdings than other institutional investors (Chhaochharia & Laeven, 2009). In other words, sovereign wealth fund invest closer to home

298

CHRISTOPHER BALDING AND YAO YAO

than other investors, recognizing the political sensitivities involved. The primary concern focuses on their returns and efficacy as investors. Research supports the idea that sovereign wealth funds act as profit oriented investors, with one study writing they are ‘‘similar to institutional investors in their demand for asset characteristics and in their impact on target firm performance y .’’ (Kotter & Lel 2010, p. 36) There are some consistent findings about the impact of sovereign wealth funds on target companies and ensuing returns. Companies that receive sovereign wealth fund investment experience a positive short term abnormal return from the announcement but return to trend with no long-term impact to firm performance (Bortolotti, Fotak, Megginson, & Miracky, 2009; DeWenter, Han, & Malatesta, 2010; Kotter & Lel, 2010). Additionally, there is evidence that sovereign wealth fund investors may help their target firms by opening up opportunities to the target firm not available to less well-connected companies (Sojli & Wham, 2010). In short, sovereign wealth funds’ returns and allocation resemble other institutional investors. There has, however, been a lack of research on whether sovereign wealth funds should resemble other institutional investors in their portfolio construction when seeking to maximize national wealth. In other words, just because sovereign wealth funds do act like other institutional investors fails to answer the question of whether they should invest like other institutions.

SOVEREIGN WEALTH FUNDS AS NATIONAL GUARDIANS Sovereign wealth fund behavior, returns, and allocation appear to resemble large international institutional investors. To date, there has been minimal research on how sovereign wealth funds should invest rather than comparing them to other institutional investors. Recent work has begun to focus on the asset allocation problem for oil-dependent sovereign wealth funds though continuing to overlook that national income and wealth implications (Brown, Papaioannou, & Petrova, 2010). Theoretical work has considered how to balance the volatility and implied currency hedging strategies of investing in foreign currency assets (Gintschel & Scherer, 2008; Scherer, 2009). This research has argued that given the volatility of oil, sovereign wealth funds should invest in high-quality fixed income to offset this risk and use these assets as an implied hedge against rapid price movement that our empirical results largely support. Research on optimal

Portfolio Allocation for Sovereign Wealth Funds

299

sovereign wealth fund allocation strategies has begun to incorporate financial assets as a potential hedging strategy to offset volatility from changes in price and ongoing depletion (Scherer, 2011). Our approach seeks to determine a balanced national wealth portfolio accounting for the implied national wealth of unmonetized natural resources. While investment research has studied the impact of including commodities into a classical portfolio, there has been little research on how sovereign wealth funds reliant on commodities should invest (Ankrim & Hensel, 1993; Jensen, Johnson, & Mercer, 2002). The primary research focus on sovereign wealth funds concerned whether they resembled institutional investors enough: what is less clear is if sovereign wealth funds should mimic institutional investors given their national wealth reliance on oil. Sovereign wealth funds manage the financial asset portion of national wealth of commodity rich states, not the national wealth.1 SWF are the guardians of financial asset wealth monetized from countries depleting their natural resources. In other words, sovereign wealth funds manage the financial assets of monetized existing commodity assets. Though significant attention has been paid to the absolute size and potential growth of sovereign wealth funds, these estimates and analyses are all misleading. These all overlook the underlying concept of national wealth. Sovereign wealth fund assets may be quite large but in most cases remain small compared to the unmonetized natural resource wealth. Countries are no more wealthy due to their decision to extract a natural resource asset and monetize it into a financial asset. In fact, monetizing existing commodity wealth into financial asset wealth has resulted in no net increase or even a decrease of wealth. It reallocates national wealth between asset classes, unmonetized commodity wealth, and monetized financial wealth, with no effective net change in national wealth.2 As states deplete their existing stock of natural resources, they reallocate their percentage of national wealth toward financial assets and away from natural resource assets. The sovereign wealth fund assets under management value represent the historical depletion of natural resources plus accumulated returns on capital. It is misleading, however, to represent the value of sovereign wealth funds as changing national wealth. Consequently, any return estimate of sovereign wealth funds should consider the larger national wealth portfolio and the accompanying risks. Including the concept of national wealth under the mandate of sovereign wealth funds has important implications for their investment strategies. Sovereign wealth fund portfolios use financial planning and investment techniques to maximize risk-adjusted returns on their financial asset portfolio. Sovereign wealth funds generally espouse a permanent income

300

CHRISTOPHER BALDING AND YAO YAO

model target providing a stream of income that will replace commodity revenue given the expectation of its eventual depletion. To meet this goal, sovereign wealth funds allocate capital in traditional methods with a balanced portfolio of equities, fixed income, and small percentage of alternative assets. For instance, the Abu Dhabi Investment Authority has well-defined and balanced guidelines for financial asset holdings (ADIA, 2009). The Norwegian Global Pension Fund (NGPF), for instance, manages their assets under a legally mandated 60/40 split between equities and fixed income (Norges Bank, 2009). Though some have argued that oil-based funds, like NGPF, should invest a higher percentage of capital in equities to meet expected outlays, this fails to account for the national wealth consequences of holding oil as the anchor asset. Russia passed legislation that gives their two funds greater latitude of investment choices with a maximum portfolio holding of 50% equity, though the Finance Ministry has, to date, preferred a more conservative portfolio of high-quality shortterm fixed-income securities (Finance Ministry of the Russian Federation, 2011). Although the exact composition of the Saudi Arabian Monetary Authority is unknown, it is believed to be in currency reserves and shorter duration high-quality fixed-income securities. The investment patterns reflect judgments and allocations that account for the risk appetite of the nation and the financial asset portfolio. There is no sovereign wealth fund that explicitly includes a national wealth risk abatement strategy. Sovereign wealth funds attempt to maximize the risk-adjusted returns of their financial assets, but have demonstrated no evidence that their portfolio risk assessment accounts for national wealth. Sovereign wealth funds may act like other investors, but should they? Sovereign wealth funds, like other investors, do not exist in isolation from underlying risks. A pension fund is seeking to match its expected long-term liabilities with its portfolio risks and capital inflows; sovereign wealth funds have similar factors to consider. Beginning from the risk-adjusted portfolio allocation of assets, sovereign wealth funds need to balance the volatility of yearly national income given their reliance on oil exports and long-term capital growth due to commodity extraction with its financial asset portfolio. As the manager of national wealth, it would appear to be a poor risk management to control financial asset risk but ignore nation income and wealth risk. These risks are not insubstantial and impact the total growth of national wealth. In Table 1, we present descriptive statistics on the total return and volatility of basic asset classes like oil, world stock, and sovereign bonds.3 There is a striking division between the asset behavior of oil and others as seen in Table 1.

301

Portfolio Allocation for Sovereign Wealth Funds

Table 1.

1951–1960 1961–1970 1971–1980 1981–1990 1991–2000 2001–2010

Asset Class Return by Decade.

Arab Light Crude

World Stock

World Government Bond

S&P 500

Dow Jones Corporate Bond Index

Emerging Market Stock

1% 0 0 0 51% 0.84 2% 0.25 9% 0.58 22% 0.44

22% 0.18 3% 0.11 16% 0.22 20% 0.28 10% 0.15 5% 0.27

2% 0.03 4% 0.05 6% 0.04 14% 0.1 8% 0.07 6% 0.08

18% 0.2 9% 0.14 10% 0.21 15% 0.13 18% 0.15 2% 0.21

2% 0.05 4% 0.05 7% 0.09 17% 0.12 9% 0.08 7% 0.05

7% 0.06 9% 0.13 21% 0.16 15% 0.27 14% 0.39 20% 0.38

Average yearly decadal return is given in percentage and volatility beneath.

From 1950 to 1970 in our sample, returns to oil were small with low volatility. During this same period, all other assets enjoyed higher returns and as would be expected, higher volatility. Sovereign wealth funds targeting a well-balanced, diversified portfolio with oil prices remaining constant were pursuing a sound strategy. However, since 1970 and the freeing of oil, prices have fluctuated wildly from year to year and across time. From 1980 to 1990, while other asset prices increased substantially, oil prices suffered an average 2% annual decline. Conversely, from 2001 to 2010, while major financial asset prices enjoyed moderate gains, oil enjoyed average annual increases of 22%. Countries blessed with large oil reserves would be remiss if they did not consider the returns on oil as the anchor of their national wealth, but this also requires additional risk management.

METHODOLOGY AND DATA We proceed by allocating between differing asset classes and the cross correlations from a basket (Sharpe, 1992, 1963). We begin by utilizing the standard portfolio optimization formula as defined Formula 1: Eðrp Þ ¼ w1 r1 þ w2 r2 1 ¼ w1 þ w2

(1)

302

CHRISTOPHER BALDING AND YAO YAO

The expected rate of return is equal to the respective weights of each asset and rate of return. To optimize the portfolio in a two-asset model, we use the standard variance minimization formula using cross correlations as defined in Formula 2: Varðrp Þ ¼ w21 s21 þ w22 s22 þ 2w1 w2 s1 s2 r

(2)

The goal of Formulas 1 and 2 is to minimize the variance at an expected rate of return maximizing the expected rate of return at a given variance. This basic formulation of the maximization of risk-adjusted returns is used by portfolio managers to purchase financial assets that provide the best riskadjusted return that minimizes correlation between assets. However, the two-asset model fails to capture the effects of national wealth anchored in a depleting and volatile natural resource. Most sovereign wealth fund states depend on the monetization of natural resource wealth to purchase financial assets. Their unmonetized natural resource wealth represents a shadow third asset altering the underlying assumption of prudent portfolio allocation when the primary asset, and a volatile one, is not considered. To measure the impact and optimize a portfolio for a country with oil as an anchor asset, we expand the traditional two-asset model to the following in Formulas 3 and 4: Eðrp Þ ¼ w0 r0 þ w1 r1 þ w2 r2 Varðrp Þ ¼ w20 s20 þ w21 s21 þ w22 s22 þ 2w0 w1 s0 s1 r0;1 þ 2w0 w2 s0 s2 r0;2 þ 2w1 w2 s1 s2 r1;2

(3)

(4)

In Formulas 3 and 4 w0 is the weighting given to oil in the national wealth portfolio. There are a few key differences in the analysis to national wealth when including oil reserves in the portfolio. First, we exclude supply shocks via new discoveries, implying the depletion or transition from unmonetized oil reserves into financial assets follows a predictable and steady time horizon. The oil as a percentage of the national wealth portfolio in most cases changes very slowly and is monotonically decreasing. Second, unlike liquid financial assets that can be turned into other assets, the oil portion of national wealth is essentially fixed in any given period and cannot be transferred into other assets except over long time horizons. Portfolios can be rebalanced but portfolios where oil is the anchor asset will continue to remain dependent on oil. Excluding oil from the national wealth calculation omits the major asset from risk analysis. Third, the total return of oilanchored portfolios versus financial asset portfolios offers stark differences

Portfolio Allocation for Sovereign Wealth Funds

303

and policy recommendations for financial professionals and policy makers. Fourth, while the portion of oil in the portfolio is essentially fixed over the short run, the risks require different responses if the focus is on national wealth rather than the narrower sovereign wealth fund portfolio. This simple expansion from two to three assets provides us the ability to expand on a better understanding of the behavior of national wealth of oil rich nations. To calculate the return on national wealth in a three-asset portfolio with oil as the anchor asset, we created optimized ex post multi-asset total return portfolios based upon widely used indexes or asset classes. We utilize only total return for widely held indexes and exclude any consideration of derivatives. The total return indexes covered 17 categories of major stock, bond, and commodity indexes such as the S&P 500, World Government fixed income, and Emerging Market Stock.4 To account for changes in national wealth via oil, we include oil prices in the form of world benchmarks Arab Light Crude and Brent Crude. Downloaded from global financial data, the data spans from 1950 to June 2010 in the form of total return indexes including all items such as dividends and capital gains. An Excel-based VBA program was then written calculating cross correlations, mean, variance, and volatility via the standard deviation of each asset to optimize the portfolio between the 17 assets and oil prices. The VBA simulation optimized the portfolio based upon Formulas 3 and 4 by determining the maximum risk-adjusted returns with the lower cross-asset correlations. The simulation was then free to determine the respective allocation, excluding oil, between financial assets to maximize the riskadjusted return of national wealth. To best describe the reality of oil-based sovereign wealth funds, we took two approaches. First, we create arbitrary portfolios where oil comprised a fixed percentage of the asset basket beginning at 99% and declining to 95%, 90%, 80%, 66%, 50%, 33%, 20%, and 10% of the portfolio. Second, we created a dynamic portfolio where the user inputs the depletion horizon of the oil reserves producing a straight-line decline in reserves. For instance, inputting a 1% yearly depletion rate would produce a 100-year time horizon to exhaustion. The yearly 1%, in this case, is then added to the financial asset basket producing the total return on national wealth.5 This produces an accurate accounting of the allocation between assets and total return. There are a few additional factors that need mentioning about the methodology. First, to create a basis of comparison we created portfolios comprised of purely financial assets and portfolios comprised of oil in both fixed and dynamic amounts. This provides a basis to compare the total

304

CHRISTOPHER BALDING AND YAO YAO

absolute return and the risk-adjusted return. Second, we optimize the portfolios both across time and within decades.6 In other words, based upon our simple trading strategy, funds optimize the portfolio for a given decade and across a longer time horizon. Given that the oldest sovereign wealth fund began in 1950, our analysis considers fund performance for every decade from 1950 to 2010 as discrete periods and over the entire period decreasing in 10-year increments. It should be noted, however, that due to the freeing of oil prices in the early 1970s, the first two decades can largely be considered anomalies for purposes of our analysis. Third, in the oil-based portfolios, the optimization strategy had the option of utilizing three-asset baskets while the pure financial portfolios could only hold two assets. Though portfolios based in oil rarely use three-asset combinations of any size until oil drops beneath 50% of the total portfolio, the option is not available to purely financial asset holders. Unlike financial asset portfolio construction, sovereign wealth funds anchored to oil enjoy a de facto fixed risk that cannot be easily shifted into other assets. Financial economists have for many years wondered why commodity-dependent economies do not make more use of the derivatives market to hedge their price risk. It appears that national wealth and prudent risk management are not primary concerns for sovereign wealth fund states managing depleting natural resources.

THE RESULTS The results from our baseline simulated optimized portfolios based upon financial assets and Arab Light Crude are presented in Table 2.7 There are two specific findings to note that also lay the groundwork for further analysis. First, on a non-risk adjusted basis post-1970, Arab Light Crude has provided superior returns to optimized two-asset financial portfolios. In only one decade, from 1981 to 1990, did the three-asset oil-based portfolio significantly underperform the two-asset portfolio benchmark. For the other decades, the three-asset portfolio exceeded the two-asset portfolio at all weightings of Arab Light Crude. From 1971 to 1980 and 2001 to 2010 specifically, the three-asset portfolio far exceeded the two-asset portfolio return, by 28.51% and 6.07%, respectively. In short, without adjusting for the riskiness of holdings in the portfolio, the three-asset portfolio with Arab Light Crude as the anchor provides superior long-term returns to a purely financial asset portfolio optimized by decade ex post. This may help explain why oil-exporting countries with the expected higher returns to oil may be reluctant to engage in a systematic price hedging strategy. Second, as has

305

Portfolio Allocation for Sovereign Wealth Funds

Table 2. Arab Light Crude Decadal Average Annual Returns Optimized Portfolios Without Risk Adjustment. Decade

1951–1960 1961–1970 1971–1980 1981–1990 1991–2000 2001–2010

Two Financial Asset Portfolio

Three Asset Portfolio with 95% Oil

Three Asset Portfolio with 80% Oil

Three Asset Portfolio with 50% Oil

4.36% 6.07% 7.60% 13.79% 8.72% 8.01%

2.33% 0.30% 49.20% –0.65% 9.34% 21.90%

5.58% 1.21% 45.24% 2.59% 10.77% 18.83%

5.30% 3.04% 36.12 9.06% 13.63% 14.08%

Results present the total average annual return by decade of ex post optimized portfolios. The two asset portfolio optimized for only financial assets. Three asset portfolios are optimized under the assumption that a given percentage of national wealth is held in oil wealth and the portfolio is optimized around this holding. The results are total return and not adjusted for risk.

been noted in previous research on oil, and will be emphasized here within the portfolio allocation framework, price volatility is extreme. From 1981 to 1990, the cumulative return on oil was negative while from 1971 to 1980 and 2001 to 2010 average annual returns equaled 49% and 22%, respectively. Not only is oil volatile across shorter time horizons by year and month but also is even across longer time periods such as decades. The superior returns generated by holding oil as an investment asset, also generate excessive risk. This matters to the economies of oil-dependent countries planning their development and public finance budgets. Managing the national wealth and sovereign wealth fund portfolio of an oil-dependent state requires mitigating the downside risk of commodity prices.

THE RISK-ADJUSTED RETURN RESULTS As noted in the previous section, while the absolute returns when including oil into a three-asset portfolio exceed two financial asset portfolios, a major concern is the large-volatility and risk-accompanying commodity prices. To account for this, we have simulated returns of national wealth adjusting returns for risk in varying oil weighted portfolios that we present in Table 3 and the specific holdings in Table 4. There are a few results of interest. First, when risk is not accounted for, three-asset oil-anchored portfolios exceed

306

CHRISTOPHER BALDING AND YAO YAO

Table 3. Decade

1951–1960 1961–1970 1971–1980 1981–1990 1991–2000 2001–2010

Arab Light Crude Decadal Average Annual Returns Optimized Portfolios Adjusted for Risk.

Two Financial Asset Portfolio

Three Asset Portfolio with 95% Oil

Three Asset Portfolio with 80% Oil

Three Asset Portfolio with 66% Oil

2.82 3.22 2.72 2.23 3.41 3.31

0.44 3.22 0.63 0.03 0.17 0.50

1.16 3.22 0.73 0.15 0.24 0.54

1.91 3.22 0.86 0.43 0.32 0.58

Results present the risk adjusted average annual return by decade of ex post optimized portfolios. The two asset portfolio optimized for only financial assets. Three asset portfolios are optimized under the assumption that a given percentage of national wealth is held in oil wealth and the portfolio is optimized around this holding. The results are adjusted for risk.

the decadal returns produced by purely financial asset portfolios, but when risk adjusted the financial portfolios far exceed oil-anchored portfolios. The financial asset portfolio returns adjusted for risk exceeded the oil-anchored portfolios by a significant margin. The risk-adjusted returns for the financial asset portfolio exceeded the oil-anchored portfolio by a factor of never less than 3 and only when oil as a percentage of the overall portfolio dropped to 66%. In most other simulated allocations, the ratio of risk-adjusted returns between the financial and oil-anchored portfolios was even higher indicating a larger risk differential. While the absolute returns to oil may exceed a purely financial asset portfolio, there is extreme risk in creating a portfolio around a primary commodity. Second, nor is the risk analysis altered by any consideration for time. Even the decades where absolute oil returns are high, the risk premium eliminates any portfolio advantage to the commoditybased portfolio. Third, when adjusting for risk in an optimized oil-anchored three-asset portfolio, most allocations require only small amounts of third assets even with low weightings of oil. In the 80% oil portfolio after 1970, optimization required a third asset only from 2001 to 2010 and in that instance it comprised 1.1% of the total portfolio. Even when the portfolio oil dependency drops significantly, the use of a third asset in constructing the portfolio is in most cases minimal. Except in the 1971–1980 decade where the remaining assets were split 40/60 between the FTSE All Share and the Food Commodity Index, the portfolio is reliant on one dominant asset to

307

Portfolio Allocation for Sovereign Wealth Funds

Table 4. Arab Light Crude Three Asset Risk Adjusted Optimized Holding Detail. Decade

80% Oil Anchor Remaining Asset(s)

20% Oil Anchor

Split of Remaining Allocation

Remaining Asset(s)

Split

1951–1960 Europe stock and S&P 500 total return

97%/3%

64%/36%

1961–1970 UK 10-year government bond and economist metals index 1971–1980 FTSE all share

93%/7%

World government bond excluding us and emerging market stock UK 10-year government bond and economist metal index FTSE all share and economist food commodity index World stock excluding US and Japan government bond Euro 16 10-year government bond and S&P 500 total return UK 10-year government bond and emerging market stock

100%

1981–1990 World stock excluding US

100%

1991–2000 S&P 500 total return

100%

2001–2010 UK 10-year government bond and emerging market stock

94%/6%

93%/7%

40%/60%

9%/91%

74%/26%

97%/3%

The 80% and 20% oil anchor columns assume that each portfolio is comprised of either 80% or 20% oil. Using the risk adjusted return simulator and optimizing with the given allocation of oil within the portfolio, the ‘‘Remaining Asset(s)’’ column lists the asset index used to optimize the portfolio within a decade given the assumed oil weighting. The ‘‘Split of Remaining Allocation’’ provides the weighting within the remainder of the total portfolio. For instance, in a portfolio comprised of 80% oil from 2001 to 2010, to optimize the risk adjusted returns of the portfolio would have necessitated holding the remaining 20% in 94% UK 10-year government bond and 6% emerging market stock.

manage the implied risk of an oil heavy portfolio. For instance, from 1991 to 2000 and 2001 to 2010, optimized three-asset portfolios were dominated by the Euro 16 10-year government bond index and the UK 10-year government index with 59% and 77% of the total portfolio, respectively. Fourth, beyond the importance of the specific second and third asset weightings are the optimized holdings. Oil-based portfolios are optimized by the inclusion of low-return and low-volatility assets. For instance, the optimized 80% oil portfolio includes such lower volatility holdings as world

308

CHRISTOPHER BALDING AND YAO YAO

stock excluding the United States, the S&P 500 total return, and the UK 10-year government bond index. The optimized 20% oil portfolio relies even more on low-volatility assets. After 1980, the dominant holdings by decades to optimize a portfolio of 20% Arab Light Crude would have been Japan government bonds, Euro 16 10-year government bond index, and the UK 10-year government bond index. The high-return, high-volatility nature of oil holdings imply that to balance the risk return ratio, a manager should include higher levels of low-return, low-volatility assets. The portfolio might not be the optimal risk return choice, but for countries reliant on oil to increase sovereign wealth fund financial assets and fund public finances, the bimodal asset allocation between high and low risk manages the downside risk faced by states.

CONCLUSION Sovereign wealth fund states find themselves in an enviable, albeit difficult position. Focusing solely on the financial asset returns of the sovereign wealth fund in the absence of the larger national wealth framework omits key factors in the long-term net asset value growth, public finances, and economic development. Oil-based sovereign wealth fund states are inextricably linked to the price of oil and their national wealth will grow in line with oil prices. Within the framework of maximizing national wealth returns, rather than focusing on risk-adjusted returns for the financial asset portfolio of sovereign wealth funds, oil-dependent states should consider the national wealth portfolio with oil as the anchor. Maximizing the risk-adjusted returns of the national wealth portfolio over the long term will outperform a narrower focus on financial asset returns. It should be noted that the results are time dependent and are not universal, especially in light of numerous sovereign debt crises. However, these results match the generally expected behavior of fixed income, equity, and commodity price movements. The empirical results presented here support theoretical models on how sovereign wealth funds should invest to minimize the risk given the national wealth portfolio. First, given the inherent price volatility of oil, sovereign wealth funds should invest in high-quality fixed-income instruments such as middle and long-term sovereign debt. The low-risk, low-return fixed-income securities will balance the portfolio anchored with high-risk, high-return oil. Second, given the high volatility of oil, asset class diversity between equities and fixed income is less problematic. Based upon our ex post portfolio optimizations, maximizing the risk-adjusted returns only requires asset class

Portfolio Allocation for Sovereign Wealth Funds

309

diversification after oil depletion approaching 50% of total reserves. Third, to maximize national wealth returns requires a reduced focus on maximizing returns and minimizing risk. Oil-dependent sovereign wealth fund states cannot escape their link to the price of oil and their dependence on it for national wealth. They can however minimize their risk by increasing their utilization of futures and options markets. Managing national wealth risk implies an increased reliance on oil options to manage price volatility, which would significantly improve risk-adjusted returns. Managing the national wealth of oil-dependent states requires portfolio managers and public policy makers to consider the larger picture, rather than the narrower risk-adjusted returns of the financial two-asset portfolio.

NOTES 1. In this study, we will focus on oil-based sovereign wealth funds to the exclusion of the non-commodity funds, primarily China, and Singapore. 2. The assumption of no net change transferring between asset types allows for extraction and other transaction costs associated with commodity monetization. In other words, national wealth is comprised of total assets minus total liabilities. National wealth is the surplus that remains after deducting extraction and transaction costs associated with turning commodities in the ground into a financial asset. 3. Please see Appendix 1 for a complete table of the asset classes considered with decadal statistics on return, variance, and volatility. 4. For a complete listing of the assets considered please see Appendix 1. 5. The Excel-based VBA program is available upon request from the corresponding author. 6. For the sake of space, only decadal returns are presented in this chapter though optimized portfolios across the entire time spans are similar to their decadal counterparts. The VBA program is available upon request from the corresponding author, which will provide additional results. 7. Portfolio simulations were conducted using the price of Brent Crude rather than Arab Light Crude and though the results are not identical, they are similar enough not to warrant additional analysis.

REFERENCES ADIA. (2009). Abu Dhabi Investment Authority Review 2009: Prudent Global Growth. Alberola, E., & Serena, J. M. (2008). Reserves, sovereign wealth funds and the resilience of global imbalances. Economic Notes, 37(3), 315–343. Ankrim, E., & Hensel, C. (1993). Commodities in asset allocation: A real-asset alternative to real estate? Financial Analysts Journal, 49(3), 20–29.

310

CHRISTOPHER BALDING AND YAO YAO

Arreaza, A., Castilla, L., & Fernandez, C. (2009). The coming age of sovereign wealth funds: Perspectives and policy issues within and beyond borders. Global Journal of Emerging Market Economies, 1(1), 1–25. Avendano, R., & Santiso, J. (2010). Are sovereign wealth funds’s investment politically biased? A comparison with mutual funds. Working Paper no. 283. OECD Development Centre, Paris, France. Backer, L. (2010). Sovereign wealth funds as regulatory chameleons: The Norwegian sovereign wealth funds and public global governance through private global investment. Georgetown Journal of International Law, 41(2). Balding, C. (2008). A portfolio analysis of sovereign wealth funds. Working Paper. Economics Department, Shenzhen, China. Bernstein, S., Lerner, J., & Schoar, A. (2009). The investment strategies of sovereign wealth funds. NBER Working Paper no. 14861. Finance Department, Cambridge, MA. Bortolotti, B., Fotak, V., Megginson, V., & Miracky, W. (2009). Sovereign wealth fund investment patterns and performance. Working Paper. Brown, A., Papaioannou, M., & Petrova, I. (2010). Macrofinancial linkages of the strategic asset allocation of commodity based sovereign wealth funds. IMF Working Paper no. WP/10/9. IMF, Washington DC. Chhaochharia, V., & Laeven, L. (2009). The investment allocation of sovereign wealth funds. Working Paper. Collier, P., & Gunning, J. (2005). Asset policies during an oil windfall: Some simple analytics. The World Economy, 28(10), 1401–1415. DeWenter, K., Han, X., & Malatesta, P. (2010). Firm values and sovereign wealth fund investment. Journal of Financial Economics, forthcoming. Doskeland, T. (2007). Strategic asset allocation for a country: The Norwegian case. Financial Market Portfolio Management, 21(2), 167–201. Finance Ministry of the Russian Federation. (2011). Investment Management. Retrieved from http://www1.minfin.ru/en/nationalwealthfund/management/. Accessed on February 21. Gintschel, A., & Scherer, B. (2008). Optimal asset allocation for sovereign wealth funds. Journal of Asset Management, 9, 215–238. Jen, S. (2007). Sovereign wealth funds. Morgan Stanley Research, October. Jensen, G., Johnson, R., & Mercer, J. (2002). Tactical asset allocation and commodity futures. The Journal of Portfolio Management, 28(4), 100–111. Knill, A., Lee, B., & Mauck, N. (2009). Bilateral political relations and the impact of sovereign wealth fund investment: A study of causality. Working Paper. Kotter, J., & Lel, U. (2010). Friends or foes? Target Selection Decision of Sovereign Wealth Funds and Their Consequences. Journal of Financial Economics, forthcoming. McKinsey Global Institute. (2007). The new power brokers: how oil, Asia, hedge funds, and private equity are shaping global capital markets, October. Norges Bank. (2009). Norges bank investment management government pension fund global annual report. Rose, P. (2009). Sovereign wealth fund investment in the shadow of regulation and politics. Georgetown Journal of International Law, 40(4). Scherer, B. (2011). Portfolio choice for oil based sovereign wealth funds. in asset allocation for central banks and sovereign wealth funds. Journal of Alternative Investments, 13(3), 24–34.

Portfolio Allocation for Sovereign Wealth Funds

311

Scherer, B. (2009). A note on portfolio choice for sovereign wealth funds. Financial Markets and Portfolio Management, 23(3), 315–327. Sharpe, W. (1992). Asset allocation: Management style and performance Measurement. Journal of Portfolio Management, 18(2), 7–19. Sharpe, W. (1963). A simplified model for portfolio analysis. Management Science, 9(2), 277–293. Sojli, E., & Wham, W. (2010). The bright side of sovereign wealth fund investments: evidence from the U.S. Working Paper. Summers, L. (2007). Sovereign funds shake the logic of capitalism. Financial Times, July 30.

312

CHRISTOPHER BALDING AND YAO YAO

APPENDIX 1 – LIST OF TOTAL RETURN ASSETS World Stock – Total return for global stocks including emerging markets based upon rough market capitalization weightings. World Stock excluding US – Same as the world stock total return index but excluding the United States. World Government Bond – Total government bond uses 10-year government bonds from developed countries and weighted by GDP. World Government Bond excluding US – Same composition of world government bond but excludes the United States. Euro 16 10-Year Government Bond Dow Jones Corporate Bond Index – Index of bullet bonds with no calls and equally weighted with a variety of maturities and industries. AAA Corporate Bond Index – AAA corporate bonds from Moody’s AAA corporate bond index with long-term maturity of 20 years or greater. Japan Government Bond UK 10-Year Government Bond Index Emerging Market Stock – Reflecting an approximate 20% Africa/Asia, 40% Latin America, and 40% East Asia weighting, the total return is meant to ‘‘reflect an evenly balanced’’ portfolio between states. Europe Stock – Includes European countries based upon their stock indexes accounting for total return based upon a weighting of total stock market capitalization. S&P 500 Total Return FTSE All Share Economist Food Commodity Price Index – Food price index weighted based upon the value of world food imports to reflect their relative importance to global prices. Economist Non-Food Agricultural Commodity Price Index – Agricultural price index based upon the world import volume to reflect broad price movements. Economist Metals Index – Non-precious metal price index with weighting based upon import weighting to reflect the share of importance. Arab Light Crude – The listed market price of Arab Light Crude oil. Brent Crude – The listed market price of Brent Crude oil.

ARE SOVEREIGN WEALTH FUNDS POLITICALLY BIASED? A COMPARISON WITH OTHER INSTITUTIONAL INVESTORS Rolando Avendan˜o and Javier Santiso STRUCTURED ABSTRACT Purpose – To study the allocation in equity markets of sovereign wealth funds’ (SWF) investments with respect to other institutional investors. To analyze the role of political regimes in the sending and recipient countries as a determinant of the allocation of SWF investments. Methodology/approach – We use mutual funds’ investments as a benchmark for SWF investment allocations. We collect data of SWF and mutual fund equity investments at the firm level and analyse them on a geographical and sector basis. We compare target investments for these two groups by looking at the political regime in the sending and recipient country, using different political indicators (Polity IV, Bertelsmann). We provide a comparison of SWFs and pension funds based on governance features related to investment. Findings – We find that the fear that sovereigns with political motivations use their financial power to secure large stakes in OECD countries is not confirmed by the data. SWF investment decisions do not differ greatly Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 313–353 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012015

313

314

ROLANDO AVENDAN˜O AND JAVIER SANTISO

from those of other wealth managers. Although there can be differences in the allocation, political regimes in the recipient countries do not play a role in explaining the allocation of sovereign wealth funds. Social implications – Investment from public institutions, such as sovereign wealth funds, can have significant implications at the economic and social level. Sovereign funds are potential sources of capital for emerging economies, and therefore can enchance economic growth. It is important to understand to what extent public institutional investors behave differently from private investors. The ‘‘political bias’’ is not a relevant factor for sovereign funds, or for other institutional investors, for allocating their capital. More often than not, their asset allocation strategies converge with other large investors, these being driven by financial and not political bias. Originality/value of the chapter – The chapter is an original contribution providing a firm-level analysis of equity holdings for two groups of institutional investors. Moreover, it emphasizes the political dimension of institutional investments, highlighting the priorities and constraints of public investors participating in financial markets. The chapter suggests that SWFs do not discriminate by the political regime of the recipient country in their asset allocation. Keywords: Sovereign wealth funds; asset allocation; regulation; political regimes; benchmark Jel Classifications: F21; G11; G18; O57

INTRODUCTION It may be referred to as the return of power brokers or as state capitalism. Whatever the label, sovereign wealth funds (SWFs) are key actors in today’s global financial landscape. The rise in their investment is impressive, with the number of deals tripling between 2000 and 2008, and jumping from US$ 4 billion to nearly US$ 130 billion.1 SWFs have captured the imagination of Western media, bankers, and policy makers. They were portrayed as politically guided institutions, using their financial strength to secure stakes in Western companies. Ironically, after being depicted as the new barbarians at the gate, politics and the media

Are Sovereign Wealth Funds Politically Biased?

315

turned them into white knights when Western financial blue chips collapsed in the midst of the global financial crisis of 2008–2009 (see the chapter by Paola Subacchi in this volume). The reality is that SWFs today are dynamic institutional investors in both industrialized and developing countries. The explosion of their investment activity is, above all, testimony to emerging markets’ wise stewardship of their national wealth.2 It also generated a wave of sceptikism and mistrust, mostly from the side of OECD member countries, which saw in these vehicles a potential threat to their financial structures and strategic industries. Cohen (2009) labeled this the ‘‘Great Trade-off’’ between the collective interest in sustaining open capital markets and the legitimate national security concerns raised by host countries.3 As they diversified their assets, criticism of SWFs spread in OECD countries. With some developed countries fearing the implications of SWFs’ entry, the increasing involvement and investment of these funds into emerging markets remained unnoticed.4 Between 2008 and 2009, emerging markets received in average 26.5% of the deals and 45.8% of the value of SWFs’ investments (Monitor Group, 2009). Allegations that SWFs lacked transparency and proper governance abated when a major crisis erupted at the core of the OECD financial system. SWFs were (legitimately) ‘‘puzzled that the standards and transparency requirements that others advocate for them go far beyond anything that had been envisaged for the highly leveraged hedge fund and private equity communities in industrial countries’’ (El-Erian, 2008). Along the same lines, it has been stressed that the need for greater transparency applies to all, including Western-based hedge funds and private equity firms (Gieve, 2008). After months of public debate and rising concerns about their investment activities, many SWFs agreed on a number of principles of investment behavior, reassuring OECD countries and international organizations about the role of public investors in the future. Under the umbrella of the IMF, the Santiago Principles5 called for higher transparency, stressing the fact that these funds should demonstrate the financial orientation of their decisions. SWFs’ investment strategy should be based on sound portfolio management principles and all relevant financial information should be publicly disclosed (GAPP 17, 18). Moreover, investments should follow an investment strategy set by a governing body. The investment policy should guide financial risk exposure, the recourse to internal/external managers and the range of activities (GAPP 18). In addition, investment decisions should aim to maximize risk-adjusted financial returns in a manner consistent with stated

316

ROLANDO AVENDAN˜O AND JAVIER SANTISO

investment policy. Any investment decisions beyond economic and financial considerations should be clearly set out in the investment policy (GAPP 19). (See also the chapter on the work of the IMF on sovereign wealth funds in this volume.) The standard OECD view regarding investments from SWFs has been nondiscriminatory, even if the Santiago Principles are fully endorsed by the Organization. Requiring sovereign funds to disclose their investment strategy and portfolio allocation would put them at a disadvantage with respect to other investors. Therefore, SWFs should be regarded in the same way as other institutional investors, and held to comply with the existing OECD Investment Guidelines, which commit their adherents to principles of transparency, nondiscrimination, liberalization, and standstill.6 (See the chapter on the work of the OECD on sovereign wealth funds in this volume.) The injunction that SWFs should be treated equal to other institutional investors stresses the importance of a comparative analysis. Still, the comparability of investment practices between public and private investors is not straightforward, and instruments for this purpose are needed. In this chapter, we consider a possible benchmark for the asset allocation of SWFs. By collecting recent data on SWF equity holdings, we analyze two dimensions of their investment: their geographical and industry allocation relative to other institutional investors (i.e., mutual funds) and the political bias of their investments. The rest of the chapter is organized as follows: the second section reviews the economic literature on asset allocation for SWFs and recent findings on this issue. The third section discusses implications for regulation and the perspective for setting a benchmark for SWFs in terms of investment. The fourth section describes the asset allocation for a group of SWFs and compares it with that of other institutional investors (mutual funds). Finally, the last section discusses the political dimension of sovereign funds’ asset allocation and concludes this chapter.

LITERATURE REVIEW AND STYLIZED FACTS The asset allocation of SWFs has been addressed from economic, legal, and political perspectives. However, the implications for asset allocation as regards the requirements that SWFs have agreed to respect (through the Santiago Principles and other agreements) are still a matter of study.

Are Sovereign Wealth Funds Politically Biased?

317

The traditional economic approach focuses on the management of reserves (Jeanne & Rancie`re, 2008; Papaioannou, Portes, & Siourounis, 2006), models of portfolio choice (Campbell, Chacko, Rodriguez, & Viceira, 2004) and contingency claims (Alfaro & Kanczuk 2005; Rozanov, 2008). More holistic approaches analyze the motives behind the establishment of each type of fund (Reisen, 2008). A recent literature on the implications of SWFs’ investments for the international financial system is rapidly growing. Bortolotti, Fotak, Megginson, and Miracky (2009) assessed the financial impact of SWFs’ investments on stock markets, stressing some similarities between SWFs and other internationally active investment vehicles such as pension funds, buy-out funds, and mutual funds. Bortolotti et al. (2009) found a significantly positive mean abnormal return on SWFs’ acquisitions of equity stakes in publicly traded companies. Sun and Hesse (2009) found that the announcement effect of SWFs’ investments was positive, and SWFs’ share purchases were positively associated with abnormal returns. Balding (2008) states that SWFs acted as economic-driven investors and that their impact on international financial markets may have been more moderate than expected. Chhaochharia and Laeven (2008) found that SWFs invest to diversify away from industries at home but do so in countries with cultural closeness, suggesting that investment rules are not entirely driven by profit maximizing objectives. In fact, the long-term performance of firms acquired by SWFs tends to be poorer than other public investors. Berstein, Lerner, and Schoar (2009) examined SWFs’ equity investment strategies and their relationship to organizational structure. They found that SWFs in which politicians were involved, these were more likely to have invested at home than those where external managers participate. At the same time, SWFs with external managers tend to invest in industries with lower price-to-earnings levels.7 Fernandes (2009) focused on SWFs’ holdings (rather than transactions) for the period 2002–2007, finding that firms with higher SWF ownership have higher valuations and better operating performances. In a companion paper, Fernandes and Bris (2009) found a stabilizing effect of SWFs’ investments on corporations. They stress the positive impacts of SWFs, notably through helping companies reduce their cost of capital.8 These findings are confirmed by other studies on the market impact of SWFs’ investment. Kotter and Lel (2010) suggested that SWFs are profit-oriented passive investors and that markets react positively to SWFs’ investment announcements. All in all, the evidence suggests that SWFs can be a stabilizing force in global financial markets.

318

ROLANDO AVENDAN˜O AND JAVIER SANTISO

Defining a Benchmark for SWFs Even if our understanding of SWF investment has improved, little is known about their benchmarks. SWFs enjoy substantial freedom in investing the funds entrusted to them (Weinberger & Golub, 2007). In contrast to international reserves, which have traditionally limited their investments to less-risky assets, the asset classes in which SWFs invest are substantially broader, including public and private debt securities, equity, private equity, hedge funds, real estate, and the use of derivative instruments. At the same time, their investment horizon is longer, and it is assumed that speculation does not play a role in their investment strategies. A number of SWFs have benchmarks for their investments, but there is great heterogeneity in their implementation and use (JPMorgan, 2008). Some funds have overall portfolio benchmarks (index or total return) whereas others use separate benchmarks for each asset class. Although the majority of benchmark indices are based on market indices, many are customized. Even if some SWFs have the mandate to target higher/riskier returns than central banks, they remain public sector institutions and are unlikely to act as hedge funds or private equity firms that engage in speculative trading and use extensive leverage (Idem, 2008). Ziemba (2008) provided an overview of disclosed benchmarks and return targets of SWFs (Table 1). Although these benchmarks might have evolved, they provide a snapshot of the indices used by some SWFs. Funds like those of Norway, Kuwait, Singapore, Saudi Arabia, the Republic of Korea, and Kazakhstan disclose some of their benchmarks in their active and passive mandates even if the holdings are not known to the public. This fact suggests that the benchmarks per se do not always reflect the investments made by SWFs. With the exception of ADIA, funds relying on portfolio investment and external managers tend to release benchmarking data. Other funds (Alberta, Alaska) disclose detailed benchmarks only for some parts of their portfolio (equity or fixed income). The most active funds focused on public and private equity stakes tend to disclose even less, sometimes only an overall return target. Other funds, such as the China Investment Corporation, include the benchmarks and targets that they expect external managers to outperform. Overall, most funds use a general index (e.g., MSCI All Country Index) as their primary (equity) target, with a range of indices being used as a global bond target (JP Morgan, Barclays, etc.). There is not, however, a unique benchmark describing the investment profile of sovereign institutions.

319

Are Sovereign Wealth Funds Politically Biased?

Table 1. Sovereign Wealth Fund China Investment Corporation

Kuwait Investment Authority Norway Bank Investment Management Saudi ArabiaSAMA Foreign Holdings Korea Investment Corporation Singapore GIC Kazakhstan National Fund Alaska Permanent Fund Alberta’s Heritage Fund

Disclosed Benchmarks for Selected SWFs, 2008a. Equities

Return net-of-fees 300 bps above MSCI All country Index for global equities 200 bps above EAFE 300 bps above MSCI EM Asia ex-Japan, benchmark suggested by manager seeking the mandate Outperform MSCI Global Index

Fixed Income

150 bps above the JP Morgan EMBI Global

FTSE large and mid-cap equity indices for the countries where it invests S&P 500, MSCI (Europe and Global), TSE (Japan)

JPM Global Bond Index, 3month Libor (cash/deposits)

MSCI world equity

Lehman global bond index (now Barclays)

MSCI world equity MSCI world equity

Lehman global bond indices Merryl Lynch 6-month T-bill index, Salomon World Government Bond Index

S&P 500, Russell 1000, Russell 2000, MSCI EAFE, EM Standard & Poor’s/ TSX Composite Index (Canada), Standard & Poor’s 1500 Index (US), MSCI EAFE

Scotia Capital Universe Bond Index

Source: Ziemba (2008). Benchmarks correspond to April 2008 (before the global crisis).

a

Toward Better Regulation If the financial crisis has temporarily mitigated some of the criticisms against sovereign wealth funds, there is still no consensus on the regulatory instruments to which SWFs’ investments should be submitted. Governments and policy makers (i.e., G8, IMF, OECD) have promoted and adopted a code of best practices to govern SWFs’ investments to appease these

320

ROLANDO AVENDAN˜O AND JAVIER SANTISO

concerns. On the contrary, advocates against regulation argue that the means for monitoring SWFs’ investments are already in place and that foreign investments in domestic companies are subject to review.9 The debate on the regulation of SWFs often focuses on the motivation and impact of their investments. Chalamish (2009) distinguished two fronts when looking at the role of international law in the SWF debate: regulating SWF activity, either in the home or the host country, and regulating protective measures taken by governments to block SWFs’ investments or diminish their impact. In addition, he identified four possible measures for increasing protection: (i) national regulation blocking foreign investment in government-owned entities; (ii) national regulation blocking investments in strategic sectors; (iii) individual screening mechanisms of proposed acquisitions or investments; and (iv) an open market policy to ensure that investments do not serve a foreign entity. Today, different modalities of national or regional legislation exist to control foreign investment, SWFs included. Indeed, initiatives for a stronger regulation of SWFs’ investments are not entirely new and federal laws already exist, notably in the United States, against potential national security threats posed by foreign direct investment (Epstein & Rose, 2009). Some proposals for regulation seek to allow SWFs to invest only through professional managers, or to limit (or deny) voting rights in the targeted companies. It has been proposed, for instance, that SWFs invest only in global index funds.10 Although the debate on new regulatory frameworks is far from closed, quantifiable objectives related to the already-agreed Santiago Principles are undeniably useful. It is indispensable, therefore, to have a reference to measure the extent to which SWFs follow these principles. A way to establish a SWF investment benchmark would be to look at other large institutional investors, such as hedge, mutual, or pension funds. Public pension reserve funds, for example, share some similarities with SWFs (Blundell-Wignall, Hu, & Yermo, 2008).11 Both are large in terms of assets, autonomous, and accountable only to governments or public sector institutions. Like SWFs, public pension reserve funds are increasingly investing abroad and moving into alternative assets. On the contrary, important differences exist in terms of objectives and funding sources.12 Compared to other institutional investors, such as mutual or hedge funds, most SWFs have long-term investment horizons. Furthermore, whereas mutual or hedge fund investors pursue profit-maximization objectives for their specific risk profile, SWFs are suspected to follow more strategic objectives.13 Moreover, some SWFs such as Norway’s must comply with

Are Sovereign Wealth Funds Politically Biased?

321

specific investment principles, requiring the companies in which they invest to fulfill specific standards (environmental, labor, transparency, etc.) that may put them at a disadvantage to purely market-driven actors (Chesterman, 2008). To contrast the equity allocation of sovereign funds with that of other investors, we provide a simple analysis of their investments, comparing them with those of a set of private (mutual) funds over a similar period. By looking at their geographical, sector, and industry allocation, we analyze whether SWFs diverge from a ‘‘benchmark’’ investor allocation, represented by the set of private (mutual) funds. Some of these funds are index funds, maintaining investments that are part of a major stock, and others are actively managed funds, attempting to outperform a stock index.14

DATA ANALYSIS Disclosed information allowing a comparison of the asset allocation of SWFs or, for that matter, mutual funds, is scarce. Therefore, we focus on information available on the stock holdings of these two groups. Information on stock holdings is obtained mainly through the FactSet/ Lionshares and Thomson Financial databases. They provide detailed information on the portfolio holdings of institutional and private funds during the past decade. They collect data from mandatory filings with national regulatory agencies (e.g., 13F filings with the Securities and Exchange Commission or Share Register in the United Kingdom), as well as information from annual reports or other primary sources. We gathered information on these two groups’ portfolios as follows: for the SWFs, we selected a group of 17 funds, including the most important in terms of assets. The sample includes nearly 14,000 observations (holdings), although some funds were excluded due to data constraints.15 Most of them are from emerging markets (11) and some from OECD countries, notably New Zealand, Norway and the United States. For the mutual fund group, we used the 25 largest mutual funds in the world. Times series were only available for some funds; therefore we restricted our analysis to a specific period, covering the last quarter of 2008, where holdings information is most complete. The total sample of mutual funds’ holdings includes 11,600 observations. Portfolio Characteristics We begin by reporting some portfolio characteristics of the two groups: holder style, capitalization group style, turnover, average price-to-earnings

ROLANDO AVENDAN˜O AND JAVIER SANTISO

322

ratio, average price-to-book ratio, average dividend yield, average sales growth, price momentum, relative strength, and market beta. Annexes 1 and 2 display the funds selected for each group and their portfolio characteristics. A straightforward comparison of portfolio characteristics between the two groups shows some similarities in their investments (Fig. 1). Fig. 1 shows that SWFs have a relatively lower beta (0.83 in average) in comparison with mutual funds (1.0 in average). The average price-toearnings (P/E) ratio is slightly higher for the SWFs group. A higher P/E ratio is associated with a higher price for each unit of net income, so the stock is more expensive. In contrast, the average price-to-book (P/B) ratio is lower for SWFs. A higher P/B ratio implies that investors expect more value Sovereign Wealth Funds

Mutual Funds

25.0 20.0 15.0 10.0 5.0 0.0 Avg Price/Earnings Ratio

Avg Price-to-Book Ratio

Average Dividend Avg Sales Growth (%) Yield(%)

1.2 1.0 0.8 0.6 0.4 0.2 0.0 Price Momentum

Beta

Fig. 1. Average Portfolio Characteristics for SWFs and Mutual Funds, 2008.16 Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009.

323

Are Sovereign Wealth Funds Politically Biased?

from the asset. There is a substantially higher average dividend yield for SWFs, which is a desirable feature for most investors. Finally, the higher average sales growth in the SWF group could be interpreted similarly to dividend yields. These indicators depict relatively small differences in the investment profile of the firms in which SWFs and mutual funds invest. Geographical Distribution To understand better the distribution of holdings, we calculated country and regional investments. Figs. 2 and 3 show the distribution for each group and the main destinations by country (10 largest recipients) and region (worldwide), as a percentage of total holdings.

Switzerland

France

Japan

Germany

China

Malaysia

United Arab Emirates

United Kingdom

Singapore

United States

Percentage of Total

SWFs 40% 35% 30% 25% 20% 15% 10% 5% 0%

Philippines

Japan

Australia

Singapore

Indonesia

United Kingdom

Hong Kong

China

Taiwan

40% 35% 30% 25% 20% 15% 10% 5% 0% United States

Percentage of Total

Mutual Funds

Fig. 2. Distribution of Fund Equity Holdings for the 10 Largest Host Countries, 2008 (Percentage of Total). Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009.

ROLANDO AVENDAN˜O AND JAVIER SANTISO

324

Africa

Latin America

Pacific

Middle East

North America

Europe

45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

Asia

Percentage of Total Percentage of Total

SWFs 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

Middle East

Africa

Latin America

Pacific

Europe

Asia

North America

Mutual Funds

Fig. 3. Distribution of Fund Equity Holdings, by Region, 2008 (Percentage of Total). Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009.

Figs. 2 and 3 show an interesting pattern of geographical distribution for each group. In the case of SWFs, where the United States is the main destination for investment, the allocation by country is more diversified than for mutual funds, where the concentration of holdings in this country is much higher.17 This could be explained by a sample bias, as the largest mutual funds here are all located in the United States. A ‘‘home bias’’ phenomenon might lay behind the fact that the United States is by far the top destination of mutual fund investments.18 In the case

Are Sovereign Wealth Funds Politically Biased?

325

of SWFs, the United States also ranks first, but the bias is less pronounced. Regarding the regional distribution of holdings, Asia is the main destination of equity investments for SWFs, followed by Europe and North America. Mutual funds, however, concentrate holdings of similar levels in North America and Asia, whereas Europe receives less of these investments. For both categories, North America, Europe and Asia rank as the top destinations. It is SWFs that show a greater diversification: while mutual funds are mostly concentrated in two regions (North America and Asia), SWFs are invested more uniformly across the three regions. They also show more presence in the Middle East, a region with much less investments from mutual funds. Sector and Industry Distribution When looking at the sector and industry distribution of assets (Figs. 4 and 5), further differences between sovereign and mutual funds come to the fore.19 Although both groups invested on average a similar share of their revenues in the finance sector (38% and 32%, respectively), SWFs focused on sectors like communication (14%), transportation (9%) and energy materials (6%). Mutual funds are clearly focused on industrial services (38%), health technology (18%), and energy materials (6%). Other sectors, with allocations below 5% each, include consumer durables and nondurables, utilities, and technology services. A more dissagregated look, at the industry level, allows the identification of more specific sub-sectors in each sector. SWFs privileged the financing sector, with significant investments in four industries: regional banks (15%), telecommunications (9%), major banks (7%), and transportation (5%). Mutual funds invested in major banks and specialty telecommunications (about 38% each), followed by pharmaceuticals (18%) and regional banks (6%). At the industry level, whereas mutual funds are heavily biased toward banks and telecommunications, SWFs have a more diversified industrial portfolio. These descriptive figures illustrate some differences between SWFs and mutual funds in terms of portfolio distribution. Sovereign funds show a higher level of diversification by country and region, and the same applies for sectors and industries. There are also differences in the industries of interest. The finance and specialty telecommunications industries are very present in the mutual fund group, where their shares are much lower for the case of SWFs.

ROLANDO AVENDAN˜O AND JAVIER SANTISO

326

Producer Manufacturing

Non-Energy Minerals

Industrial Services

Electronic Technology

Health Technology

Utilities

Consumer Non-…

Consumer Durables

Energy Minerals

Transportation

Communications

Finance

Percentage of Total

SWFs 40% 35% 30% 25% 20% 15% 10% 5% 0%

Communications

Producer Manufacturing

Technology Services

Consumer Durables

Consumer Services

Utilities

Retail Trade

Consumer Non-…

Energy Minerals

Health Technology

Industrial Services

40% 35% 30% 25% 20% 15% 10% 5% 0% Finance

Percentage of Total

Mutual Funds

Fig. 4. Distribution of Fund Equity Holdings, by Sector, 2008 (Percentage of Total). Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009.

Investment in OECD and non-OECD Countries Fig. 6 shows some noteworthy differences in sector allocation within and outside the OECD, suggesting that SWFs and mutual funds tend to invest in slightly different sectors.20 It is interesting to see the differences in the

327

Are Sovereign Wealth Funds Politically Biased?

Multi-Line Insurance

Real Estate…

Oil & Gas Production

Steel

Telecomms. Equipment

Multi-Line Insurance

Motor Vehicles

Engineer & Construction Packaged Software

Pharmaceuticals: Major

Electric Utilities

Airlines

Integrated Oil

Electric Utilities

Wireless Telecomm.

Major Banks

Other Transportation

Regional Banks

Specialty Telecomm.

Percentage of Total

SWFs 40% 35% 30% 25% 20% 15% 10% 5% 0%

Real Estate…

Major Banks

Integrated Oil

Semiconductors

Wireless Telecomm.

Regional Banks

Pharmaceuticals: Major

Major Banks

40% 35% 30% 25% 20% 15% 10% 5% 0% Specialty Telecomm.

Percentage of Total

Mutual Funds

Fig. 5. Distribution of Fund Equity Holdings for the 12 Largest Industries, 2008 (Percentage of Total). Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009.

allocation in each group, as OECD are often associated to industrialized countries. This provides a rough comparison of the SWF allocation between industrialized and developing economies.

60% 50% 40% 30% 20% 10% 0%

Fig. 6. Process Industries

Not Classified

Energy Minerals

Consumer Services

Non-Energy Minerals

Electronic Technology

Utilities

Consumer Durables

Industrial Services

Transportation

Communications

Finance

Retail Trade

Technology Services

Producer Manufacturing

Non-Energy Minerals

Communications

Electronic Technology

Utilities

Consumer Durables

Health Technology

Consumer Non-…

Energy Minerals

Finance

Percentage of Total

SWF investments in non-OECD

Mutual fund investments in non-OECD

60% 50% 40% 30% 20% 10% 0%

SWF and Mutual Fund Equity Investments in OECD and Non-OECD by Sector, 2008 (Percentage of Total). Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009. ROLANDO AVENDAN˜O AND JAVIER SANTISO

Industrial Services

Consumer Non-…

Consumer Durables

Process Industries

Transportation

Retail Trade

Utilities

Non-Energy Minerals

Communications

Energy Minerals

Mutual fund investments in OECD

Electronic…

Finance

Energy Minerals

Utilities

Consumer Durables

Producer Manufacturing

Consumer Services

Technology Services

Retail Trade

Consumer Non-…

Health Technology

Communications

Electronic Technology

Finance

Percentage of Total

SWF investments in OECD 328

60% 50% 40% 30% 20% 10% 0% 60% 50% 40% 30% 20% 10% 0%

Are Sovereign Wealth Funds Politically Biased?

329

Regarding SWF presence in OECD countries, about 27% of investments go to finance, whereas this figure is higher (35%) in non-OECD destinations (which includes emerging markets). Although investments in OECD countries are also focused on sectors like energy, consumer nondurables and health technology, in the case of non-OECD countries they concentrate on infrastructure and ICT-related industries, such as communications and transportation. Regarding mutual funds, a slightly different landscape emerges. Mutual funds invested in finance in OECD and non-OECD countries at similar levels (21% and 30%), followed by electronic technology (10% and 17%). Mutual funds also showed some similarity in their sector investment profile between OECD and non-OECD regions.21 These observations could suggest that SWFs (which include mostly nonOECD countries in our sample) may have different investments profiles in OECD and non-OECD countries. The benchmark investor, mutual funds in this case, shows a more homogenous sector distribution among the two country groups.

SWF INVESTMENTS: THE POLITICAL DIMENSION So far, our study of asset allocation has not taken political factors into account. Leaving aside issues of political economy surrounding the rise of SWFs, we focus on analyzing the relationship between asset allocation and political regimes. We assess whether asset allocation for SWFs is independent or not from the political regime, particularly in host countries. The relationship between investment and democratic regimes has been studied in the past, particularly in the context of multinational enterprises (MNEs). From the seminal contribution of Barro (1996) on democracy and economic growth, the effects of political regimes on growth enhancers, such as investment, have been studied. Busse (2003), for example, examined the relationship between democratic regimes and FDI, suggesting that investments by MNEs are significantly higher in democratic countries. Jensen (2006, 2009) showed that U.S. MNEs tend to restrict the size of their investments in authoritarian regimes relative to democratic ones. Li and Resnick (2003) studied the effect of democratic institutions on FDI inflows whereas democracy hinders FDI by limiting oligopolistic or monopolistic behaviors of MNEs, it encourages FDI inflows by promoting credible property rights protection, reducing risks and costs to investors. The net effect of democracy on FDI inflows is contingent on these two forces.

330

ROLANDO AVENDAN˜O AND JAVIER SANTISO

Regarding aid portfolios in developing countries, Frot and Santiso (2008, 2009) found that OECD donor countries do not reward transitions toward democracy and official aid donors invest indifferently in democratic and autocratic countries. However, little research has been devoted to the impact of the political regime on institutional investors’ behavior, and none to our knowledge on the relation between SWF asset allocation and political regimes of home/ host countries. To address some of these issues, we provide a series of indicators on governance (for SWFs and pension funds) and political regimes in home and host countries (for SWFs and mutual funds).

Internal Governance and Investment Strategy We start by comparing SWFs with another institutional investor, specifically pension funds, using well-known data from Truman (2008) on governance and functioning for the two groups. The use of pension funds as a benchmark for internal governance derives from the fact that they are publicly owned, and have specific governance and allocation standards, in contrast to other private institutional investors. Data availability on corporate governance was more accessible for these investors. A more detailed analysis would be needed to analyze the political economy behind the emergence of SWFs. Fascinating work has been devoted to the case of the Singapore Government Investment Corporation (Clark & Monk, 2009) and the institutional context behind the emergence of SWFs.22 Using a survey of SWF and pension fund managers, Truman (2008) collected valuable data regarding these funds’ investment strategy, transparency, fiscal treatment and management, among others. We focus on those variables most relevant to asset allocation strategies: transparency level, existence of an investment strategy, use of a benchmark, a policy of specific investments, and credit ratings.23 Fig. 7 provides a simple comparison for these variables, taking the average of each group, SWFs and pension funds.24 Fig. 7 suggests that pension funds surpass SWFs in all measured criteria: they show higher transparency levels, the investment strategy is communicated more clearly, the use of benchmarks is more frequent, investments are more constrained by credit rating minimums, and their policy towards specific investments is more defined. Clearly, the heterogeneity of SWFs is not reflected on these indicators, but nevertheless suggests the existence of a gap in the investment policies between the two groups. When comparing

331

Are Sovereign Wealth Funds Politically Biased? Sovereign Wealth Funds

Pension Funds

1.20

Index 0–1

1.00 0.80 0.60 0.40 0.20 0.00 Transparency

Investment Strategy

Benchmarks

Credit Ratings

Specific Investments

Fig. 7. Investment Criteria in SWFs and Pension Funds, 2008a (Index 0–1). Source: Truman (2008). aAll scores standardized to values between 0 and 1.

OECD and non-OECD SWFs (see Annex 5), the differences in their investment strategies are stressed.25 Moreover, regarding commodity and noncommodity funds (Annex 6), a clear disparity exists between the two groups, with noncommodity funds having higher levels of transparency, investment strategy, investment benchmarks and credit ratings constraints.26 Political Regimes Considering the issue of political regimes and investments, we take into account two dimensions: (a) The political regime for the (investment) home country. For this, we use data from Polity IV and Truman’s fund sample, to compare political characteristics between SWFs and pension funds.27 (b) More importantly, the political regime for the (investment) host country. For this, we use our database of holdings for SWFs and mutual funds, and look at the political characteristics of each destination (again using data from Polity IV regime characteristics and transitions). We test the hypothesis that SWFs and mutual funds do not discriminate their investments by the host country’s regime.28 We control the results with a second database, the Bertelsmann Transformation Index (BTI) which provides two political indices: first, the Status Index that ranks countries according to their state of democracy and

ROLANDO AVENDAN˜O AND JAVIER SANTISO

332

market economy (as of Spring 2007); second, the Management Index ranks them according to their leadership’s management performance between 2005 and 2007. Although the first indicator is static, the second provides a dynamic indicator of performance.29 Fig. 8 details differences in political regime between the home countries of SWFs and pension funds, using the Truman database.30 Note that in this case we only compare country average values. Not surprisingly, the level of autocracy is higher in the case of SWFs, whereas the polity score and various measures of political competition are higher in the pension fund group. More often than not, SWF investors belong to autocratic regimes rather than democratic ones. Sovereign Wealth Funds

Pension Funds

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Institutionalized Democracy

Autocracy

Polity Score

Sovereign Wealth Funds

Regulation of Chief Competitiveness of Executive Executive Recruitment Recruitment Pension Funds

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Openness of Exec. Recruitment

Executive Constraints (Decision Rules)

Regulation of Participation

Competitiveness of Participation

Executive Recruitment

Political Competition

Fig. 8. Political Regime for SWFs and Pension Funds by Home Country, 2008 (Index 0–1). Source: Authors’ calculation, based on LionShares, Thomson Financial and Polity IV databases, 2009.

333

Are Sovereign Wealth Funds Politically Biased?

Most interesting is the political regime of the host countries targeted by SWFs and mutual funds. This allows us to exploit our database at the holding-level. Fig. 9 provides a comparison of political characteristics between countries attracting SWF investments and mutual fund investments. The definition for each of the political variables may be found in Annex 4. Revealingly, there are more similarities than differences when looking at the political regime and corporate governance of firms targeted by SWFs and mutual funds.31 The indicator of institutionalized democracy, which

Sovereign Wealth Funds

Mutual Funds

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Institutionalized Democracy

Autocracy

Polity Score

Sovereign Wealth Funds

Regulation of Chief Executive Recruitment

Competitiveness of Executive Recruitment

Mutual Funds

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Openness of Exec. Recruitment

Executive Constraints (Decision Rules)

Regulation of Participation

Competitiveness of Participation

Executive Recruitment

Political Competition

Fig. 9. Political Regime Characteristics in SWF and Mutual Fund Host Countries Target, 2008 (Index 0–1). Source: Authors’ calculation, based on LionShares, Thomson Financial and Polity IV Project, 2009.

334

ROLANDO AVENDAN˜O AND JAVIER SANTISO

reflects the competitiveness of political participation, is very similar for both types of investors. The regulation levels of chief executive recruitment, referring to the procedures for transferring executive power, are nearly equal. The same occurs for the indicators of competitiveness of executive recruitment, which refers to the extent that subordinates have equal opportunities to become superordinates, or political competition. In sum, when considering the host countries of SWFs and mutual funds in their investments, there is no significant gap in the political regime or corporate governance characteristics between the two groups. This reinforces the hypothesis that SWFs are in fact more oriented toward risk-return and profit-maximization objectives than often thought. Although there are some exceptions (notably some GCC funds who invest more in autocratic regimes), we see very little difference between the political profiles of SWFs and mutual funds’ investment destinations. Both invest in democratic and nondemocratic regimes, and are in fact indifferent to this political dimension. These results were confirmed by Bortolotti et al. (2009) in their analysis of host countries for SWFs’ investments. They found that the United States, the largest OECD democracy, is the most targeted country, with 22% of SWF deals’ value. Just behind is China, the largest autocratic emerging economy, explained by the focus of the China Investment Corporation (CIC) on domestic firms. Other than that, popular target countries are indifferently democratic and autocratic regimes: India, the largest emerging market democracy, but also the United Kingdom and Australia, Malaysia and Singapore were among the other major recipients of SWF investments, along less democratic host countries like Saudi Arabia, Vietnam, Libya, and Tunisia. The Bertelsmann indicators complement those explored with Polity IV, by introducing a static/dynamic dimension into the analysis. We thus calculate the average Bertelsmann Democracy, Status and Management indices for both SWFs and mutual funds, looking at individual holdings in the host countries. The results are summarized in Fig. 10. Although some differences between group means for the key indices exist, they are equivalent for others.32 However, the gap between the two groups does not seem to be pronounced. The management index is particularly revealing, as it indicates the dynamic improvement in the quality of governance; the score difference between SWFs and mutual funds for 2007 could indicate a ‘‘democratic premium’’ of private investors toward countries showing improvement in this direction. For 2009, the indicator attains similar levels for both groups.

335

Are Sovereign Wealth Funds Politically Biased? Average Mutual funds

Average SWF funds

Democracy Status

Status Index

Management Index

0

2

4

Average Mutual funds

6

8

10

Average SWF funds

Democracy Status

Status Index

Management Index

0

2

4

6

8

10

Fig. 10. Bertelsmann Index for SWF and Mutual Fund Holdings, 2007 vs. 2009 (Index Value). 2007–2009. Source: Authors’ calculation, based on LionShares, Thomson Financial and Bertelsmann Transformation Index, 2009.

CONCLUSIONS The recent debate on the regulation of SWFs culminated with the adoption of the Santiago Principles on fund transparency, investment orientation and accountability (see the chapter on the work of the IMF in this volume for more on the Santiago Principles). The implementation of these principles supposes that SWFs should be considered on the same basis as other institutional investors. They should follow investment practices similar to

336

ROLANDO AVENDAN˜O AND JAVIER SANTISO

those of, say, public pension, mutual or hedge funds. With that objective in mind, we compared different dimensions of investment between two institutional investors, SWFs and mutual funds. Although differences exist in the allocation of SWFs and other funds, they do not suggest that their investment motives are radically different. We also introduced a new dimension in analyzing SWF investments: the political regime in the home and host countries. Although it is unsurprising that differences in the political regime of home countries exist (with SWFs’ regimes tending to be less democratic), we find that SWF investments are not different from mutual fund investments in terms of political regime characteristics in the targeted countries. This evidence suggests that they do not discriminate by this criterion in their asset allocation. Both invest in countries with democratic and autocratic regimes. More often than not, their asset allocation strategies converge, these being driven by financial and not political bias. Some policy implications can be drawn from this analysis. First, in a world of post-2008-financial-collapse, applying double standards is and will be more difficult to legitimize than in the past. Emerging economies, starting with China or Singapore, have little time to be lectured by rich countries that set off a major global financial crisis. OECD countries, do not have the monopoly on best practices. Some emerging economies have proven that they can also generate best practices and be more virtuous in applying sound policies. This has a practical consequence, as shown by the joint efforts of Western based institutions and SWFs to generate shared principles: double standards should be avoided. What could be requested from some (e.g., disclosure and more transparency from SWFs) should also be asked from others (public pension funds, mutual funds, central banks33). More importantly, the definition of such principles should be done jointly and shared; in this regard, the inclusive process of the IMF-led International Working Group of Sovereign Wealth Funds (IWG) is a promising one to replicate. The rise of SWFs offers in the end an excellent opportunity to invent more inclusive global structures and processes.34 Second, SWFs, like mutual funds, are investing in countries because it is financially rewarding, regardless of their political regime. They should resist requests to make investments without a good financial rationale to do so. Taking into consideration nonfinancial objectives, even if they are ethically rewarding, can be a double-edge sword. Finally, if SWFs wish to avoid future criticism, they should, like their Norwegian or Chilean peers, increase disclosure levels in a balanced way.

Are Sovereign Wealth Funds Politically Biased?

337

Disclosure and transparency levels are much higher today than in the past, as in the case of Temasek or Mubadala.35 It is important for them to follow sound corporate governance policies that generate confidence in home countries as well as host countries. Other strategies to achieve higher standards could involve the creation of international advisory boards, as was done in 2009 by the CIC. This chapter leaves open other questions related to the political rationale of SWFs’ investment. One dimension that requires further analysis is the domestic political economy dimension of capital exporters. Do SWFs contribute not only to the wealth but also to the welfare of their own citizens?36 Another direction would be to focus on the emergence of new SWFs, the political tradeoffs through which they rise and their relations with stakeholders and local governments. Lastly, given the recent increase of domestic investments by SWFs during the 2008 crisis, it would be revealing to know whether such rebalancing was based on commercial conditions (e.g., currency risks arbitrage, information asymmetries, and low-priced assets) or on political criteria (bolstering domestic corporations, preserving jobs, protecting heavily debt companies, etc.).37

NOTES 1. According to Monitor Group and Fondazione ENI Enrico Mattei (2009). Estimates of SWFs assets under management in 2009 ranged from US$ 1.5 trillion to US$ 3 trillion. For a discussion of the estimates, see http://blogs.cfr.org/setser/2009/ 08/02/how-much-do-sovereign-wealth-funds-manage/. Most of the SWFs have been heavily hit by the 2008 global financial crisis. For specific and detailed re-estimations of SWFs’ assets and losses before, during and after the crisis, see for example on Gulf countries, Setser and Ziemba (2009) and Kern (2009). 2. See some definitions of the term in Jen (2007a, 2007b, 2008) and Rozanov (2005, 2007). Later SWFs from emerging markets have developed their own analysis. See reports on CIC by Chen (2008, 2009). 3. Ironically, OECD based SWFs promoted investments abroad incorporating political and ethical considerations in their decisions. In the case of Norway, it resulted in controversial divestment decisions from Wal-Mart and companies operating in Myanmar (Chesterman, 2008). 4. For some research regarding the south-south dimension of SWF investment and their contributions to development in other emerging and developing countries, (2008). For a special focus on Arab SWFs, see Behrendt and Kodmani (2009). 5. Reprinted as an annex in this volume. 6. In addition to the investment standards that the OECD demands from public and private investors, some principles have been highlighted for sovereign funds. The transparency/predictability principle refers to the codification and publication of

338

ROLANDO AVENDAN˜O AND JAVIER SANTISO

laws regarding investment, prior notification to interested parties about plans to modify investment strategies, consultation of these strategies with other counterparts and the disclosure of investment policy actions. The regulatory proportionality principle stipulates that restrictions on investment should not be greater than is needed to protect national security. Finally, the accountability principle is an objective for guaranteeing periodic regulatory impact assessments, parliamentary oversight, and other supervision activities. See also OECD (2008a, 2008b) for the OECD declaration on SWFs, the Freedom of Investment Process and the OECD General Investment Policy Principles. 7. SWFs have been associated in structure and objectives to hedge funds (Ferreira & Matos, 2007; Klein & Zur, 2006), in that SWFs are also stand-alone, unregulated pools of capital, managed by investment professionals, and often take large stakes in publicly traded companies. See Bortolotti et al. (2009) for a more complete review of the literature. 8. Importantly, SWFs tend to be passive rather active investors. The typical position taken by a SWF is not a controlling stake: on average, an SWF takes 0.74% of the shares outstanding in a company. With such a limited stake, SWFs can hardly be viewed as possessing control over companies, at least directly (some SWFs externalized the management of their assets to investment firms – most of them located in OECD countries; this is the case of ADIA (Abu Dhabi Investment Authority) which had 70–80% of its assets before the 2008 crisis managed by external asset managers according to JP Morgan). See JP Morgan (2008). 9. Currently, in the United States, the regulatory framework for institutional investors has similarities to the one for private investors. They have to make disclosures pursuant of the Securities Exchange Act of 1934 if they acquire a 5% or higher equity stake in a public company. In addition, a number of U.S. statutory regimes restrict foreign control in certain sensitive industries, like nuclear energy and airlines. In the United States, this is done by the interagency Committee on Foreign Investment (CFIUS). Under this Committee, any transaction that could result in a foreign entity’s control of a company engaged in interstate commerce in the United States is subject to a review to determine the effects of the transaction on national security. See the chapters by Larson et al. and Lowery in this volume. 10. See Aizenman and Glick (2008) for a proposal for investment in diversified global equities to prevent destabilization. 11. For clarifying definitions and differences between public pension funds and SWFs see Monk (2008a). 12. Although commodity funds are set up to protect the domestic economy against fluctuations in commodity prices, public pension funds serve as long-term financing vehicle of public pensions. 13. There are also important differences in regulation for each of these market participants. Unlike hedge funds, mutual funds are required to register with the Securities Exchange Commission in the United States. Hedge funds are not required to have specific investment strategies, or prohibit specific investments. 14. Although index funds provide a representative allocation strategy, a broader comparison with actively managed funds is more enriching for the analysis. 15. Equity holdings for a number of funds are not available or incomplete in either Lionshares or Thomson One database.

Are Sovereign Wealth Funds Politically Biased?

339

16. An index of concentration (Herfindahl-Hirschman) by region illustrates this pattern (a value of 0.12 for SWFs and 0.19 for mutual funds). A low HH index (close to zero) indicates a high degree of diversification of investment destinations. A high HH (close to 1) indicates a higher concentration of investments. In the case of sector concentration (0.10 for SWFs vs. 0.30 for mutual funds) and industry concentration (0.04 for SWFs vs. 0.33 for mutual funds) the difference is even more important. 17. See Hau and Rey (2008) for a review. 18. Following the Factset classification, we included 23 sectors, as follows: finance, communications, transportation, energy minerals, consumer durables, consumer nondurables, consumer nondurables, utilities, health technology, electronic technology, industrial services, nonenergy minerals, producer manufacturing, technology services, consumer services, retail trade, process industries, commercial services, distribution services, health services, miscellaneous, and government. The industry classification includes around 130 categories. See Annex 3 on sectors and industries. 19. Founded in 1961, the OECD is a forum of countries committed to democracy and the market economy. It currently has 34 member countries: Australia, Austria, Belgium, Canada, Chile, Czech Republic, Denmark, Estonia, Finland, France, Germany Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the United States. 20. The Herfindahl-Hirschman concentration index for SWF investments are 0.08 (in OECD) and 0.19 (non-OECD), and for Mutual funds 0.05 (in OECD) and 0.1 (in non-OECD). These results suggest a higher concentration of investment sectors for SWFs in OECD countries. 21. In another piece, Clark (2009) focused on the governance of SWFs from the perspective of competing political interests. Moreover, a detailed analysis is devoted to the Future Fund (from Australia). 22. Specifically, we focus on the following questions from the Truman survey: (1) Is the overall investment strategy clearly communicated? (2) Does the strategy use benchmarks? (3) Do regular reports on the investments by the SWF include information on the specific investments? (4) Does the strategy limit investments based on credit ratings? 23. See Annex 7 for a description of funds included in each sample. 24. In the case of OECD/non-OECD funds, T-test for the sample revealed significant differences at 5% for transparency, investment strategy and credit ratings. For commodity/non-commodity funds, differences are significant at 5% for investment strategy, benchmarks and credit ratings. 25. Remarkably, OECD-based and noncommodity funds showed, in average, similar levels to those of pension funds. 26. Polity IV is a comprehensive database examining concomitant qualities of democratic and autocratic authority in governing institutions. The Polity IV dataset covers all major, independent states (i.e., states with a total population of 500,000 or more in the most recent year; currently 163 countries) over the period 1800–2008. The Polity IV Project constantly monitors regime changes in all major countries and provides annual assessments of regime authority characteristics and regime changes, for purposes of comparative, quantitative analysis. The project has become one of the most widely used resources for monitoring regime change and studying the

340

ROLANDO AVENDAN˜O AND JAVIER SANTISO

effects of regime authority. For more information see http://www.systemicpeace.org/ polity/polity4.htm. 27. For each equity holding in our database (14,435 for SWFs and 11,600 for mutual funds) we identify a host country. For each destination, we determine the Polity IV scores for political regime, in order to calculate averages. 28. The Status Index explores the state of development achieved by countries on their way to democracy under the rule of law and a market economy flanked by sociopolitical safeguards, as of Spring 2007. Status Index scores result from the combined scores given for the status of political and economic transformation. The Management Index evaluates the quality of governance among decision makers from 2005 to 2007. 29. Although ideally the comparison should be done between SWFs and mutual funds, we use data on pension funds, since all the mutual funds in our sample (the 25 largest funds) were based in the United States. 30. T-tests show no significant differences (at 5%) for some, if not all, of the variables in Fig. 9. Tests for differences in means using very small thresholds illustrate a minor gap between both groups. 31. T-tests indicate that differences between means for Democratic status and management index are different from zero (1% of significance) and zero for the status index in 2007. For 2009, the management index is statistically equal for both samples (1% significance). Other indicators (i.e., macrostability, property rights, and economic performance) are statistically equal for both samples. 32. As stressed by the Bank for International Settlements (BIS), very few central banks in the world have full disclosure of their holdings in terms of asset or currency allocation; notable exceptions include the Bank of Canada, the European Central Bank and the Bank of England (BIS, 2008). 33. For a discussion on the emerging international regime related to SWFs, see Helleiner (2009), Arreaza, Castilla, Ferna´ndez (2009), and Ochoa and Keenan (2009). 34. For Temasek, see the report at http://www.temasekholdings.com.sg/temasekreview/2008/index.html. For Mubadala, see http://www.mubadala.ae/media-files/ 2009/04/23/20090423_FINAL.pdf. See Elson (2008) on the high degree of transparency of Temasek. By mid-2009, Singapore Temasek SWF was leading with Emirates based Mubadala, the Linaburg-Maduell Transparency Index developed by the Sovereign Wealth Fund Institute, ahead the ones of Ireland, Alaska (USA) and Norway. See http://www.swfinstitute.org/research/transparencyindex.php. 35. For a good introduction to this approach one could refer to Kennan (2009). 36. Many Gulf SWFs in 2008 and 2009 injected capital into local banks hit by the global financial crisis. Early 2009, KIA from Kuwait or Bahrain’s SWF – Mumtalakat Holding Company – both decided to continue focusing their investments in Kuwait and Bahrain, respectively, instead of overseas. Such behavior has been followed by OECD SWFs too.

REFERENCES Aizenman, J., & Glick, R. (2008). Sovereign wealth funds: Stylized facts about their determinants and governance, National Bureau of Economic Research, NBER Working Papers 14562.

Are Sovereign Wealth Funds Politically Biased?

341

Alfaro, L., & Kanczuk, F. (2005). Sovereign debt as a contingent claim: A quantitative approach. Journal of International Economics, 65(2), 297–314. Arreaza, A., Castilla, L. M., & Ferna´ndez, C. (2009). The coming age of sovereign wealth funds. Global Journal of Emerging Market Economies, 1, 25–41. Balding, C. (2008). A portfolio analysis of Sovereign Wealth Funds. Mimeo University of California Irvine. Barro, R. (1996). Democracy and growth. Journal of Economic Growth, 1(1), 1–27. Behrendt, S., & Kodmani, B. (Eds). Managing Arab sovereign wealth in turbulent times—and beyond. Carnegie Endowment for International Peace. Number 16, 2009. Berstein, S., Lerner, J., & Schoar, A. (2009). The investment strategies of sovereign wealth funds. Harvard Business School Working Paper 09–112. Retrieved from http:// www.hbs.edu/research/pdf/09-112.pdf Bank for International Settlements. (2008). ‘‘FX reserve management: trends and challenges.’’ BIS Papers 40 (2008), Retrieved from http://www.tau.ac.il/Balexcuk/pdf/Paper.pdf Blundell-Wignall, A., Hu, Y., & Yermo, J. (2008). Sovereign wealth and pension fund issues. OECD Working Papers on Insurance and Private Pensions 14. Bortolotti, B., Fotak, V., Megginson, W., & Miracky, W. (2009). Sovereign wealth fund investment patterns and performance, University of Oklahoma Working Paper. Busse, M. (2003). Democracy and FDI. Hamburg Institute of International Economics Working Paper 220. Retrieved from http://ssrn.com/abstract¼384480 Campbell, J., Chacko, G., Rodriguez, J., & Viceira, L. M. (2004). Strategic asset allocation in a continuous-time VAR model. Journal of Economic Dynamics and Control, 28(11), 2195–2214. Chalamish, E. (2009). Rethinking Global Investment Regulation in the SWFs Era. Asian Society of International Law and National University of Singapore, Working paper 2009/5. Chen, C. (2008). Sovereign wealth funds, macroeconomic policy alignment and financial stability. China Investment Corporation 2008. Unpublished; available at ssrn database. Chen, C. (2009). The theoretical logic of sovereign wealth funds. China Investment Corporation 2009 (unpublished). Chesterman, S. (2008). The turn to ethics: Disinvestment from multinational corporations for human rights violations-the case of Norway’s sovereign wealth fund. NYU Law School Public Law Research Paper No. 08-25 (2008), (unpublished). Retrieved from http:// www.simonchesterman.com/ Chhaochharia, V., & Laeven, L. (2008). Sovereign wealth funds: their investments strategies and performance. CEPR Working Paper 6959 (2008). Clark, G. (2009). Temptation and the virtues of long-term commitment: The governance of sovereign wealth fund investment. Oxford University Working Papers in Employment, Work and Finance 09-01 (2009), (unpublished). Retrieved from http://www.geog.ox. ac.uk/research/transformations/wpapers/wpg09-01.html Clark, G. and Monk, A. (2009). Government of Singapore Investment Corporation: insurer of last resort and bulwark of nation-state legitimacy. Oxford University Working Papers in Employment, Work and Finance 09-05(2009). Retrieved from http://www.geog.ox.ac.uk/ research/transformations/wpapers/wpg09-05.html Cohen, B. (2009). Sovereign wealth funds and national security: The great trade-off. paper presented at the annual meeting of the ISA’s 50th Annual Convention, ‘‘Exploring the

342

ROLANDO AVENDAN˜O AND JAVIER SANTISO

past, anticipating the future,’’ New York, 2009. Retrieved from http://www.polsci. ucsb.edu/faculty/cohen/working/pdfs/SWF_text.pdf El-Erian, M. (2008). When markets collide: Investment strategies for the age of global economic change (p. 183). New York: McGrawHill. Elson, A. (2008). The sovereign wealth funds of Singapore. World Economics, 9(3), 73–96. Epstein, R., & Rose, A. (2009). The regulation of sovereign wealth funds: The virtues of going slow. University of Chicago Law Review, 76, 111–134Retrieved from http://lawreview. uchicago.edu/issues/archive/v76/76_1/Epstein.pdf Fernandes, F. (2009). Sovereign wealth funds: Investment choices and implications around the world. IMD International, Working Paper (unpublished). Retrieved from http:// www.fcee.ucp.pt/docentes/url/nfernandes/ Fernandes, N., & Bris, A. (2009). New life for sovereign wealth funds: valuable contributors to long term shareholder value. Retrieved from http://www.imd.ch/research/challenges/ TC015-09.cfm Ferreira, M. A., & P. Matos (2007). The color of investors’ money: The role of institutional investors around the world. Journal of Financial Economics, 88(3), 499–533, June 2008. Frot, E., & Santiso, J. (2008). Development aid and portfolio funds: Trends, volatility and allocation. OECD Development Centre Working Paper 275. Retrieved from http:// www.oecd.org/dataoecd/46/52/41890256.pdf Frot, E., & Santiso, J. (2009). Herding in aid allocation. OECD Development Centre Working Paper, 279. Gieve, J. (2008). Sovereign wealth funds and global imbalances. Bank of England (2008), (unpublished). Retrieved from http://www.bankofengland.co.uk/publications/speeches/ 2008/speech339.pdf Hau, H. H., & Rey, H. (2008). Home bias at the fund level. American Economic Review, 98(2), 333–338. Helleiner, H. (2009). The geopolitics of sovereign wealth funds: An introduction. Geopolitics, 14(2), 300–304. Jeanne, O., & Rancie`re, R. (2011). The optimal level of reserves for emerging market countries: Formulas and applications. Economic Journal, forthcoming. Jen, S. (2007a). Sovereign wealth funds: What they are and what’s happening. World Economics, 8(4), 1–7. Jen, S. (2007b). The Definition of a Sovereign Wealth Fund (London, Morgan Stanley Global Research, October, 2007). Retrieved from http://sovereignwealthfunds.files.wordpress. com/2008/01/the-definition-of-a-sovereign-wealth-fund-morgan-stanley-october-2007.pdf Jen, S. (2008). Celebrating the birth of Russia SWF (London, Morgan Stanley Global Research. January). Retrieved from http://sovereignwealthfunds.files.wordpress.com/2008/02/ celebrating-the-birth-of-russias-swf-morgan-stanley-janury-2008.pdf. Jensen, N. (2006). Nation states and the multinational corporation: A political economy of foreign direct investment (Princeton University Press, Princeton). Jensen, N. (2009). Firms level response to politics: Political institutions and the operations of U.S. multinationals, unpublished. Retrieved from http://nathanjensen.wustl.edu/research.html. JP Morgan (2008). Sovereign wealth funds: a bottom-up primer. JP Morgan Research, June Report. Kennan, P. J. (2009). Sovereign wealth funds and social arrears: Should debts to citizens be treated differently than debts to other creditors. Retrieved from http://works.bepress. com/cgi/viewcontent.cgi?article¼1004&context¼patrick_keenan

Are Sovereign Wealth Funds Politically Biased?

343

Kern, S. (2009). Sovereign wealth funds – global update. Deutsche Bank Report. Klein, A., & Zur, E. (2006). Entrepreneurial shareholder activism: hedge funds and other private investors. New York University Working Paper (2007). Kotter, J., & Lel, U. (2010). Friends or foes? The stock price impact of sovereign wealth fund investments and the price of keeping secrets. Journal of Financial Economics, 101(2), 360–381. Li, Q., & Resnick, A. (2003). Reversal of fortunes: Democratic institutions and foreign direct investment inflows to developing countries. International Organization, 57, 175–211. Monitor Group (2009). Monitor Group and Fondazione ENI Enrico Matteri, Sovereign Wealth Fund Investment Behaviour: Analysis of Sovereign Wealth Funds Transactions During Q2 2009 (Cambridge, MA and Milan). Retrieved from http://www.monitor.com/Portals/0/ MonitorContent/documents/Monitor_SWF_Q1_2009_Report.pdf Monk, A. (2008a). Recasting the sovereign wealth fund debate: trust, legitimacy and governance. Boston College. Ochoa, C., & Keenan, P. (2009). The human rights potential o sovereign wealth funds. Indiana University Maurer School of LawResearch Paper 132 (unpublished). OECD (2008a). Freedom of Investment, National Security and ‘‘Strategic’’ Industries: Progress Report by the Investment Committee. Retrieved from http://www.oecd.org/dataoecd/ 34/9/40408735.pdf OECD (2008b). OECD, Sovereign Wealth Funds and Recipient Countries: Working Together to Maintain and Expand Freedom of Investment. Retrieved from http://www.oecd.org/ dataoecd/0/23/41456730.pdf Papaioannou, E., Portes, R., & Siourounis, G. (2006). Optimal currency shares in international reserves: The impact of the euro and the prospects for the dollar. Journal of the Japanese and International Economies, Elsevier, 20(4), 508–547. Reisen, H. (2008). How to spend it: Commodity and non-commodity sovereign wealth funds. Deutsche Bank Research Notes (28). Rozanov, A. (2005). Who holds the wealth of nations? Central Banking Journal, 15(4), 52–57. Rozanov, A. (2007). Sovereign wealth funds: Defining liabilities, State Street Global Advisors. Retrieved from http://www.ssga.com/library/esps/Soverign_Wealth_Funds_Andre_ Rozanov_4.27.07rev2CCRI1182371372.pdf Rozanov, A. (2008). A liability-based approach to sovereign wealth. Central Banking Journal, 18(3), 15–20. Setser, B., & Ziemba, R. (2009). GCC sovereign wealth funds: Reversal of fortune, Council on Foreign Relations Working Paper. Retrieved from http://www.cfr.org/content/publications/ attachments/CGS_WorkingPaper_5.pdf Sun, T., & Hesse, H. (2009). Sovereign wealth funds and financial stability: An event study analysis, IMF Working Paper 09/239. Truman, E. (2008). Sovereign wealth funds: New challenges from a changing landscape, testimony before the Subcommittee on Domestic and International Monetary Policy, Trade and Technology, Financial Services Committee, U.S. House of Representatives. Retrieved from http://www.iie.com/publications/papers/truman0508.pdf Weinberger, F., & Golub, B. (2007). Asset allocation and risk management for sovereign wealth funds. In: J. Johnson-Calari & M. Rietveld (Eds.), Sovereign wealth management. London, UK: Central Banking Publications. Ziemba, R. (2008). So, what are sovereign wealth funds targeting? Assessing benchmarks, RGE Ecomonitor.

344

ROLANDO AVENDAN˜O AND JAVIER SANTISO

DATABASES AND OTHER RESOURCES: Thomson Datastream. (2010). Thomson Financial 2010. Lionshares. (2010). Factset Research Systems Inc., 2010. Polity IV. Polity IV Project. Political Regime Characteristics and Transitions, 1800–2008. Bertelsmann Transformation Index. (2009). Bertelsmann Institute, 2009.

ANNEX 5 TRUMAN’S INVESTMENT INDICATORS FOR OECD AND NON-OECD SWFS, 2008A (INDEX 0–1)

ANNEXES ANNEX 1 PORTFOLIO CHARACTERISTICS FOR SELECTED SWFS, 2009 Fund

Holder Manager Cap Turnover Average Average Average Average Price Relative Beta Style Style Group P/E P/B Div Sales Momentum Strength Style Ratio Ratio Yld (%) Growth (%)

New Mexico Alabama Retirement System

GARP Specialty Yield GARP

N/A N/A

0 11.41

0 2.59

0 5.48

0 17.12

0 0.96

0 1

0 0.83

Alaska Retirement Management Board Dubai World Group

Yield

Medium

36.69

1.85

8.88

7.28

0.77

14

1.5

N/A

1.29

0.13

4.29

21.51

0.61

0.69

0.76

N/A

4.92

3.09

7.55

3.65

1.88

1.88

1.56

N/A

7.98

2.2

8.08

63.67

0.84

0.92

0.56

Large Very Cap Low Large N/A Cap

12.25

2.28

5.77

15.05

0.97

1.14

1.01

12.08

2.42

5.26

22.79

0.96

1.08

1.04

Mid N/A Cap Large N/A Cap

19.66

2.87

5.86

14.31

0.98

1.1

0.48

13.73

1.07

6.97

6.86

1.04

1.03

1.32

Emirates Investment Services Ltd. Abu Dhabi Investment Authority (Investment Management) Temasek Holdings Pte Ltd. (Investment Management) Government of Singapore Investment Corp. Pte Ltd. (Invt Mgmt) Saudi Arabian Monetary Agency Qatar Investment Authority (Investment Management)

Yield Yield Yield

Large Cap Generalist Multi Cap Mid Cap Small Cap Mid Cap

Yield Yield

Yield Yield

Specialty

ANNEX 1. (Continued) Oman Arab Bank Investmet Management Group Norges Bank Investment Management Guardians of New Zealand Superannuation Kuwait Investment Co. Khazanah Nasional Bhd. (Investment Management) Alberta Investment Management Corp. Brunei Investment Agency (Investment Management)

Yield Yield

Core Value

Yield Yield Yield

Growth

GARP Generalist GARP

Large Cap Large Cap Multi Cap Small Cap Mid Cap

N/A

8.63

2.06

6.51

40.28

0.88

1.1

0.57

Medium

14.42

2.21

4.25

11.91

0.94

4.43

0.91

N/A

15.26

2.07

4.7

11.37

0.95

5.03

0.88

N/A

9.48

1.28

11.06

45.15

0.53

0.58

0.47

N/A

45.36

1.49

3.77

15.5

0.98

1.13

0.24

0

0

0

0

0

6.32

1.28

0.77

0.96

0.4

N/A Small N/A Cap

0 26.53

0 22.28

Source: Authors’ calculation, based on FactSet and Thomson Financial databases, 2009. Note: Dubai World group and Dubai International Capital were assigned to UAE-Investment Corporation of Dubai. Emirates Investment Services assigned to Emirates Investment Authority. Abu Dhabi Investment Co. and Abu Dhabi Fund for Development assigned to Abu Dhabi Investment Authority. Temasek Ho Chi Min included in Temasek holdings (code 39). Data Malaysian Timber Council not included. Data from Kuwait Investment Office and Kuwait Investment Co included in Kuwait Investment Authority. No fund from Republic of Korea included (SWF not identified). Botswana fund not included (SWF not identified). For Oman fund we used data from December 2008 holdings (only for total SWF). Data for Dubai from Dubai World Group. Data for Abu Dhabi from Abu Dhabi Investment Authority. Data for Kuwait from Kuwait Investment Co. For mutual funds, large funds not included: Vanguard Emerging Markets Stock Index Fund, American Funds American Mutual and Dodge & Cox Balanced Fund. Historic data on their holdings not available.

ANNEX 2 PORTFOLIO CHARACTERISTICS FOR SELECTED MUTUAL FUNDS, 2008 Name

PIMCO Convertible Fund American Funds AmCap Fund Vanguard Total International Stock Index Fund Fidelity Low Priced Stock Fund Fidelity Magellan Fund iShares MSCI Emerging Markets Index Fund Fidelity Growth Company Fund Dodge & Cox International Stock Fund Fidelity Advisor Aggressive Growth Fund Fidelity Diversified International Fund Dodge & Cox Stock Fund American Funds Fundamental Investors American Funds New Perspective Vanguard Wellington Fund Franklin Income Fund American Funds American Balanced

Holder Style

Cap Group Style

Turnover Average P/ Average P/ Average Div E Ratio B Ratio Yld (%)

Average Sales Growth (%)

Price Momentum

Relative Beta Strength

Value Value

Multi Cap Large Cap

Medium Medium

7.78 17.3

1.78 2.67

0.8 1.97

23.86 13.5

1.7 1.2

39.23 2.28

0.86 1.5

Index

Large Cap

N/A

14.31

2

4.74

12.46

0.94

1.78

0.94

Deep Small Cap Value Value Large Cap Index Large Cap

High

12.3

2.8

1.7

8.36

Very low Very low

16.86 14.65

2.72 2.7

1.5 2.54

GARP

Large Cap

Very low

24.3

4.55

Yield

Large Cap

Medium

15.71

Growth

Mid Cap

High

Yield

Large Cap

Deep Large Cap Value GARP Large Cap Yield Yield Yield Yield

1

15.54

1

1.91 15.69

1.6 1.7

24.57 14.28

1.1 1.5

1.39

28.15

1.6

24.19

0.89

1.6

4.29

6.22

0.96

5.5

1.7

21.85

4.18

0.58

21.9

1.1

28.29

0.87

Low

14.49

2.63

3.56

13.46

0.97

7.41

0.95

Low

13.31

1.71

2.84

4.74

0.95

22.9

1.1

Medium

15.9

3

2.77

12.4

1.1

16.82

0.96

Large Cap

Medium

14.1

2.74

3.23

11.62

1

8.88

0.94

Large Cap Large Cap Large Cap

Low Medium Low

12.62 12.22 13.27

2.82 1.48 2.96

3.46 5.46 3.9

8.45 4.63 8.2

16.44 9.13 13.55

0.88 0.76 0.93

0.98 0.91 0.95

ANNEX 2. (Continued) Vanguard Institutional Index Fund Fidelity Contrafund American Funds Investment Company of America American Funds Income Fund of America American Funds Capital World Growth & Income American Funds EuroPacific Growth Vanguard 500 Index Fund American Funds Capital Income Builder Vanguard Total Stock Market Index Fund

Index

Large Cap

Very low

13.86

3.5

2.95

1.31

0.97

8.28

0.93

GARP Yield

Large Cap Large Cap

Low Low

19.5 12.91

4.8 3.18

1.55 3.77

17.8 9.87

1 0.97

18.5 13.22

0.74 0.92

Yield

Large Cap

Medium

13.75

2.83

5.53

1.95

0.91

6.46

0.86

Yield

Large Cap

Low

12.6

2.57

5.6

12.41

0.95

4.28

0.93

Yield

Large Cap

Medium

14.73

2.41

3.89

11.71

0.98

4.6

0.93

Index Yield

Large Cap Large Cap

Very low Medium

13.86 12.92

3.5 2.63

2.95 5.87

1.3 12.8

0.97 0.91

18.28 2.98

0.93 0.88

Index

Large Cap

Very low

14.85

2.98

2.71

11.71

0.97

19.45

0.93

Source: Authors’ calculation, based on FactSet and Thomson Financial databases.2009. Note: For the sake of argument, we included some financial definitions for indicators in Fig. 1 and Annex 2. The price-to-earnings ratio is the valuation of a company’s current share price compared to its per-share earnings, and is calculated as the ratio between the market value per share and the earnings per share; in general, a high P/E ratio indicates that investors are expecting higher earnings growth in the future. This ratio is usually compared to other companies in the same industry, or the market in general. The price-to-book ratio is used to compare a stock’s market value to its book value. It is calculated by dividing the current closing price of the stock by the latest quarter’s book value per share. The Dividend Yield shows how much a company pays out in dividends each year relative to its share price. The dividend yield is calculated as the ratio between the annual dividends per share and the price per share. The Average Sales Growth indicates the percentage change in sales over a certain period. The Price Momentum, highly regarded by investors, indicates the rate of acceleration of a stock’s price. The Relative Strength is a measure of price trend that indicates how a stock is performing relative to other stocks in its industry and it is calculated dividing the price performance of a stock by the price performance of an appropriate index for the same time period. The financial beta is a measure of the volatility, or systematic risk, of a security or a portfolio in comparison to the market as a whole.

ANNEX 3 SECTOR AND INDUSTRY CLASSIFICATION Sectors

Industries

Finance Industrial Services

Major Banks Specialty Telecommunications

Health Technology Energy Minerals Consumer Nondurables Retail Trade Utilities Consumer Services Consumer Durables Technology Services Producer Manufacturing Communications Transportation Process Industries Commercial Services Miscellaneous Electronic Technology Distribution Services Health Services Nonenergy Minerals Government

Pharmaceuticals: Major Regional Banks Semiconductors Wireless Telecommunications Integrated Oil Major Banks Real Estate Development Electric Utilities Packaged Software Steel Telecommunications Equipment Multi-Line Insurance Electrical Products Real Estate Investment Trusts Electronic Production Equipment Gas Distributors Electronics/Appliance Stores Motor Vehicles Oil & Gas Production Cable/Satellite TV Investment Trusts/Mutual Funds Medical Specialties Precious Metals

Department Stores Electronics/AppliancesAdvertising/ Marketing Services Homebuilding Household/Personal Care Tobacco Apparel/Footwear Home Improvement Chains Trucks/Construction/Farm Machinery Broadcasting Publishing: Newspapers Other Metals/Minerals Internet Software/Services Computer Peripherals Wholesale Distributors Apparel/Footwear Retail Aerospace & Defense Life/Health Insurance Beverages: Nonalcoholic Biotechnology Other Transportation Electronic Equipment/Instruments Insurance Brokers/Services Restaurants Home Furnishings Industrial Machinery

Office Equipment/Supplies

Industrial Specialties Food Distributors Pharmaceuticals: Other Pharmaceuticals: Generic Data Processing Services Chemicals: Major Diversified Industrial Conglomerates Auto Parts: OEM Hospital/Nursing Management Trucking Savings Banks Recreational Products Personnel Services Water Utilities Computer Communications Containers/Packaging Medical/Nursing Services Environmental Services Forest Products Miscellaneous Manufacturing Commercial Printing/Forms Medical Distributors Aluminum

ANNEX 3. (Continued) Engineering & Construction Investment Managers Investment Banks/Brokers Contract Drilling Finance/Rental/Leasing Computer Processing Hardware Coal Miscellaneous Commercial Services Chemicals: Specialty Airlines Food: Specialty/Candy Construction Materials Beverages: Alcoholic Oil Refining/Marketing Specialty Stores

Financial Conglomerates Food: Meat/Fish/Dairy Electronic Components Food Retail Other Consumer Services Railroads Marine Shipping Drugstore Chains Managed Health Care Media Conglomerates Agricultural Commodities/Milling Information Technology Services Oilfield Services/Equipment Hotels/Resorts/Cruiselines Food: Major Diversified Pulp & Paper Air Freight/Couriers Tools & Hardware Property/Casualty Insurance Chemicals: Agricultural

Source: Avendan˜o and Santiso (2009) based on FactSet database and Thomson Financial, 2009.

Casinos/Gaming Financial Publishing/Services Metal Fabrication Electronics Distributors Automotive Aftermarket Publishing: Books/Magazines Other Consumer Specialties Catalog/Specialty Distribution Building Products Specialty Insurance Movies/Entertainment Textiles Consumer Sundries Discount Stores Alternative Power Generation Oil & Gas Pipelines General Government Miscellaneous Internet Retail Services to the Health Industry

Variable

Description

Polity Fragmentation

This variable codes the operational existence of a separate polity, or polities , comprising substantial territory, and population within the recognized borders of the state and over which the coded polity exercises no effective authority (effective authority may be participatory or coercive) Additive 11-point scale (0–10). The operational indicator of democracy is derived from coding of the competitiveness of political participation Defined in terms of the presence of a distinctive set of political characteristics. Constructed additively. The operational indicator of autocracy is derived from coding of the competitiveness of political participation, the regulation of participation, the openness, and competitiveness of executive recruitment Defined as the difference between the Institutionalized Democracy and the Autocracy s core Regulation refers to the extent to which a polity has institutionalized procedures for transferring executive power Competitiveness refers to the extent that prevailing modes of advancement give subordinates equal opportunities to become superordinates Openness of Executive Recruitment: Recruitment of the chief executive i s ‘‘open’’ to the extent that all the politically active population has an opportunity, in principle, to attain the position through a regularized process Operationally, this variable refers to the extent of institutionalized constraints on the decision making powers of chief executives, whether individuals or collectivities Existence of binding rules on when, whether, and how political preferences are expressed The competitiveness of participation refers to the extent to which alternative preferences for policy and leadership can be pursued in the political arena Combined information of the following components: Regulation of Chief Executive Recruitment, Competitiveness of Executive Recruitment and Openness of Executive Recruitment Indicator of authority patterns

Institutionalized Democracy (0–10) Autocracy

Polity Score Regulation of Chief Executive Recruitment Competitiveness of Executive Recruitment Openness of Executive Recruitment: Executive Constraints (Decision Rules ) Regulation of Participation The Competitiveness of Participation Executive Recruitment Political Competition

351

Source: Avendan˜o and Santiso (2009), based on LionShares, Thomson Financial and Polity IV Project (2009).

Are Sovereign Wealth Funds Politically Biased?

ANNEX 4 POLITICAL REGIMES AND FUND INVESTMENTS: DEFINITION OF POLITICAL VARIABLES

ROLANDO AVENDAN˜O AND JAVIER SANTISO

352

Average Emerging SWFs

Average OECD SWFs

1 0.9 0.8

Index 0-1

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Transparency

Investment Strategy

Benchmarks

Credit Ratings

Specific Investments

Source: Truman (2008). a

SWFs included in the survey are the following: Nonpension Funds: Algeria Revenue Regulation Fund, Azerbaijan State Oil Fund of the Republic, Botswana Pula Fund, Brunei Darussalam Brunei Investment Agency, Canada (Alberta) Alberta Heritage Savings Trust Fund, Chile Economic and Social Stabilization Fund, China Investment Corporation, Hong Kong Exchange Fund, Iran Oil Stabilization Fund, Kazakhstan National Fund for the Republic of Kazakh, Kiribati Revenue Equalization Reserve Fund, Korea Investment Corporation, Kuwait Investment Authority, Malaysia Khazanah Nasional, Mexico Oil Income Stabilization Fund, Nigeria Excess Crude Account, Norway Government Pension Fund – Global, Oman State General Reserve Fund, Qatar Investment Authority, Russia Reserve Fund and National Wealth Fund, Sa˜o Tome´ and Prı´ ncipe National Oil Account, Singapore Government of Singapore Investment Corporation, Singapore Temasek Holdings, Sudan Oil Revenue Stabilization Account, Timor-Leste Petroleum Fund for Timor-Leste, Trinidad and Tobago Heritage and Stabilization Fund, United Arab Emirates (Abu Dhabi) Abu Dhabi Investment, United Arab Emirates (Abu Dhabi) Mubadala Development, United Arab Emirates (Dubai) Istithmar World, Alaska Permanent Fund, New Mexico Severance Tax Permanent, Wyoming Permanent Mineral Trust Fund, Venezuela Macroeconomic Stabilization Fund , Venezuela National Development Fund. Pension Funds: Australia Future Fund, Canada Pension Plan, Canada (Que´bec) Caisse de de´poˆt et placement du Que´bec, Chile Pension Reserve Fund, China National Social Security Fund, Fonds de re´serve pour les retraites (France), Ireland

353

Are Sovereign Wealth Funds Politically Biased?

National Pensions Reserve Fund, Japan Government Pension Investment Fund, Netherlands Stichting Pensioenfonds ABP, New Zealand Superannuation Fund, Thailand Government Pension Fund, California Public Employees Retirement System.

ANNEX 6 TRUMAN’S INVESTMENT INDICATORS FOR COMMODITY AND NONCOMMODITY SWFS, 2008 (INDEX 0–1) Average Non-commodity SWFs

Average Commodity SWFs

1 0.9 0.8

Index 0-1

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Transparency

Investment Strategy

Source: Truman (2008).

Benchmarks

Credit Ratings

Specific Investments

SOVEREIGN WEALTH FUND ACQUISITIONS: A COMPARATIVE ANALYSIS WITH MUTUAL FUNDS Narjess Boubakri, Jean-Claude Cosset and Nabil Samir STRUCTURED ABSTRACT Purpose – Run a comparative analysis between investments of sovereign wealth funds (SWFs) and mutual funds, focusing on firm-level, country-level, and institutional variables. Methodology/approach – We use a hand-collected sample of 1,845 acquisitions around the world over the last 25 years (251 for SWFs and 1,594 for mutual funds). We then run univariate parametric and nonparametric tests to assess the differences in the investments of both subsamples. Findings – We review the literature on the determinants of SWFs’ investment decisions. Our analysis adds to the scarce available literature on the investment decisions of SWFs and their comparison with other institutional investors. Our results show that, compared to mutual funds, SWFs indeed exhibit different preferences: for instance, SWFs prefer to acquire stakes in larger, less liquid companies which are financially distressed but which also have a higher level of growth opportunities. They Institutional Investors in Global Capital Markets International Finance Review, Volume 12, 355–389 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1569-3767/doi:10.1108/S1569-3767(2011)0000012016

355

356

NARJESS BOUBAKRI ET AL.

also prefer less innovative firms with more concentrated ownership, which are located in less developed but geographically closer countries with whom they do not necessarily share cultural and religious backgrounds. Social implications – Our results are important for practitioners and firms seeking to attract a given type of institutional investment. They also add insights to the debate on the ‘‘hidden’’ political objectives behind SWF investments in the Western world. Originality/value of paper – This is the first attempt to empirically assess the differences in the investment choices of SWFs and mutual funds. Keywords: Sovereign wealth funds; mutual funds; investment decisions Jel classification: G38; G32; G15

INTRODUCTION The recent financial crisis (2007–2010) has drawn attention to a type of institutional investor until then unknown to the general public, namely sovereign wealth funds (SWFs), with their very publicized investments in Citigroup, Bear Stearns, Merrill Lynch, Barclays, UBS, and Morgan Stanley (Bortolotti, Fotak, Megginson, & Miracky, 2009). The particularity of these funds is that they are government owned and draw their funds from natural resources (e.g., the Gulf oil monarchies or Alberta) and/or from the accumulation of foreign reserves because of strong exports (e.g., China and Singapore). These funds have been created to ensure local macroeconomic stability and the welfare of future generations. Although these funds have existed for over half a century, the soaring prices of natural resources and the strength of Asian exports have contributed to their considerable growth over the past five years. Since most of these funds are based in emerging markets, a heated debate among political analysts has arisen regarding sovereign funds’ real motives for investing in developed countries. To some, these countries are trying to diversify their economy, while contributing to global financial stability, particularly during the most recent crisis when they came to the rescue of troubled banks around the world. Other analysts are, however, troubled by

Sovereign Wealth Fund Acquisitions

357

the lack of transparency of most of these funds, and argue that political motives are driving their investments (Kotter & Lel, 2011). In this chapter, we describe these funds, their activities and their investments, and we then run a comparative analysis between the characteristics of SWF and mutual fund investments. At the same time, we review the literature on the determinants of SWF investment decisions. Our analysis adds to the scarce literature available on the investment decisions of SWFs, and compares them with other institutional investors. Our results show that, compared to mutual funds, SWFs indeed exhibit different preferences; for instance, SWFs prefer to acquire stakes in larger, less liquid companies which are financially distressed but which also have a higher level of growth opportunities. They also prefer less innovative firms with more concentrated ownership, which are located in less developed but geographically closer countries with whom they do not necessarily share cultural and religious backgrounds. In what follows, we begin by describing the different categories of SWFs. We then review the available (scarce) literature on the determinants of institutional investors’ choices. We then describe our sample and variables, before we analyze our results. Finally, we conclude and discuss the future prospects and challenges facing SWFs.

DEFINITION AND TYPES OF SOVEREIGN WEALTH FUNDS Although some SWFs have existed for over half a century, the term sovereign wealth fund was first introduced in 2005 (Rozanov, 2008). However, there is currently no single clear-cut definition of these funds (Bean, 2009): it is commonly accepted that these funds are financial investment vehicles that are owned and controlled by a sovereign state and that report to none but that state (Monk, 2008). The literature distinguishes SWFs from other types of public funds administered by the state such as official reserves or public pension funds (Monk, 2008; Rozanov, 2008). For instance, although they are managed by government agencies, the beneficiaries of the proceeds from the investment of pension funds are the participants in the pension plan, and not the state. We propose in this section to introduce the main categories of SWFs, based on the taxonomy developed by the International Monetary Fund (IMF) in its report on SWFs in 2008.

358

NARJESS BOUBAKRI ET AL.

The IMF identifies five categories of sovereign funds: The first class of SWFs, called stabilization funds, aims to neutralize the impact of a massive outflow of capital that would trigger inflation in the economy. They can also be employed, in the case of capital flows related to natural resource revenues, to offset the volatility of these revenues by smoothing income over time. Indeed, stabilization funds serve as a buffer between inflows in an economy and their assignment to the state budget. Thus, all incoming flows will first be allocated to the stabilization fund that will in turn allocate them so as to stabilize the state budget. Any discontinuity or volatility will thus be absorbed by the stabilization fund (Johnson, 2007). The second category of funds manages currency reserves. These funds are intended to more optimally invest the foreign exchange reserves accumulated by a country. Thus, monetary authorities who maintain official foreign exchange reserves to support their monetary policies will typically invest in assets with a low risk profile, strong liquidity, and short horizons (e.g., U.S. Treasury bills). If the accumulated foreign exchange reserves exceed the needs of the reserves, they can be invested through SWFs in higher risk profiles and less liquid assets over longer horizons, fostering a return on investment that is higher than 3–5% (Balin, 2009), while at the same time lowering the cost of holding official reserves. The objective of the third category, called development funds, varies depending on the economic objectives of the country (Chhaochharia & Laeven, 2009). For instance, for countries that are highly dependent on oil revenues, development funds can be used for diversification. In this vein, Gulf countries are investing heavily in sectors such as infrastructure, industry, and finance to lessen their dependence on oil revenues. Similarly, exporting economies such as the Asian countries tend to invest their development funds in their national champions, but also in regional economies with whom they maintain strong commercial ties (Johnson, 2007). The savings management funds, which form the fourth category, are designed to preserve the accumulated current wealth for future generations. Thus, in the case of countries exporting natural resources, accumulated savings from natural resources are invested in diversified assets that generate revenue for the benefit of future generations. These funds work well as mechanisms for intergenerational transfer (Balin, 2009). Finally, funds for future commitments are intended to channel the resources to the benefit of the state. An example is the Australian Future Fund whose purpose is to cover the growing needs of the pension system (SWF Institute, 2009). For this reason, the Australian government decided

Sovereign Wealth Fund Acquisitions

359

to allocate part of its surplus and income from privatization to a dedicated fund, which aims to build wealth to relieve future governments from the weight of the retirement system. It should nevertheless be noted that these objectives are not mutually exclusive. A fund may have several objectives at once, or pursue goals that may evolve over time. Thus, most Middle Eastern sovereign funds, whose surplus is drawn from oil revenues, were originally developed for stabilization purposes, namely to offset the volatility of oil prices on the domestic economy. Subsequently, as the amount of available funds increased, their objectives switched to savings management, economic development, and diversification (Jen, 2007b, Figs. 1 and 2).

Fig. 1.

Fig. 2.

Geographical Distribution of Assets Managed by Sovereign Wealth Funds. Source: Authors’ adaptation based on SWF Institute (2009).

Distribution by Source of Assets Managed by Sovereign Wealth Funds. Source: Authors’ adaptation based on SWF Institute (2009).

360

NARJESS BOUBAKRI ET AL.

THE GROWTH OF SWFs SWFs represent only a small share of the total assets under management by institutional investors. This share is estimated at about 4%, which puts them close to private equity funds (private equity) and hedge funds, but far behind insurance funds, mutual funds, or pension funds (Bortolotti & Miracky, 2009). In recent years, the growth of SWFs has been staggering. Assets under management rose by about 1,000 billion U.S. dollars in 2000 to an estimated 4,000 billion U.S. dollars in 2008. During the same period, no fewer than 29 new SWFs were created (Bortolotti & Miracky, 2009). This strong growth was the result of macroeconomic imbalances that have fueled major SWF inflows. In January 2008, a barrel of oil reached the $100 mark for the first time, far above the $10 per barrel recorded in the 1990s. Moreover, the large trade imbalances have allowed Asian exporters to accumulate foreign exchange reserves in excess of their needs, thereby fostering growth (Figs. 3 and 4). This momentum slowed somewhat during the recent financial crisis that affected the assets of SWFs. Bortolotti et al. (2009) estimate the total losses of SWFs at nearly 57 billion U.S. dollars solely on investments in listed shares. In 2009, total assets under management stagnated at 4,000 billion

Fig. 3. Assets Under Management and Growth Estimate, in U.S.$ billion. Source: Authors’ adaptation based on Bortolotti and Miracky, 2009 (IFSL Estimate).

Sovereign Wealth Fund Acquisitions

361

U.S. dollars, bringing to an end the double-digit growth rate of these funds in previous years. However, this has not dampened the desire of the SWF countries to optimize the use of their wealth, with no less than 11 new SWFs being created in 2009. The future growth of SWFs in the short term is difficult to predict in the current economic climate of volatility and imbalances. The medium and long-term perspectives are also uncertain. However, given the structural nature of SWFs as engines of growth, it is very likely that the global economic recovery will foster a massive return of SWF investments around the world. The recent economic and financial crisis is not the only element of uncertainty that influencing the future of SWFs. Indeed, their success also depends on their ability to continue their pace of investment and to convince the host countries that their objectives are purely economic and not at all political. A number of reservations have been raised in recent years regarding this issue and could pose challenges for SWFs (Kotter & Lel, 2011). The main complaint about SWFs is their lack of transparency (Bortolotti et al., 2009). Indeed, with the exception of a few players such as the Norwegian fund, most SWFs do not communicate their investment strategy, nor do they disclose the composition of their assets portfolio. This lack of transparency is amplified by the fact that the largest funds in terms of size, which often come from developing countries, are generally the least

Fig. 4.

Number of Sovereign Wealth Funds Created Per Year Since 1953. Source: Authors’ adaptation based on Bortolotti and Miracky (2009).

362

NARJESS BOUBAKRI ET AL.

transparent. By way of illustration, Beck and Fidora (2008) estimate that the seven largest of the least transparent funds account for more than half of the assets under management by all SWFs. Some politicians argue that the willingness of sovereign funds to maximize the return on their investments is hiding political objectives. As a result, SWFs are facing an increased protectionism by host countries, which on several occasions have denied sovereign funds the possibility of investing domestically (Jen, 2007a). Many examples abound: the failed attempts of the Chinese company CNOOC to take over Unocal and of the UAE DP World to buy P&O in 2005 are revealing in this sense (Bean, 2009). It is to overcome these suspicions that the major sovereign wealth funds (23 in total) met under the auspices of the IMF as part of the International Working Group of Sovereign Wealth Funds (IWG) to agree the Santiago Principles (IWG, 2008). In keeping with the Santiago Principles, some SWFs (e.g., ADIA and CIC) have disclosed information for the first time about their structure, objectives, and investment strategies. However, this disclosure was considered by many to be superficial and insufficient (Fotak, 2010; Hill & Knowlton, 2010). It should be noted that financial markets do not seem to share the fears of politicians – quite the contrary. In fact, the acquisition of a sovereign fund is usually accompanied by an average positive reaction of the market of 1.25%, which is comparable to what is observed for other categories of institutional investors (Bortolotti et al., 2009). Thus, unlike politicians, financial players do not view SWFs as being motivated by political purposes.

DETERMINANTS OF ACQUISITIONS BY INSTITUTIONAL INVESTORS This section presents a review of the literature on the determinants of acquisitions by institutional investors in an international context, with a special emphasis placed on mutual funds and SWFs. Firm Size Institutional investors’ preference for large firms is widely documented in the literature. Indeed, Dahlquist and Robertsson (2001) and Gompers and Metrick (2001) conclude, respectively, in the context of Swedish and U.S. financial markets, that the main determinant of the level of institutional

363

Sovereign Wealth Fund Acquisitions

ownership, compared to other investors, is the size of the targeted firms. Regarding mutual funds in particular, Falkenstein (1996) analyzes the holdings of U.S. funds for the years 1991 and 1992 and notes a preference for large firms. We find the same conclusions in Aggarwal, Klaper, and Wysocki (2005) and Bradshaw, Bushee, and Miller (2004). Ferreira and Matos (2008) point out, for their part, that institutional investors try to minimize the transaction costs and information asymmetries they face in an international context by focusing on large firms. It is therefore not surprising that SWFs show a similar interest in large companies. This has been highlighted in recent empirical studies on the choices of sovereign wealth funds (e.g., Bortolotti et al., 2009; Fernandes, 2009; Kotter & Lel, 2011). A descriptive study by Balding (2008) reveals that out of the four largest SWFs, only Government of Singapore Investment Corporation (GIC) (Singapore), has a relatively balanced proportion of large and small public companies in its portfolio. The other funds analyzed – Temasek (Singapore), Norges Bank Investment Management (NBIM) (Norway), and ADIA (UAE) – are biased toward large firms, with allocations of 75–25%, 85–15%, and 80–20%, respectively. SWFs, concludes the author, are thus minimizing their exposure to the information asymmetry and agency problems that are typically associated with small firms, consistent with the existing finance literature (Kang & Stulz, 1997). As indicated in Kotter and Lel (2011), the preferences of SWFs are similar to those of public pension funds. According to Balding (2008), the preference of SWFs for large companies also stems from a desire to respond to critical comments that have been addressed by some observers. The author asserts that the appeal of large firms reflects the conservative attitude of SWFs. Given that SWFs are not bound – unlike mutual funds – by legal constraints or regulatory authorities (Fernandes, 2009), the average size of SWF holdings is likely to be higher than that of mutual funds, implying that SWFs prefer larger companies than do mutual funds.

Liquidity Institutional investors differ markedly from other players by requiring a higher liquidity from their investments (Gompers & Metrick, 2001) allowing them to convert their investments into cash whenever needed at minimum cost. Analyzing a sample of U.S. equity funds between 1991 and 1992, Falkenstein (1996) reaches the same conclusion with regard to mutual funds. The author explains the preference of this type of institutional investors for

364

NARJESS BOUBAKRI ET AL.

liquidity by their particular sensitivity to transaction costs. Ferreira and Matos (2008) also note the tendency of mutual funds to focus on liquidity as compared to other institutions like insurance companies and banks. Foreign institutional investors are thus inclined to prefer liquid shares. Gompers and Metrick (2001) show a preference for more liquid firms by large institutional investors. In an international context, Dahlquist and Robertsson (2001) and Tesar and Werner (1995) also find that foreign investors favor more liquid shares whose volume of transaction is higher. Kang and Stulz (1997) share this view and add that the appeal of higher liquid stocks to institutional investors stems from political risk and the possibility of expropriation by the host country that might hinder the repatriating of assets by foreign investors. The attitude adopted by SWFs nonetheless seems to be opposed to that of mutual funds. Indeed, Fernandes (2009) notes that SWFs are not particularly concerned about the liquidity of the firms they acquire. The author argues that liquidity is a feature valued by investors for short-term investments, suggesting that SWFs have long-term investment horizons compared to mutual funds.

Operating Performance According to Ferreira and Matos (2008), institutional investors, including mutual funds, prefer profitable companies that are considered to be safer. As large financial players, institutional investors can potentially provide external monitoring of the company’s managers (Gillan & Starks, 2003), thus contributing to better corporate governance. Given that more indebted firms are already benefiting from the supervision of their creditors, Ferreira and Matos (2008) conclude in their study that institutional investors are less interested in leveraged firms. According to these authors, institutional investors prefer to invest in companies with significant cash holdings. Other empirical studies draw similar conclusions to Ferreira and Matos (2008). Kang and Stulz (1997) show that foreign investors in Japan invest in firms with low leverage and satisfactory performance. Aggarwal et al. (2005) find that U.S. mutual funds overweight firms with low levels of debt relative to the benchmark MSCI. Covrig, Ng, and Chan (2005) also note the preference of mutual funds for firms with a high return on equity. Analyzing foreign investors in Sweden, Dahlquist and Robertsson (2001) argue that their preferences are tilted toward cash-rich companies. Kotter and Lel (2011), for their part, show that SWFs, unlike other institutional investors, prefer to invest in financially distressed companies. By comparing firms owned by

Sovereign Wealth Fund Acquisitions

365

SWFs to all other firms covered in Worldscope without SWF participation, the authors document the preference of these funds for firms that exhibit financial difficulties, and explain that these firms are valued by SWFs to signal that their intentions are not political in nature. In summary, by preferring companies with weak financial performance, SWFs differ markedly from other institutional investors, including mutual funds. Stock Price Certain categories of investors have shown a significant preference for shares whose price has recently experienced a significant appreciation. These choices are based on momentum investment strategies that have historically resulted in positive abnormal returns (Gompers & Metrick, 2001). Grinblatt and Keloharju (2000) analyzed all the transactions performed in Finland between 1995 and 1996 to examine this issue and clearly show the link between the level of sophistication of the investor and the latter’s adoption of a momentum investment strategy. The authors note that those investors more likely to use the recent stock market performance as a criterion for acquisition on the Finnish market are foreign institutional investors, mutual funds, hedge funds, and investment banks. These findings are also shared by Froot, O’Connell, and Seasholes (2001) and Choe, Kho, and Stulz (1999). Grinblatt and Keloharju (2000) also point out that less sophisticated investors adopt the opposite strategy to that of foreign institutional investors, and focus primarily on recently devalued stocks rather than those with an upward tendency (contrarian investment strategy). The authors’ conclusions, based on the Finnish market, were later validated in an international context by Tesar and Werner (1995), Bohn and Tesar (1996), and Ferreira and Matos (2008). SWFs are perceived to be less sophisticated than other institutional investors (see Grinblatt & Keloharju, 2010’s classification). In practice, Fernandes (2009) finds that SWFs underweigh high value stocks, arguing that this could be due to their long-term investment horizons. Kotter and Lel (2011) concur with this view, noting the preference of SWFs for companies whose price has recently performed negatively. They relate this evidence to the SWF preference for distressed firms. A recent descriptive study by Avendan˜o and Santiso (2009) compares SWFs to mutual funds, noting that mutual funds tend to focus more than SWFs on the recent stock performance as a determinant of their acquisitions.

366

NARJESS BOUBAKRI ET AL.

To summarize, SWFs are more likely, because of their particular profile, to underweight recent stock price performance of the target firm in their investment decision.

Valuation of the Firm Due to their sophistication, institutional investors are more likely, in theory, to identify and exploit a given company’s underpriced shares (Ferreira & Matos, 2008). Lakonishok, Shleifer, and Vishny (1994) show that U.S. institutional investors have historically preferred to invest in ‘‘glamour’’ shares. Ferreira and Matos (2008) explain that the attractiveness of these shares for institutional investors lies in their growth potential. Institutional investors prefer to invest in companies with significant investment opportunities, in line with prudent man rules that are implicitly or explicitly imposed on them by the fiduciary responsibility they carry (Del Guercio, 1996). Foreign investors, primarily institutional investors, show similar preferences, as discussed by Kang and Stulz (1997) in Japan and Dahlquist and Robertsson (2001) in Sweden. Foreign institutional investors prefer high market to equity value ratios. Mutual funds, in particular, exhibit the same behavior (Aggarwal et al., 2005). SWFs seem to differ in this respect. To illustrate this point, Kotter and Lel (2011) report that an Abu Dhabi Investment Authority (ADIA) officer identified stock undervaluation as a main criterion for acquisitions: ‘‘the fund buys stakes of $500 million to $1.5 billion from the World’s 500 Biggest Publicly Traded Companies perceived to have undervalued stocks’’ (McSheehy & Suzuki, 2007). Chhaochharia and Laeven (2009) also share this view. They find that SWFs prefer foreign value stocks with investment opportunities. In summary, SWFs show a greater preference than mutual funds for depreciated shares (value stocks).

Dividend Yield In an international setting, foreign investors prefer capital gains to dividends (Ammer, Holland, Smith, & Warnock, 2005). This contention is confirmed by Dahlquist and Robertsson (2001) and Dahlquist, Pinkowitz, Stulz, and Williamson (2003) on the Swedish market. Similarly, Ferreira and Matos (2008) note a marked aversion by international institutional investors for

Sovereign Wealth Fund Acquisitions

367

shares with high dividend yields. Gompers and Metrick (2001) provide similar evidence for U.S. mutual funds. Analyzing the features of SWF acquisitions, Fernandes (2009) and Kotter and Lel (2011) find that dividend yields do not appear to matter in the selection process of SWFs. Unlike pension funds for example, SWFs do not require a regular income stream from their investments. We can thus expect that unlike mutual funds, SWFs are not likely to prefer high dividend yield firms. Ownership Concentration Several authors, including Choe, Kho, and Stulz (2005) show that American investors tend to avoid concentrated ownership structures, more likely to result in expropriation by insiders. As noted by Leuz, Nanda, and Wysocki (2003), information asymmetries associated with ownership concentration frequently result in higher earnings management and lower quality of accounting information. Dahlquist and Robertsson (2001), based on a sample of firms listed on the Swedish market, show that foreign institutional investors adopt a similar behavior, markedly avoiding companies whose ownership is concentrated. Institutional investors prefer firms with diluted ownership structures, indicative of better governance. Ferreira and Matos (2008) show that mutual funds also prefer such firms. Anecdotal evidence indicates however that SWFs do not exhibit this behavior, as they tend to invest in firms with high ownership concentration. To summarize, ownership concentration, an indicator of corporate governance, does not seem to be a major selection criterion for SWFs, contrary to mutual funds. The Level of Investor Protection in the Country The literature agrees on the importance to investors of an environment that provides legal protection against expropriation by insiders. The importance of the level of investor protection is that it is strongly linked to the quality of financial information (Leuz et al., 2003). Weak investor protection is a significant barrier to investment. Higher investor protection by contrast minimizes the information asymmetries and agency problems that result from a weak legal environment and has been shown to be a major determinant in developing financial markets (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2000). Countries whose legal origin is the common law system provide a better protection for shareholders (La Porta, Lopez-de-Silanes, Shleifer, & Vishny,

368

NARJESS BOUBAKRI ET AL.

1998), and hence exhibit more developed financial systems. Ferreira and Matos (2008) empirically show that institutional investors, particularly mutual funds, prefer to invest in countries where minority shareholders are better protected. In the same vein, the level of investor protection is shown by Khorana, Servaes, and Tufano (2005) to be the major determining factor that explains cross-country differences in the size of the mutual fund industry. In reviewing the international allocations of U.S. mutual funds, Aggarwal et al. (2005) note that U.S. mutual funds invest more, compared to the MSCI International Index, in companies that operate in countries with high investor protection. Fernandes (2009) shows that SWFs are comparable to mutual funds in this respect. The author notes their significant preference for companies operating in countries with better legal institutions. This seems surprising given that most SWFs originate from countries with weak legal and institutional environments. However, this can be explained by the willingness of these funds to diversify the risk profile of their political and institutional assets. To summarize, mutual funds and SWFs value the level of investor protection in the host country.

The Level of Economic Development of the Host Country The level of economic development in the host countries (i.e., where the target companies are based), as measured by GDP per capita, has only a slight influence on institutional investors’ decisions (Ferreira & Matos, 2008). Evidence on mutual funds appears in Covrig et al. (2005) and Aggarwal et al. (2005). With regard to SWFs, Fernandes (2009) asserts that the level of economic development in the host country is not a determining factor in the investment decisions of SWFs. The characteristics of the firms and the quality of institutions in the country appear to be more relevant.

The Level of the Financial Development of the Host Country The level of financial development, measured in the literature by the ratio of the domestic market capitalization to GDP, plays a role in attracting investors because it reflects lower transaction costs. Covrig et al. (2005) show, for example, that mutual funds prefer to invest in developed financial markets. SWFs, according to Kotter and Lel (2011), show similar

Sovereign Wealth Fund Acquisitions

369

preferences to mutual funds by overweighting companies that operate in more developed equity markets. Geographic Proximity of the Company Financial theory says that investors have a preference for investments that are geographically (physically) close (Coval & Moskowitz, 2001 in the U.S. market and Grinblatt & Keloharju, 2001 in the Finnish market). The rationale behind these results is that investors prefer companies with which they are more familiar, hence minimizing the asymmetry of information to which they are exposed, the cost of acquiring information being proportional to the distance between the investor and his investment. The importance of geographical proximity also applies in a regional and international context. Portes and Rey (2005) confirm this conjecture by examining the international flow of equity investment. Tesar and Werner (1995) show, for example, a greater emphasis by U.S. investors on Canadian companies in their international portfolios. Mutual funds in particular, as Covrig et al. (2005) note, exhibit a clear regional bias, as they tend to invest more in neighboring countries than in other markets. The empirical literature also reveals that SWFs follow the same behavior. Indeed, the descriptive analysis presented by Balding (2008) shows, for example, that over 80% of investments by the three largest public SWFs in the world (ADIA, Temasek, and NBIM) are regional, that is to say respectively in the Middle East, Southeast Asia, and the West (Europe and United States). In the same vein, Miracky et al. (2008) find that the majority of transactions made by SWFs take place in emerging markets rather than in OECD countries. Avendan˜o and Santiso (2009) concur with this view. They note that the first place where SWF investment is made is Asia, which is also the home of a large proportion of SWFs. This regional concentration of SWF investments seems to contradict the strong media coverage of their stakes in Western companies. These acquisitions in fact seem to provide SWFs with an additional diversification for their portfolios. In conclusion, we can say that SWFs as well as mutual funds prefer to invest in companies that are geographically close. Religious and Cultural Factors According to Stulz and Williamson (2003), culture and religion are the two most significant variables explaining international capital flows. Grinblatt

370

NARJESS BOUBAKRI ET AL.

and Keloharju (2001) validate this assumption in Finland. They show, for instance, that Finnish investors whose native language is Swedish prefer to invest in companies whose managers share the same characteristics. A recent study on international investment allocations by mutual funds shows that, similar to other institutional investors, mutual funds prefer to invest in markets with which they share similar cultural considerations, typically those resulting from a common language (Covrig et al., 2005). With regard to SWFs, Chhaochharia and Laeven (2009) show a similar relationship. The authors believe that cultural and religious factors are particularly relevant because of the state influence on these investors. The influence of religious affiliation largely explains, according to Chhaochharia and Laeven (2009), SWFs’ attraction to some companies, particularly in comparison with the choices made by U.S. pension fund CALPERS and U.S. mutual funds. For example, the favorite investment destinations of Middle Eastern SWFs are Muslim nations. To summarize, cultural and religious affinities appear to be relevant factors to both mutual fund investors and sovereign funds.

Firm Visibility Information asymmetry is an implicit barrier to investment involving additional costs for the researching, processing, and analyzing of data so as to identify appropriate investments (Covrig et al., 2005). The theory of market segmentation (Merton, 1987) stipulates that investors, with information on a limited number of shares, will give priority to firms with which they are familiar. Several measures of recognition by investors have been used in literature, including the level of institutional ownership (Ferreira & Matos, 2008), the number of analysts following the firm (Aggarwal et al., 2005), size (Merton, 1987), and the frequency of citations in the press (Falkenstein, 1996). In an international setting, several additional indicators of visibility are introduced in the literature, including export activity (Dahlquist & Robertsson, 2001) and inclusion in an investment index (Covrig et al., 2005). Ferreira and Matos (2008) argue that mutual funds prefer more visible firms. Bradshaw et al. (2004), analyzing the international choices of U.S. institutional investors, note the importance in these choices of the number of analysts covering a company’s stock. Aggarwal et al. (2005) in this regard maintain that analyst following is a determining factor in the U.S. mutual funds’ investments. These conclusions confirm evidence provided by Falkenstein (1996), who emphasizes

Sovereign Wealth Fund Acquisitions

371

the preference of U.S. mutual fund managers for stocks that are cited more frequently in the press. With respect to SWFs, Kotter and Lel (2011) argue that they show a marked interest for companies with a strong international presence, principally for diversification purposes. In conclusion, SWFs, like mutual funds, show a strong preference for companies with broad international exposure. In what follows, we describe our sample and we present the main results from our univariate analysis of the comparison between SWFs and mutual funds.

SAMPLE AND DESCRIPTIVE STATISTICS We follow Bortolotti et al’s. (2009) approach to choosing the data sources needed to identify the sample of transactions made by SWFs and the control sample of transactions by mutual funds in order to avoid generating a selection bias due to heterogeneous sources. Data on purchases made by SWFs and mutual funds come from the same secondary data sources, namely Zephyr (Bureau Van Dijk) and SDC Platinum M&A (Thomson Reuters). Zephyr includes nearly 500,000 transactions occurring from 1997 to 2011, while SDC M&A includes 672,000 transactions involving U.S. companies acquired over the past 30 years and international firms acquired over the past 25 years. We only take into account the acquisitions of public companies because of lack of financial and accounting information for private firms. Likewise, transactions involving preferred shares, convertible bonds, or swaps are ignored. Out of the 1,845 acquisitions identified in our transaction sample, 251 were made by SWFs and 1,594 by mutual funds. To identify SWFs, we use the IMF list. We thus exclude acquisitions by state-owned enterprises, pension funds (such as CalPERS), and funds managed by states whose ultimate beneficiaries are individuals. In addition, subsidiaries of sovereign wealth funds are included only if their mission is similar to that of their parent company and thus diverges from a strictly operational role. The list of SWFs considered in the study includes the following: ADIA, Alberta Investment Management Corporation (AIMCO), Mumtalakat Holding Company BSC Bahrain, Brunei Investment Agency (BIA), China Investment Corp (CIC), Dubai Holding LLC, Dubai International Group, Dubai World, the Government of Singapore Investment Corporation Private Limited (GIC), Khazanah Nasional Berhad (KNB), Korea Investment

372

NARJESS BOUBAKRI ET AL.

Corporation (KIC), Kuwait Investment Authority (KIA), Libyan Arab Foreign Investment Company (LAFICO), the Libyan Investment Authority (LIA), Mubadala Development Company (MDC), the National Pensions Reserve Fund (NPRF), the New Zealand Superannuation Fund, NBIM, Oman Investment Fund (OIF), Qatar Investment Authority (QIA), the Strategic Investment Fund (SIF), and Temasek Holdings Limited. The selection of mutual funds is based on Avendan˜o and Santiso (2009). Based on this list, we draw from the sources cited above (i.e., Zephyr and SDC M&A) the transactions conducted by these funds between 1985 and 2010 (Table 1).

Descriptive Statistics of the Sample Our final sample includes 251 acquisitions by SWFs and 1,594 acquisitions by mutual funds. In terms of transactions, the average (median) size of SWF transactions is 287.32 (54.81) million dollars compared to 59.48 (11.71) for mutual funds. The acquisitions of SWFs, because of their relative rarity, involve much higher amounts than those of mutual funds. Their investments are thus more concentrated than those of mutual funds, as they are distributed over a smaller number of companies. The same conclusion can be drawn from the analysis of the following figures: the average (median) percentage acquired by SWFs amounts to 14.13% (6.79%). By comparison, mutual funds buy a mean (median) of 3.66% (3.03%). Finally, the average (median) ownership stake held after the transaction is 17.45% (9.01%) for SWFs and 6.66% (5.4%) for mutual funds. The two categories of institutional investors analyzed in this study generally remain minority investors in target companies. This behavior seems more pronounced for mutual funds, as shown in Table 2 using the percentage owned after the transaction. Table 1. Type of Investors

Sovereign wealth funds Mutual funds

Acquisition Data.

Transaction Values ($)

Acquired (%)

Held After Transaction (%)

287.32 (54.81) 59.48 (11.71)

14.13 (6.79) 3.66 (3.03)

17.45 (9.01) 6.66 (5.40)

Source: Authors’ calculations, based on Zephyr and SDC, (2010).

373

Sovereign Wealth Fund Acquisitions

Table 2. Type of Investors Sovereign wealth funds Mutual funds

Distribution of Acquisitions by Size. Minority (o50%)

Influent (20–50%)

Majority (W50%)

79 100

12 0

9 0

Source: Authors’ calculations, based on Zephyr and SDC (2010).

Table 3. Year 1985–1989 1990–1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009–2010

Distribution of Acquisitions by Year. Sovereign Wealth Funds (%)

Mutual Funds (%)

2 16 2 2 2 6 6 9 10 20 12 13

0 4 1 1 3 9 14 16 16 10 6 20

Source: Authors’ calculations, based on Zephyr and SDC (2010).

The analysis of the distribution by year of the sample highlights the rise of SWFs as a new phenomenon in recent years. It also shows their major activities during the recent financial crisis in 2007, as over one fifth of total transactions recorded for SWFs occur in 2007. We also note, from the distribution of transactions, the marked slowdown in 2008 and the impact of economic recovery starting in 2009. The sample of mutual funds shows, in turn, the marked growth of the industry during the 2000s (Table 3). The analysis of the distribution of the SWF transaction sample highlights the dominance of SWFs from Singapore and the UAE (see Table 4). The combined share of the Singaporean Temasek and GIC is 48% of the entire sample (divided equally), followed by the UAE ICD, ADIA, and MDC, which make up 21% of the total transactions. Finally, it is important to note that the underrepresentation of the Norwegian fund is likely due to its diversification policy, which involves the distribution of its assets into thousands of small-sized transactions.

374

NARJESS BOUBAKRI ET AL.

Table 4.

Distribution of Acquisitions by Sovereign Wealth Fund.

Sovereign Wealth Funds

Number of Transactions

% of Sample

61 59 32 17 17 15 13 11 8 3 15

24 24 13 7 7 6 5 4 3 1 6

Temasek GIC ICD ADIA KNB QIA NBIM KIA CIC MDC Others (10 funds)

Source: Authors’ calculations, based on Zephyr and SDC (2010).

Table 5. Distribution of Acquisitions by Host Country. Countries Australia China United States France India Italy Japan Malaysia United Kingdom Singapore Other countries (49)

Sovereign Wealth Funds (%)

Mutual Funds (%)

4 7 7 2 9 4 1 9 8 10 48

7 2 7 5 3 5 10 1 30 1 32

Source: Authors’ calculations, based on Zephyr and SDC (2010).

The geographical distribution of target companies in Table 5 shows that, for mutual funds, the most frequently observed host countries are the United Kingdom (30% of all acquisitions), followed by Japan (10%) and the United States (7%). SWFs on the other hand are mostly interested in Southeast Asian countries exhibiting high economic growth, such as Singapore (10% of total transactions), Malaysia (9%), India (9%), and China (7%). We note that SWFs invest in China and the United States in the same proportions (7% of total transactions respectively). SWF investments seem to be distributed between emerging and developed

375

Sovereign Wealth Fund Acquisitions

Table 6. Region Asia Europe North America Middle East Pacific Latin America Africa

Distribution of Acquisitions by Host Region. Sovereign Wealth Funds (%)

Mutual Funds (%)

51 24 8 6 5 4 2

20 60 9 0 8 3 0

Source: Authors’ calculations, based on Zephyr and SDC (2010).

economies, suggesting that these investments are dictated by diversification objectives as well as the desire to minimize information asymmetries. Regarding mutual funds, Hau and Rey (2008) and Avendan˜o and Santiso (2009), for example, show that the United States is the leading destination for this category of investors. Our sample shows however that the United Kingdom is well ahead of the United States as the favorite destination for acquisitions by mutual fund managers (30% compared to 7%, respectively). Table 6 reports the breakdown of the acquisitions by region. The regional distribution of investments shows that, as stated by Balding (2008) and Chhaochharia and Laeven (2009), SWFs prefer to invest in countries with common cultural or religious values. Mutual funds, mainly from the G20 countries, follow the same trend, as 60% of their transactions take place in Europe. Additionally, we note that the two regions favored by institutional investors are Europe and Asia, but the distribution of acquisitions shows that SWFs invest more in Asia than in Europe while the opposite is true for mutual funds. Finally, the analysis of the industrial distribution of acquisitions in Table 7 indicates that SWFs show, in general, the same preferences as mutual funds, apart from an overweight in the financial industry at the expense of manufacturing and technologies. Description of Variables In our analysis we focus on the following variables: SIZE is the size of the target company as measured by the natural logarithm of total assets in millions of U.S. dollars for the fiscal year preceding the year of the transaction (Bortolotti et al., 2009; Fernandes (2009)).

376

NARJESS BOUBAKRI ET AL.

Table 7. Industrial Sector Consumer goods Finance Manufacturing Basic material Oil & gas Health Consumer services Utilities Technologies Telecommunications

Distribution of Acquisitions by Industrial Sector. Sovereign Wealth Funds (%)

Mutual Funds (%)

10 37 19 6 4 4 11 2 4 3

10 12 22 9 6 6 20 2 11 3

Source: Authors’ calculations, based on Zephyr, SDC and Datastream (2010).

LIQUID represents the liquidity of common stock of the target firm, measured by the average number of shares traded in one day as a percentage of the total number of shares outstanding (Fernandes, 2009; Ferreira & Matos, 2008). PROFIT is profitability measured by return on equity (ROE) for the fiscal year preceding the year of the transaction (Bortolotti et al., 2009; Fernandes, 2009; Ferreira & Matos, 2008). LEV represents the leverage of the target firm measured by total debt as a percentage of total assets (Fernandes, 2009) for the fiscal year before the year of the transaction. CASH is the level of cash available to the target company. It is measured by the amount of cash as a percentage of total assets for the fiscal year preceding the year of the transaction (Fernandes, 2009; Ferreira & Matos, 2008). RETURN is the historic dividend yield of common stock of the target company (Fernandes, 2009; Ferreira & Matos, 2008) calculated with the adjusted price for the year preceding the transaction. PBR is the price to book on the date of the transaction (Avendan˜o & Santiso, 2009). OPP captures the growth opportunities of the target firm as measured by the annual growth of net sales, as in Durnev and Kim (2005) and Ferreira and Matos (2008). DIVY represents the dividend yield, measured on the day of the transaction. The dividend yield is used by Avendan˜o and Santiso (2009), Bortolotti et al. (2009), Fernandes (2009), Ferreira and Matos (2008).

Sovereign Wealth Fund Acquisitions

377

CONC represents the level of ownership concentration of the target company measured on the day of the transaction by the number of shares held by insiders as a percentage of the total number of common shares outstanding (Ferreira & Matos, 2008; Kotter & Lel, 2011). LEG represents the quality of the legal environment of the host country, measured by the Anti-Self Dealing Index developed by Djankov, La Porta, Lopez-de-Silanes, and Shleifer (2008). This index, used by Fernandes (2009), quantifies the level of legal protection afforded to minority shareholders against expropriation by insiders. We used data on the Anti-Self Dealing Index directly from Djankov et al. (2008) for the year 2003. DEV represents the level of economic development of the target country (Fernandes, 2009; Ferreira & Matos, 2008). This indicator of economic development is measured by the natural logarithm of GDP per capita in current U.S. dollars (World Development Indicators) for the year of the transaction. FIN represents the level of financial development of host countries, measured by total market capitalization as a percentage of total GDP. We draw the data from World Development Indicators for the year of the transaction. DIST is the geographical distance between the acquirer and the host country (Ferreira & Matos, 2008). This variable is measured by the natural logarithm of the distance in kilometers between the capitals of the fund and target country. The distances are obtained from Google Maps mapping system. CULT represents the cultural affinity between the acquirer and target country, measured by a dichotomous variable equal to 1 if the language most commonly used in both countries is the same according to the CIA World Factbook 2010. RELIG represents the religious affinity between the acquirer and the target country, and is measured by a dummy variable equal to 1 if both countries practice the same religion according to the CIA World Factbook 2010. ANALYST represents financial visibility, measured by the number of analysts following the firm in of the year of the transaction (Fernandes, 2009; Ferreira & Matos, 2008). The number of financial analysts is extracted from the database IBES (Institutional Brokers’ Estimate System), which aggregates the estimates of financial analysts on nearly 45,000 companies across 70 markets worldwide. VISINT represents the international visibility, measured by sales abroad as a percentage of total net sales for the fiscal year preceding the year of the transaction

378

NARJESS BOUBAKRI ET AL.

INNOV is the level of expertise and innovation of the target company as measured by the intangible assets as a percentage of total assets for the fiscal year preceding the year of the transaction. We use this variable following Kotter and Lel (2011). It is important to note that we preferred to use intangible assets as a measure of the level of expertise and innovation rather than the amount of research and development (Fernandes, 2009) due to the greater availability of data for this measure in Datastream and Worldscope.

UNIVARIATE ANALYSIS In this section, we analyze the differences and similarities between the characteristics of the main sample (SWFs) and the control sample (mutual funds). Specifically, we test whether there is a difference between the variables for each of the subsamples using the t-test of mean difference and the Wilcoxon z-test of median difference. Table 8 displays the results. We note that the size of acquisitions made by SWFs is significantly larger than those by mutual funds (at the 1% level of significance). Indeed, the average (median) total assets reaches 45 billion dollars (1.5) for the main sample (SWF acquisitions), as compared to only 13.64 (0.8) for the control sample (mutual fund acquisitions). As expected, the liquidity ratio mean and median of SWF targets are significantly (at 5% and 1%) lower than those of companies acquired by mutual funds. SWFs seem not to prefer companies that experienced a recent satisfactory financial performance. Three variables are used to validate this assumption. First, the mean and median leverage of firms acquired by SWFs is significantly (at the 5% and 1% level) higher than those of the control sample. The mean and median cash of SWF targets are, in turn, significantly (1%) lower than those of the control sample. There is, however no significant difference between the levels of profitability of the two groups. Looking at the historical average and median of the dividend yield in SWF acquired firms shows that they are lower (at the 5% and 1% level, respectively) than firms in which mutual funds have acquired interests. PBR – The average and median valuation of companies in the main sample is significantly lower than that in the control sample. SWFs thus prefer companies that are more affordable compared to mutual funds. OPP – Because of the preference of SWFs for affordable shares (value stocks) rather than growth stocks, we conjecture that the investment opportunities of SWF targets are lower than in mutual funds targets. The

379

Sovereign Wealth Fund Acquisitions

Table 8. Variables

SIZE LIQUID LEV PROFIT CASH RETURN PBR OPP DIVY CONC LEG DEV FIN DIST CULT RELIG VISFIN VISINT INNOV

Univariate Analysis.

SWFs

Mutual Funds

SWFs vs. Mutual Funds

Mean (Median)

Mean (Median)

t-stat (Wilcoxon z)

7.45 (7.34) 1.30 (0.84) 0.25 (0.24) 11.79% (13.21%) 6.95% (3.65%) 18% (1%) 1.53 (1.67) 32% (15%) 2.29% (1.54%) 36.46% (36.90%) 0.64 (0.65) 9.11 (9.77) 113.70 (92.86) 6.47 (8.33) 0.32 (0.00) 0.39 (0.00) 8.76 (5.00) 31% (19%) 5% (0%)

6.77 (6.78) 1.69 (1.10) 0.22 (0.20) 10.39% (12.04%) 12.63% (6.78%) 30% (17%) 3.09 (2.22) 29% (11%) 2.03% (1.71%) 28.86% (25.52%) 0.64 (0.65) 10.28 (10.53) 112.14 (128.11) 7.16 (8.68) 0.59 (1.00) 0.88 (1.00) 9.29 (8.00) 41% (38%) 15% (5%)

4.25 (3.41) 2.54 (3.17) 2.27 (2.83) 0.63 (1.45) 4.46 (4.77) 2.19 (2.97) 6.17 (4.92) 0.42 (2.03) 1.63 (0.01) 4.28 (2.71) 0.07 (0.56) 16.66 (12.73) 0.38 (2.24) 2.94 (7.06) 7.90 (7.77) 20.68 (18.59) 0.86 (2.22) 3.16 (3.29) 7.08 (9.25)

Our subsamples respectively include 251 targets by SWFs and 1,594 targets by mutual funds. The definition of the variables appears in Appendix B. Significant at the 1% level, significant at the 5% level, significant at the 10% level, respectively.

380

NARJESS BOUBAKRI ET AL.

Wilcoxon z-test shows, with a 5% level of significance, that investment opportunities in acquisitions by SWFs are higher than in those made by mutual funds. DIV – The dividend yield in SWF targets and mutual fund targets is similar while the mean and median ownership concentration in SWF targets are significantly (at the 1% level) higher than those in firms targeted by mutual funds. Finally, looking at the country characteristics, we find the following. There seems to be no significant difference between the level of investor protection in the targets of SWFs and mutual funds. There is however a significant difference in the level of the economic development of the host countries. At the 1% level of significance, the level of economic development identified for the main sample is lower than for the control sample. However, there is no significant difference between subsamples based on the level of financial development. Geographic proximity plays an important role in the choice of investments. Indeed, we find a significant difference (at the 1% level) between the two samples, with the average (median) distance between mutual funds and the companies in which they acquire stakes being higher than that separating SWFs and their targets. This suggests that the geographic distance of the targets is an important factor in the decision of SWFs. Our results also show the existence of a cultural and religious bias in mutual fund investments. Indeed, the mean and median values of RELIG and CULT are significantly higher for the control sample than the original sample, suggesting that mutual funds are more likely to choose targets in countries that are culturally and religiously close than are SWFs. SWFs and mutual funds show the same degree of preference for target visibility as the tests reveal no significant differences between the two samples in terms of the number of analysts following targets. However, the median difference test, which is generally accepted as being more relevant, reports a significant difference (at the 5% level) for the control sample. Using the percentage of export sales, we find it to be significantly (1%) lower for targets of SWFs compared to targets of mutual funds. As expected, the average and median level of innovation and business expertise of firms acquired by sovereign funds are significantly lower (at the 1% level of significance) than those for firms targeted by mutual funds.

Sovereign Wealth Fund Acquisitions

381

To summarize, we find that, compared to mutual funds, sovereign funds prefer to acquire stakes in larger, less liquid companies which are financially distressed but which also have a higher level of investment opportunities. They also prefer less innovative firms with more concentrated ownership, which are located in less developed but geographically closer countries with whom they do not necessarily share cultural and religious backgrounds.

CONCLUSION In this study, we compare the characteristics of firms targeted by SWFs around the world to a control sample of targets acquired by mutual funds (251 for SWFs and 1,594 for mutual funds). Although SWFs are not exactly like other investors, their behavior does not differ markedly. Indeed, SWFs generally act as rational investors: compared to mutual funds, they prefer larger, less liquid, and less innovative public firms with more concentrated ownership. They also prefer to target firms in financial distress, with undervalued stocks and high growth opportunities. In addition, they seem more likely to choose firms that are physically but not necessarily culturally or religiously close, in order to reduce information asymmetry. Our results show that SWFs are not Machiavellian investors that pursue political objectives, contrary to what has been argued by many. Nevertheless, SWFs still face serious challenges to improve their public image and reverse the protectionist trend that affects their investments in Western countries. Examples abound: the failed attempts of the Chinese company CNOOC to take over Unocal and of the UAE DP World to buy P&O in 2005 are revealing in this sense (Bean, 2009). One major issue is the lack of transparency of SWFs (Bortolotti et al., 2009). Indeed, with the exception of a few players such as the Norwegian fund, most SWFs do not communicate their investment strategy, nor do they disclose the composition of their asset portfolio. This lack of transparency is amplified by the fact that the largest funds in terms of size, which often come from developing countries, are generally the least transparent. By way of illustration, Beck and Fidora (2008) estimate that the seven largest of the least transparent funds account for more than half of the assets under management by all SWFs. In an effort to overcome these suspicions, major SWFs from around the world (23 in total) met under the auspices of the IMF as part of the International Working Group of Sovereign Wealth Funds (IWG) to agree on the Santiago Principles (IWG, 2008). In conformity with these

382

NARJESS BOUBAKRI ET AL.

principles, some SWFs (e.g., ADIA and CIC) have disclosed information for the first time about their structure, objectives, and investment strategies. However, this disclosure was considered by many to be superficial and insufficient (Fotak, 2010; Hill & Knowlton, 2010). It should finally be noted that financial markets do not seem to share the fears of politicians – quite the contrary. In fact, the acquisition of a sovereign fund is usually accompanied by an average positive reaction of the market of 1.25%, which is comparable to what is observed for other categories of institutional investors (Bortolotti et al., 2009). Thus, unlike politicians, financial players do not view SWFs as being motivated by political purposes. To alleviate or confirm these concerns, more research is needed on the subject, particularly on the comparative behavior of SWFs and other major institutional investors, their impact on firm valuation and cost of capital, and their spillover effect on financial stability and economic development.

REFERENCES Aggarwal, R., Klaper, L., & Wysocki, P. D. (2005). Portfolio preferences of foreign institutional investors. Journal of Banking and Finance, 29, 2919–2946. Ammer, J. M., Holland, S. B., Smith, D. C., & Warnock, F. E. (2005). Look at me now: What attracts U.S. shareholders? International Finance Discussion Paper no. 815. Board of Governors of the Federal Reserve System, Washington. Avendan˜o, R., & Santiso, J. (2009). Are sovereign wealth funds’ investments politically biased? A comparison with mutual funds. Working Paper no. 283. OCDE Development Centre, Paris. Balding, C. (2008). A portfolio analysis of sovereign wealth funds. Working Paper no. University of California Irvine, California, CA. Balin, B. (2009). Sovereign wealth funds: A critical analysis. Working Paper no. The Johns Hopkins University School of Advanced International Studies (SAIS), Washington. Bean, B. W. (2009). Attack of the sovereign wealth funds: Defending the republic from the threat of sovereign wealth funds? Working Paper no. 08-01, Legal Research Paper Series, Michigan State University of Law, Michigan, MI. Beck, R., & Fidora, M. (2008). The impact of sovereign wealth funds on global financial markets. Occasional Paper Series no.91. European Central Bank, Frankfurt. Bohn, H., & Tesar, L. (1996). U.S. equity investment in foreign markets: Portfolio rebalancing or return chasing?. American Economic Review, 86, 77–81. Bortolotti, B., & Miracky, W. F. (2009). Back on Course-Sovereign Wealth Fund Activity in 2009. Monitor Group-FEEM. Fondazione Eni Enrico Mattei, Milan. Bortolotti, B., Fotak, V., Megginson, W. L., & Miracky, W. (2009). Sovereign wealth fund investment patterns and performance. Working Paper no. 22.

Sovereign Wealth Fund Acquisitions

383

Bradshaw, M. T., Bushee, B. J., & Miller, G. S. (2004). Accounting choice, home bias, and US investment in non-US firms. Journal of Accounting Research, 42, 795–841. Chhaochharia, V., & Laeven, L. (2009). Sovereign wealth funds: Their investments strategies and performance. Working Paper no. 6959. International Monetary Fund, Washington. Choe, H., Kho, B.-C., & Stulz, R. M. (2005). Do domestic investors have an edge? The trading experience of foreign investors in Korea. Review of Financial Studies, 18(3), 795–829. Choe, H., Kho, B.-C., & Stulz, R. M. (1999). Do foreign investors destabilize stock markets? The Korean experience in 1997. Journal of Financial Economics, 54, 227–264. Coval, J. D., & Moskowitz, T. J. (2001). The geography of investment: Informed trading and asset prices. Journal of Political Economy, 109, 811–841. Covrig, V., Ng, L., & Chan, K. (2005). What determines the domestic bias and foreign bias? Evidence from equity mutual fund allocations worldwide. Journal of Finance, 60(3), 1495–1534. Dahlquist, M., & Robertsson, G. (2001). Direct foreign ownership, institutional investors, and firm characteristics. Journal of Financial Economics, 59, 413–440. Dahlquist, M., Pinkowitz, L., Stulz, R., & Williamson, R. (2003). Corporate governance, investor protection and the home bias. Journal of Financial Quantitative Analysis, 38, 87–110. Del Guercio, D. (1996). The distorting effect of the prudent man law on institutional equity investments. Journal of Financial Economics, 40, 31–62. Djankov, S., La Porta, R. L., Lopez-de-Silanes, F., & Shleifer, A. (2008). The law and economics of self-dealing. Journal of Financial Economics, 88, 430–465. Durnev, A., & Kim, E. H. (2005). To steal or not to steal: Firm attributes, legal environment, and valuation. Journal of Finance, 60, 1461–1493. Falkenstein, E. G. (1996). Preferences for stock characteristics as revealed by mutual fund portfolio holdings. Journal of Finance, 51, 111–135. Fernandes, N. (2009). Sovereign Wealth Funds: Investment choices and implications around the world. Working Paper no. IMD International, Lausanne. Ferreira, M., & Matos, P. (2008). The colors of investors’ money: The role of institutional investors around the world. Journal of Financial Economics, 88, 499–533. Froot, K. A., O’Connell, P. G., & Seasholes, M. S. (2001). The portfolio flows of international investors. Journal of Financial Economics, 59, 151–193. Fotak, V. (2010). The failure of self-regulation. Monitor-FEEM Sovereign Wealth Fund Activity Report for 2009. Gillan, S., & Starks, L. (2003). Corporate governance, corporate ownership, and the role of institutional investors: A global perspective. Journal of Applied Finance, 13, 4–22. Gompers, P., & Metrick, A. (2001). Institutional investors and equity prices. Quarterly Journal of Economics, 116, 229–259. Grinblatt, M., & Keloharju, M. (2000). The investment behavior and performance of various investor types: A study of Finland’s unique data set. Journal of Financial Economics, 55, 43–67. Grinblatt, M., & Keloharju, M. (2001). How distance, language, culture influence stockholdings and trades. Journal of Finance, 56(3), 1053–1073. Hau, H., & Rey, H. (2008). Home bias at the fund level. American Economic Review, 98, 333–338. Hill & Knowlton. (2010). Economic recovery ignites interest in sovereign wealth funds but poor sovereign identity may limit investment opportunities. Retrieved from http://www.

384

NARJESS BOUBAKRI ET AL.

hillandknowlton.com/content/economic-recovery-ignites-interest-sovereign-wealth-fundspoor-sovereign-identity-may-limit. Accessed on December 2. IFSL. (2010). IFSL research: Sovereign wealth funds. Retrieved from http://www.ifsl.org.uk/ output/ReportItem.aspx?NewsID¼20. Accessed on December 12. IWG. (2008). Sovereign wealth funds, generally accepted principles and practices. October. Jen, S. (2007a). How big could sovereign wealth funds be by 2015? Morgan Stanley Global Research. Jen, S. (2007b). Sovereign wealth funds: What they are and what’s happening. World Economics, 8(4), 1–7. Johnson, S. (2007). The rise of sovereign wealth funds. Finance & Development, IMF Quarterly Magazine, 44(3). Washington, DC. Kang, J.-K., & Stulz, R. M. (1997). Why is there a home-bias? An analysis of foreign portfolio equity ownership in Japan. Journal of Financial Economics, 46, 3–28. Khorana, A., Servaes, H., & Tufano, P. (2005). Explaining the size of the mutual fund industry around the world. Journal of Financial Economics, 78, 145–185. Kotter, J., & Lel, U. (2011). Friends or foes? Target selection decisions of sovereign wealth funds and their consequences. Journal of Financial Economics, 101, 360–381. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (1998). Law and finance. Journal of Political Economy, 106, 1115–1155. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (2000). Investor protection and corporate governance. Journal of Financial Economics, 58, 3–27. Lakonishok, J., Shleifer, A., & Vishny, R. (1994). Contrarian investment, extrapolation, and risk. Journal of Finance, 49, 1541–1578. Leuz, C., Nanda, D., & Wysocki, P. (2003). Investor protection and earnings management: An international comparison. Journal of Financial Economics, 69, 505–527. McSheehy, W., & Suzuki, H. (2007). Dubai fund buys Sony shares, first Japan investment. Bloomberg News, November 27. Merton, R. (1987). A simple model of capital market equilibrium with incomplete information. Journal of Finance, 43, 483–510. Miracky, W. F., Dyer, D., Fisher, D., Goldner, T., Lagarde, L., & Piedrahita, V. (2008). Assessing the risks: The behaviors of sovereign wealth funds in the global economy. Monitor Group. Monk, A. H. B. (2008). Recasting the sovereign wealth fund debate: Trust, legitimacy, and governance. Working Paper Series, Oxford University, Oxford. Portes, R., & Rey, H. (2005). The determinants of cross-border equity flows. Journal of International Economics, 65, 269–296. Rozanov, A. (2008). A liability-based approach to sovereign wealth. Central Banking Quarterly Journal, XVIII-3, London. Stulz, R. M., & Williamson, R. (2003). Culture, openness, and finance. Journal of Financial Economics, 70(3), 313–349. SWF Institute (2009). Sovereign wealth funds institute. Retrieved from http://www.swfinstitute.org/fund-rankings. Accessed on December 12, 2010. Tesar, L., & Werner, I. (1995). Home bias and high turnover. Journal of International Money and Finance, 14, 467–493.

385

Sovereign Wealth Fund Acquisitions

APPENDIX A SWFs from the Middle East

Country

UAE – Abu Dhabi Saudi Arabia Kuwait Qatar UAE – Dubai UAE – Abu Dhabi UAE – Abu Dhabi Bahrain Oman Saudi Arabia UAE – Ras Al Khaimah UAE-Federal Oman UAE – Abu Dhabi

Name

Assets in M$

Abu Dhabi Investment 627 Authority (ADIA) SAMA Foreign Holdings 415 Kuwait Investment 202.8 Authority (KIA) Qatar Investment 85 Authority (QIA) Investment Corporation of 19.6 Dubai (ICD) International Petroleum 14 Investment Company (IPIC) Mubadala Development 13.3 Company Mumtalakat Holding 9.1 Company State General Reserve 8.2 Fund Public Investment Fund 5.3 RAK Investment 1.2 Authority Emirates Investment ND Authority Oman Investment Fund ND Abu Dhabi Investment ND Council

Source: SWF Institute (2009).

Date of Creation

Source

1976

Oil

ND 1953

Oil Oil

2005

Oil

2006

Oil

1984

Oil

2002

Oil

2006

Oil

1980

Oil and gas

2008 2005

Oil Oil

2007

Oil

2006 2007

Oil Oil

386

NARJESS BOUBAKRI ET AL.

SWFs from Asia

Country

Azerbaijan Brunei

Name

State Oil Fund Brunei Investment Agency China SAFE Investment Company China China Investment Corporation China National Social Security Fund China China-Africa Development Fund China – Hong Hong Kong Monetary Kong Authority Investment Portfolio South Korea Korea Investment Corporation Indonesia Government Investment Unit Iran Oil Stabilisation Fund Malaysia Khazanah Nasional Singapore Government of Singapore Investment Corporation Singapore Temasek Holdings Timor Timor-Leste Petroleum Oriental Fund Vietnam State Capital Investment Corporation Source: SWF Institute (2009).

Assets in M$

Date of Creation

Source

21.7 30

1999 1983

Oil Oil

347.1

1997 Non-commodity

332.4

2007 Non-commodity

146.5

2000 Non-commodity

5

2007 Non-commodity

227.6

1993 Non-commodity

30.3

2005 Non-commodity

0.3

2006 Non-commodity

23 25 247.5

1999 Oil 1993 Non-commodity 1981 Non-commodity

133 6.3

1974 Non-commodity 2005 Oil and gas

0.5

2006 Non-commodity

387

Sovereign Wealth Fund Acquisitions

Other SWFs

Country

Norway Russia Libya Australia

Name

Assets Date of in M$ Creation

Government Pension 512 Fund-Global National Welfare Fund 142.5 Libyan Investment 70 Authority Australian Future Fund 59.1

Algeria

Revenue Regulation Fund Kazakhstan Kazakhstan National Fund United States – Alaska Permanent Alaska Fund Ireland National Pensions Reserve Fund France Strategic Investment Fund Chile Social and Economic Stabilization fund Canada Alberta’s Heritage Fund United States – New Mexico State New Mexico Investment Council New Zealand New Zealand Superannuation Fund Brazil Sovereign Fund of Brazil Botswana Pula Fund United States – Permanent Wyoming Wyoming Mineral Trust Fund

Source

1990

Oil

2008 2006

Oil Oil

2004

56.7

2000

Noncommodity Oil

38

2000

Oil

35.5

1976

Oil

33

2001

28

2008

21.8

1985

Noncommodity Noncommodity Tin

13.8

1976

Oil

12.9

1958

12.1

2003

Noncommodity Noncommodity

8.6

2009

6.9

1994

3.6

1974

Noncommodity Mining and diamonds Mining

2.9

2000

Oil

388

Trinidad and Tobago Venezuela Nigeria Kiribati Mauritania

NARJESS BOUBAKRI ET AL.

Heritage and Stabilization Fund FEM Excess Crude Account Revenue Equalization Reserve Fund National Fund for Hydrocarbon Reserves

0.8 0.5 0.4

1998 2004 1956

Oil Oil Phosphates

0.3

2006

Oil and gas

Source: SWF Institute (2009).

APPENDIX B Name

Definition

SIZE

Size, neperian logarithm of total assets in millions of USD Financial leverage, total debt as a percentage of total assets

LEV

PBR

P/B, price/book value ratio

DIVY

Dividend yield

VISINT

International visibility, foreign sales as a percentage of total net sales Financial visibility, number of financial analysts following the firm’s shares Expertise/innovation: Intangible assets as a percentage of total assets

VISFIN

INNOV

PROFIT

Profitability: Return on equity (ROE)

CASH

Cash as a percentage of total assets

Source Worldscope, WC07230 Worldscope, (WC03255/ WC02999) Worldscope, WC09304 Worldscope, WC09404 Worldscope, WC08731 I/B/E/S

Worldscope, (WC02649/ WC02999) Worldscope, WC08301 Worldscope, (WC02003/ WC02999)

389

Sovereign Wealth Fund Acquisitions

Name

Definition

CONC

Ownership concentration, closely held shares as a percentage of total outstanding ordinary shares Stock returns of target firm calculated with adjusted price Stock liquidity: Number of shares traded in one day as a percentage of total outstanding shares Growth/investment opportunities: Two-year arithmetical mean of the annual net sales growth Economic development of host country: Neperian logarithm of GDP in current USD Financial development of host country: Total market capitalization of host country’s locally listed firms as a percentage of GDP Legal environment: Anti-self dealing index, measure of the legal protection of minority shareholders against expropriation by insiders Geographical distance : Neperian logarithm of the distance in kilometers between the capital of the fund’s home country and the capital of the host country Cultural affinity: A dichotomous variable, equal to 1 if the language most used in the fund’s home country and the host country is the same Religious affinity: Dichotomous variable, equal to 1 if the religion most practiced in the fund’s home country and the host country is the same

RETURN LIQUID

OPP

DEV

FIN

LEG

DIST

CULT

RELIG

Source Worldscope, (WC05475/ WC05301) Datastream, P Datastream, (VO/NOSH) Worldscope, WC08631 World Development Indicators WDI World Development Indicators WDI Djankov et al. (2008)

Google Maps

CIA World Factbook

CIA World Factbook