After the Crash: The Future of Finance

Contributors: John E. Core, Wharton School, University of Pennsylvania ◆ Christopher C. Geczy, Wharton School, University of Pennsylvania ◆ Wayne R. Guay, Wharton School, University of Pennsylvania ◆ Tetsuya Kamiyama, Nomura Institute of Capital Markets Research ◆ Kei Kodachi, Nomura Institute of Capital Markets Research ◆ Alan McIntyre, Oliver Wyman Group ◆ Michael Zeltkevic, Oliver Wyman Group YASUYUKI FUCHITA is a senior managing director at the Nomura Institute of Capital Markets Research in Tokyo. He coedited Prudent Lending Restored (Brookings, 2009) with Richard J. Herring and Robert E. Litan and Pooling Money (Brookings, 2008) with Litan. RICHARD J. HERRING is the Jacob Safra Professor of International Banking and professor of finance at the Wharton School, University of Pennsylvania, where he is also codirector of the Wharton Financial Institutions Center. ROBERT E. LITAN is a senior fellow in Economic Studies at the Brookings Institution and vice president for research and policy at the Kauffman Foundation. His many books include Good Capitalism, Bad Capitalism, and the Economics of Growth Prosperity (Yale University Press, 2007), written with William J. Baumol and Carl J. Schramm.

NOMURA INSTITUTE OF CAPITAL MARKETS RESEARCH Tokyo www.nicmr.com/nicmr/english

Cover design and illustration montage by Claude Goodwin

BROOKINGS / NICMR

BROOKINGS INSTITUTION PRESS Washington, D.C. www.brookings.edu

After the Crash

In After the Crash, noted economists Yasuyuki Fuchita, Richard Herring, and Robert Litan bring together a distinguished group of experts from academia and the private sector to take a hard look at how the financial industry and some of its practices are likely to change in the years ahead. Whether or not you agree with their conclusions, the authors of this volume—the most recent collaboration between Brookings, the Wharton School, and the Nomura Institute of Capital Markets Research—provide well-grounded insights that will be helpful to financial practitioners, analysts, and policymakers.

Fuchita / Herring / Litan

A

s the global economy continues to weather the effects of the recession brought on by the financial crisis of 2007–08, perhaps no sector has been more affected and more under pressure to change than the industry that was the locus of that crisis: financial services. But as policymakers, financial experts, lobbyists, and others seek to rebuild this industry, certain questions loom large. For example, should the pay of financial institution executives be regulated to control risk taking? That possibility certainly has been raised in official circles, with spirited reactions from all corners. How will stepped-up regulation affect key parts of the financial services industry? And what lies ahead for some of the key actors in both the United States and Japan?

After the

Crash The Future of Finance

Yasuyuki Fuchita, Richard J. Herring, and Robert E. Litan, Editors

AFTER THE CRASH

00-0404-1 fm.indd 1

7/12/10 5:58 PM

00-0404-1 fm.indd 2

7/12/10 5:58 PM

yasuyuki fuchita richard j. herring robert e. litan Editors

After the Crash The Future of Finance

nomura institute of capital markets research Tokyo

brookings institution press Washington, D.C.

00-0404-1 fm.indd 3

7/12/10 5:58 PM

Copyright © 2010

the brookings institution nomura institute of capital markets research All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means without permission in writing from the Brookings Institution Press. After the Crash: The Future of Finance may be ordered from:

brookings institution press c/o HFS, P.O. Box 50370, Baltimore, MD 21211-4370 Tel.: 800/537-5487; 410/516-6956; Fax: 410/516-6998 Internet: www.brookings.edu Library of Congress Cataloging-in-Publication data After the crash : the future of finance / Yasuyuki Fuchita, Richard J. Herring, and Robert E. Litan, editors. â•…â•… p.â•… cm. â•… Includes bibliographical references and index. â•… Summary: “Examines the ramifications of the 2007–08 financial crisis on the financial services industry and some of its practices and how these are likely to change in the future”— Provided by publisher. â•… ISBN 978-0-8157-0404-1 (pbk. : alk. paper) â•… 1. Financial institutions.╇ 2. Financial institutions—Deregulation.╇ 3. Financial services industry.╇ 4. Financial crises—United States—21st century.╇ 5. Financial futures.╇ I. Fuchita, Yasuyuki, 1958–╇ II. Herring, Richard.╇ III. Litan, Robert E., 1950–╇ IV. Title. â•… HG173.A38╇ 2010 â•… 332.1—dc22 2010026369 987654321 Printed on acid-free paper Typeset in Adobe Garamond Composition by Cynthia Stock Silver Spring, Maryland Printed by R. R. Donnelley Harrisonburg, Virginia

00-0404-1 fm.indd 4

7/12/10 5:58 PM

Contents

Preface 1 After the Crash: Will Finance Ever Be the Same?

83

Christopher C. Geczy

5 Is There a Case for Regulating Executive Pay in the Financial Services Industry?

53

Kei Kodachi and Tetsuya Kamiyama

4 The Future of the Hedge Fund Industry

11

Alan McIntyre and Michael Zeltkevic

3 Regulatory Changes and Investment Banking: Seven Questions

1

Yasuyuki Fuchita, Richard J. Herring, and Robert E. Litan

2 The Uncertain Future of U.S. Commercial Banking

vii

115

John E. Core and Wayne R. Guay

Contributors

141

Index

143 v

00-0404-1 fm.indd 5

7/12/10 5:58 PM

00-0404-1 fm.indd 6

7/12/10 5:58 PM

Preface

I

n 2004 the Brookings Institution joined with Nomura Institute of Capital Markets Research to showcase research on selected topics in financial market structure and regulation of interest to policymakers, scholars, and market practitioners in the United States and Japan, as well as elsewhere. Initially led by Brookings senior fellow Robert E. Litan and Yasuyuki Fuchita, senior managing director of Nomura Institute of Capital Markets Research, the collaboration was joined in 2008 by Richard J. Herring of the Financial Institutions Center at the Wharton School of the University of Pennsylvania. A conference has been convened each year since 2004, leading to four volumes published by Brookings Institution Press, most recently Prudent Lending Restored: Securitization after the Mortgage Meltdown (2009).1 The chapters in this fifth volume in the series are based on presentations made at a conference, After the Credit Crash: The Future of Finance, held on October 16, 2009, at the Wharton School in Philadelphia. The conference considered the future of the financial services industry after the crisis of 2007–08 and focused on commercial banks, investment banks, and hedge funds in particular. All of the chapters represent the views of the authors and not necessarily those of the staff, 1. The first five of these conferences were sponsored by the Tokyo Club Foundation for Global Studies (now part of Nomura Foundation) and the Brookings Institution.

vii

00-0404-1 fm.indd 7

7/12/10 5:58 PM

viii

preface

officers, or trustees of the Brookings Institution, the Nomura Institute of Capital Markets Research, or the Wharton Financial Institutions Center. The editors thank Adriane Fresh for research assistance and for checking the factual accuracy of the manuscript; Diane Hammond for careful editing; and Lindsey Wilson for organizing the conference and providing administrative assistance. Both the conference and this publication were funded in part by Nomura Foundation.

00-0404-1 fm.indd 8

7/12/10 5:58 PM

yasuyuki fuchita richard j. herring robert e. litan

1

After the Crash: Will Finance Ever Be the Same?

T

he financial crisis of 2007–08, which led to what is now known as the Great Recession of 2008–09, will go down in history as one of the most troubling economic events of the postwar era. Although some prescient analysts forecast that the housing bubble in the United States, which triggered the crisis, eventually would burst, we suspect that few foresaw the crisis bringing the United States and other global economies nearly to their knees. Certainly, no mainstream forecaster or high-profile policymaker predicted this outcome. Even now, after the dust has settled somewhat and a halting recovery is under way, many questions about the future of the global financial services industry remain. After receiving massive government infusions of capital and experiencing large numbers of failures, what will the U.S. commercial banking industry look like in the years ahead? Further, with only one major independent investment bank left in the United States after the crisis, what impact will new regulations have on the investment banking business, under whatever corporate structure it is conducted? The same question can be asked of the hedge fund industry, which went into the crisis largely unregulated. And finally, what is the evidence that the executive compensation structures of some financial companies contributed to the crisis (a criticism leveled by regulators and many in the media)? Should compensation regulation be imposed on the financial services industry? And if so, what form should it take?

1

01-0404-1 chap1 1

7/12/10 4:48 PM

2

yasuyuki fuchita, richard j. herring, and robert e. litan

These are important questions not just for those who own shares in or work for financial services companies but also for the policymakers designing a regulatory framework and for concerned citizens, who fear another disruption of their lives, destruction of their wealth, and the fiscal consequences of government spending on cleaning up after such crises. It is appropriate, then, that these questions were also the subjects of a research conference jointly organized by the Nomura Institute of Capital Markets Research, the Brookings Institution, and the Wharton Financial Institutions Center in October 2009. This volume contains the revised presentations made at the conference, which came just one year after the worst of the crisis unfolded. During the third week in September 2008—after Lehman Brothers declared bankruptcy, after Merrill Lynch fled to safety in the arms of Bank of America, after the Federal Reserve improvised an unprecedented bailout of the creditors of AIG, and after the U.S. Treasury rushed to guarantee the more than $3 trillion held in U.S. money market funds—many observers believed the future of the financial services industry was utterly bleak. We provide in this introductory chapter a summary of the chapters that follow. A broad theme that runs through these chapters is that each of the segments of the financial services industry we review has been significantly affected by the crisis and is likely never to be the same again. Alan McIntyre and Michael Zeltkevic, of Oliver Wyman Group, focus in chapter 2 on the industry in which many of the problems first surfaced, the U.S. commercial banking industry, and examine its future. But first the authors briefly revisit the industry’s recent past, specifically what they call its golden era, the decade between 1993 and 2003. Cognizant of the savings and loan and banking crises of the previous decade, banks during the golden era recapitalized (at the direction of new legislation) and earned returns on equity of roughly 14 percent, the highest of any decade since the 1920s. The industry’s performance began to deteriorate in 2004, however, as the Federal Reserve reversed the loose money policy it had pursued in the wake of the 2000–01 recession. With a flatter yield curve, the spread between bank lending and deposit rates (upon which banks traditionally relied to earn most of their profits) narrowed. To cover their fixed costs, banks turned to asset growth, especially subprime and alt-A mortgage lending to make up the difference.1 Larger 1. Alt-A mortgages (alternative-A paper) are considered riskier than prime mortgages (A paper), because borrowers have less than full documentation, lower credit scores, higher loan-to-value ratios, or more investment properties than prime borrowers. Alt-A mortgages, however, are considered safer than subprime mortgages.

01-0404-1 chap1 2

7/12/10 4:48 PM

will finance ever be the same?

3

banks seemingly hedged the risks of such lending in one of two ways: one, by packaging the loans into securities and then selling them to third-party investors, who might mix them with other securitized assets and resecuritize the pool as collateralized debt obligations; or two, by holding them in theoretically off-balancesheet affiliates, the so-called structured investment vehicles. As we all know now, when the residential real estate market began to sour in 2007, those less-than-prime loans, and the securities that used them for collateral, turned toxic. Banks that had such “assets” on their balance sheets suffered, as did the large institutions that were forced by reputation—and arguably by contractual liquidity arrangements—to provide liquidity or take the structured investment vehicles and their newly troubled assets back onto their balance sheets. As a result, by 2008 the industry’s overall return on equity had turned negative and a number of institutions, large and small, either failed or had to be rescued through arranged mergers. What lies ahead? McIntyre and Zeltkevic projected in October 2009 that banks and insurers around the world still had substantial credit losses to come and not just in securities backed by mortgage loans but also in commercial real estate loans held directly or in the form of securities. But the authors’ focus here is on the period after these losses have been absorbed; they spell out three future scenarios for bank performance. Each scenario reflects different assumptions about the key drivers of performance, including macroeconomic factors, the regulatory environment, and intensity of competition between banks and between banks and nonbanks, including capital markets. In the authors’ baseline scenario, which is the one they believe the most likely, the industry’s return on equity will average 10 percent, or roughly the historical norm over the past seven or eight decades. Their benign scenario sees the industry returning to golden era profitability, with credit losses returning to precrisis levels and the global economy recovering reasonably smartly. In the third scenario banks barely break even, the macroeconomic environment is poor, and U.S. unemployment remains in the 9 percent range. The scenarios depict industry averages, but within any average, some institutions will outperform, others will underperform. What factors are likely to make the difference? The authors suggest that the keys to superior performance include positioning in rapidly growing markets and the ability to manage risks and to take advantage of disruptive change (perhaps through targeted acquisitions). The authors also expect wider dispersion in performance among banks in what they call the new normal (less-buoyant) business environment. The investment banking business was a major casualty of the financial crisis and the Great Recession. Two leading institutions—Bear Stearns and

01-0404-1 chap1 3

7/12/10 4:48 PM

4

yasuyuki fuchita, richard j. herring, and robert e. litan

Lehman—disappeared (one by forced merger, the other by failure). A third, Merrill Lynch, merged with Bank of America, pushed by the Treasury and the Federal Resrve. And the two remaining bulge-bracket investment banks—Goldman Sachs and Morgan Stanley—hastily converted to bank holding company status to ensure access to the Fed’s discount window and to forestall liquidity pressures. At the time of the October conference, Nomura Securities was the only major independent international investment bank standing. In chapter 3 Kei Kodachi and Tetsuya Kamiyama of the Nomura Institute examine the future of the investment banking business through the lens of eight major regulatory changes that, at the time of the conference, were contemplated by the G-20: strengthening the risk-weighted bank capital standards, raising capital charges for banks’ (commercial and investment) trading books, adopting and enforcing leverage ratios, extending certain banking rules to nonbanks, tightening the rules governing securitization, increasing regulation of over-the-counter derivatives, regulating short selling, and regulating hedge funds. The chapter describes each of these initiatives and argues that, in general, these changes individually and collectively would harm investment banking. Accordingly, they offer alternatives that address the perceived problems but in ways the authors believe to be less harmful and more cost effective. For example, higher capital standards in general carry the danger of giving investment and commercial banks incentives to engage in “regulatory arbitrage,” thereby moving such activity to unregulated markets or affiliates. Instead, the authors suggest improvements in risk control by the institutions and the regulators that oversee them. Similarly, higher capital charges for trading activities could be counterproductive by reducing trading activity and therefore liquidity, which could lead to increased price volatility. An alternative approach is to limit the kinds of assets in trading books. A simple leverage ratio ignores the quality of assets on the balance sheet (the reason regulators have adopted risk-based standards) and would have a disproportionately negative impact on countries, such as those in Asia, where banks are more important than markets as providers of credit. The authors suggest instead that country-specific leverage or capital requirements be adopted. Because the crisis has revealed that large nonbanks (such as investment banks) may pose just as much systemic risk as large commercial banks, there is much interest within the G-20 in extending banklike regulation to large nonbanks. But this could lead to the same difficulties as with stringent bank regulation. Accordingly, the authors urge regulators to look for alternatives to the Basel II (risk-based) capital framework for reducing the systemic risks associated with large nonbanks.

01-0404-1 chap1 4

7/12/10 4:48 PM

will finance ever be the same?

5

As for the lack of incentives for prudence in securities origination, the authors argue that there is no need for a mandated “skin-in-the-game” requirement for mortgage originators, because the market, they assert, already imposes such requirements. There are clear systemic risks in over-the-counter derivatives markets, but these can be handled with greater standardization of the instruments traded and greater reliance on central clearinghouses. With respect to naked short selling, the authors see some regulation as inevitable. And with respect to the systemic risk posed by hedge funds, the authors urge greater regulatory supervision of prime brokers and reporting by the funds of their leverage. More broadly, independent of the specific regulatory changes that may be coming, the authors outline a future of the U.S. financial service industry in particular, which contains a mix of financial conglomerates (some services dominated by commercial banks, others by investment banks), megaregional banks, and perhaps a few “pure play” investment banks. Although much concern had been expressed in the years before the financial crisis about the systemic risks posed by hedge funds, these financial institutions in fact played little or no role in the crisis. That does not mean, however, that they will be immune from forward-looking reporting and perhaps regulatory requirements (for those few large funds that regulators deem to be systemically important). At the same time, however, the crisis has had a major impact on the hedge fund industry. Although some funds earned record profits, many incurred substantial losses during the decline in equity markets. And although some of the bleeding stopped when equity prices picked up in the spring of 2009, many hedge funds have closed their doors (or have been obliged to shrink their asset base by their prime brokers, who withdrew much of their leverage). What does the future hold now for hedge funds? In chapter 4, Christopher Geczy of the Wharton School seeks to answer that question, among others: what the consultants who advise institutional investors have been saying about the future; the past and likely future performance of the hedge fund industry; efforts to develop new products aimed at replicating the performance of hedge funds; how hedge fund exposures have changed over time; and the likely changes in regulation and enforcement in the post-Madoff era. The consultants who advise institutional investors, according to a survey conducted by the author in late 2009 and the first quarter of 2010, report significant postcrisis changes in the advice they give their clients. There is much more focus now on transparency of hedge fund activities and risk exposures, greater attention to liquidity, more emphasis on lowering the fees the funds charge, an expectation that hedge funds will be subject to more regulation, and the likelihood that

01-0404-1 chap1 5

7/12/10 4:48 PM

6

yasuyuki fuchita, richard j. herring, and robert e. litan

many investors will want to invest in “hedge fund replicators” rather than in the funds directly. Gezcy’s survey of consultants also reveals that most institutional investors are believed to have limited knowledge of hedge funds. Of those pension funds that allocate some investments to hedge funds, the allocation tends to range between 2 percent and 10 percent. A majority of consultants recommend (and expect to observe) a modest increase in hedge fund investment. Gezcy’s chapter reports more specific results on various provisions in hedge fund arrangements, including the typical length of “lockups” (six to twelve months), risk measures, fees, application of fair value principles to the valuation of hedge fund investments, and concerns about fraud and the quality of due diligence. Gezcy next turns to the performance of hedge funds, beginning by offering a typology of four types of fund, each with different investment strategies. The central conclusion is that one should not measure aggregate performance of all hedge funds because of the differences in risk factors to which they are exposed. The only meaningful comparison is to examine the performance over time of funds in a particular category, adjusted for the risk factors in that category. A recent phenomenon is the development of a new class of funds that seek to replicate the performance of hedge funds of a given type without actually making the investments and following the precise strategies of those funds. A key advantage of replicators is that they can be constructed as mutual funds or exchange-traded funds and thus are open to a much broader class of investors than typical hedge funds, which are open only to sophisticated individuals of means or institutional investors. There are pros and cons to these tracker, or replicator, investment vehicles. Some argue that they are more transparent than the hedge funds they track. Critics argue that replicators generally track only the average or aggregate performance of funds in a category and thus miss out on any superior returns offered by star funds. A key claim of hedge funds is that they in fact offer their investors superior returns, adjusted for risk, to other investment vehicles: alpha, for short. Gezcy evaluates this claim, noting that measures of alpha obviously depend heavily on the measure of risk used. His overall assessment is mixed, with only about half of the funds he studied reporting statistically significant alpha. Gezcy also evaluates through standard statistical techniques the impact of Statement of Financial Accounting Standards 157 on fair value measurements (FAS 157) on hedge funds, which became fully effective in November 2008—or right in the middle of the financial crisis. FAS 157outlines the conditions under which certain assets, such as the kinds of mortgage-backed securities making up

01-0404-1 chap1 6

7/12/10 4:48 PM

will finance ever be the same?

7

many hedge funds, have to be marked to market. Gezcy finds that FAS 157 did, in fact, have a noticeable impact on returns reported by some hedge funds. Gezcy ends his chapter with several tentative conclusions about the future of the hedge fund industry. He shows that private equity strategies and hedge fund strategies have been converging and thinks it likely that the trend will continue. He believes that one of the key lessons hedge fund managers learned from the crisis is the need to place much tighter controls on the mismatch between the duration of their assets and liabilities. Gezcy expresses some skepticism that the interest of consultants in separately managed accounts (as a means of making hedge funds more transparent) will change the structure of the industry, given its passion for secrecy. Although it is tempting to forecast the demise of the traditional compensation standard (2 percent of assets under management and 20 percent of profits above a high-water mark) because of the advent of replication approaches and other more liquid, transparent, and cheaper ways of gaining access to some hedge fund strategies, he concludes that, instead of a compression of fees, we are likely to see a bimodal distribution, in which the hedge fund managers with the best track records will continue to command stratospheric fees while other fund managers will be forced to reduce fees to meet the competition from the cheaper replication techniques. Policymakers around the world and many in the media and among the public in fact blame the compensation structures of financial institutions for creating the crisis or, at the very least, for making it worse. The main purported villain: salaries or bonus arrangements tied in some fundamental way to the volume of business generated (such as mortgage origination), regardless of the downstream or longer-term consequences. Is this criticism correct? And if so, what kinds of compensation regulation might be appropriate? John Core and Wayne Guay of the Wharton School at the University of Pennsylvania take up these questions in chapter 5 and provide some unconventional answers. The authors begin by placing their topic in the larger context of mounting concerns over CEO compensation generally, concerns related in their view to even broader concerns about growing income inequality. There is no doubt that CEO compensation, especially for those heading the largest corporations, is high. But is it too high? By one standard, the answer is—not really: the authors point out that economic theory would suggest that compensation of managers, and CEOs especially, should rise as the size of their entities increases, since larger organizations

01-0404-1 chap1 7

7/12/10 4:48 PM

8

yasuyuki fuchita, richard j. herring, and robert e. litan

tend to be more complex and more difficult to manage. And it turns out that, empirically, CEO compensation is correlated with firm size. Corporate CEO compensation is also not high when compared to compensation of hedge funds and private equity funds (which, in some ways, may be less difficult to manage than the typical large corporation). The authors note that U.S. corporate CEOs do make more than their counterparts in the United Kingdom but that U.S. executives also bear greater equity risks. If, then, there is a plausible defense of the level of CEO pay among U.S. corporations, can the same be said about CEO compensation in the financial services industry? Based on their empirical analysis of the 1992–2006 period, the authors find that both the levels and the composition of the pay packages of financial services CEOs are comparable to those of CEOs at nonfinancial firms. The authors also cite evidence rebutting the common view that financial executives’ compensation is too short-term oriented; to the contrary, the bulk of their compensation consists of stock and options, which the authors argue give the executives a long-term outlook. This finding is consistent with recent evidence that financial executives took heavy losses during the financial crisis and did not cash out in advance. Nonetheless, it is not surprising that, in the wake of the crisis, financial compensation has become an explosive political issue. U.S. Treasury Secretary Timothy Geithner suggested in June 2009 that, going forward, financial institutions should pay their top executives in a manner consistent with a number of key principles. Perhaps the most important of these are that financial executives should be rewarded in relation to the performance of their institutions over the long run, not the short run and, further, that pay practices be aligned with sound risk management of the institutions. The authors find these principles noncontroversial and argue that compensation practices have been largely consistent with them. Still, since mid-2009 the Treasury Department has strictly limited financial executive compensation at institutions that received government money under the Troubled Asset Relief Program (TARP). The authors are critical of a number of aspects of Treasury’s rules in this regard and the ways they have been implemented by the department’s pay czar, Ken Feinberg. For example, the requirement that independent directors approve compensation plans is already required by stock exchange listing standards. The push to have financial executives’ pay consist partially or primarily of restricted stock is puzzling to the authors in light of their evidence that executive compensation already conforms largely to this model. The authors level substantive critiques of other aspects of the pay rules, including the attacks on severance payments, tax gross-ups, and nonbinding

01-0404-1 chap1 8

7/12/10 4:48 PM

will finance ever be the same?

9

say-on-pay votes by shareholders. While not all of the participants at the conference shared the authors’ resistance to greater regulation of financial executive compensation, their chapter marshals the best evidence available supporting the notion that further regulation is unwarranted. The financial crisis of 2007–08 clearly was a watershed event in the financial and economic history not only of the United States but also of the rest of the world. The financial institutions and industries at the heart of that crisis—commercial and investment banking—as well as the hedge fund industry, which some believe could be at the heart of a future crisis, clearly have changed and will undergo more change in the future. We hope that the chapters in this volume shed light on what these changes are likely to be and how, in some cases, current and future policy might affect them.

01-0404-1 chap1 9

7/12/10 4:48 PM

01-0404-1 chap1 10

7/12/10 4:48 PM

alan m c intyre michael zeltkevic

2

The Uncertain Future of U.S. Commercial Banking

C

ommercial banking in the United States since about 1930 has for the most part been a simple business, with adequate but not particularly attractive returns to shareholders. Then for a golden decade, between the early 1990s and the first half of the 2000s, it outperformed, offering high and stable returns to shareholders and accounting for an increasing share of corporate profits. At the time, few sought to identify the confluence of factors that led to such a positive environment for banking or to ask whether that environment was sustainable. With a decline that began somewhat slowly in the middle of the 2000s and then accelerated with the giddy fall of 2008 and 2009, commercial banking underwent a reverse alchemy, turning gold into lead. With eleven straight quarters of declining profitability, enormous industry dislocation, and unprecedented financial support from the government, some may now be tempted to declare commercial banking a bad business, and there are certainly reasons to do so.1 A combination of investor fear, increased market discipline, more intrusive supervision, and stricter regulations is likely to raise both regulatory and de facto capital levels, reduce balance-sheet leverage, and generally limit risk taking and its returns. The financial crisis also sensitized consumers and regulators to certain bank fees, considered excessive and opaque, resulting in quick action on credit 1. FDIC (various years b).

11

02-0404-1 chap2.indd 11

7/12/10 6:00 PM

12

alan m c intyre and michael zeltkevic

card fees and legislation focused on deposit fees, debit interchange, and consumer financial protection. In addition to the threat from both prudent and more politically motivated regulation, the industry faces an uncertain macroeconomic environment. The next few years could see either the spectre of inflation reemerge—and with it higher interest rates and corresponding challenges for profits—or a tepid recovery with persistently high unemployment and low interest rates, which would also challenge the banking industry (as the Japanese discovered in the 1990s). Given these developments, commercial banking could face an extended period of low returns, even after the current credit losses have worked their way through the system. Yet despite these not immaterial challenges, it is too early to declare commercial banking to be a poor, or even a mediocre, business. Banking services remain a basic need for the vast majority of the population, and only 10 percent of U.S. households lack a checking account.2 Financial services are an essential lubricant in the gears of the broader economy, offering not only transaction services but also the stores of value through which the results of all other economic activities are ultimately measured. Nor will increased government involvement in the industry necessarily erode profits. On the contrary, increased regulation may create barriers to entry that prevent profits from being competed away and may permanently eradicate the shadow banking sector that thrived during the 1990s and the first half of the 2000s. The government’s implicit promise to bail out any large failing commercial bank also amounts to an economic subsidy, as it massively reduces banks’ cost of debt capital and strengthens their competitive position vis-à-vis institutions without that government backing. As we show in our historical analysis of the profitability of the industry, the fortunes of commercial banks depend primarily on such structural factors as the rate of economic growth, the shape of the yield curve, and the regulatory regime under which they operate. How the commercial banking industry emerges from its currently fragile state, and which are the best business strategies to maximize returns, depend on the direction taken by these structural factors. Nobody can anticipate such developments with certainty, but some possible and quite plausible scenarios—and their likely effects on the industry—can be considered. The three scenarios we envision are benign, malign, and a base that lies between these two. In our three scenarios we attempt to look beyond 2010 and 2011 to a point at which current credit losses have worked their way through the system and new domestic and international regulation is in place. We also look at current 2. Federal Reserve (2007).

02-0404-1 chap2.indd 12

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

13

markets (and in particular bond markets) to tell us what macroeconomic environment we might expect two years out and how that environment will affect bank economics. In our base scenario we posit a combination of macroeconomic, regulatory, and competitive conditions that we think is plausible for the industry by 2012– 13. In this scenario, the industry’s post-tax return on tangible equity returns to its long-run average of around 10–11 percent. In this scenario, commercial banking as an industry may struggle to exceed its cost of capital, is certainly not the high-returns industry of the golden decade, and is more closely coupled with the overall performance of the economy than in other scenarios. However, there is a wide error band around this estimate. The three key drivers—macroeconomic conditions, regulatory change, and the competitive environment—need not move in sync: it is this lack of correlation that creates a plausible wide range of industry profitability. We estimate that industry return on equity could range between 1 percent in our malign scenario and 17 percent in our benign scenario. Whatever scenario unfolds for the industry as a whole, there will be a distribution of returns; and we expect that distribution to be wider than it has been in the past. While the industry may return to its long-run average profitability, it is unlikely to return to being a homogeneous sector. There will be opportunities for individual institutions, through a combination of good positioning and execution, to outperform and emerge as industry leaders. (Later we address the factors that will drive the variation in performance of individual banks and suggest some common characteristics of institutions that will not only survive but may actually thrive over the medium term.)

The Historical Performance of U.S. Commercial Banking Commercial banking revenues are derived from four activities: —Treasury: Matching the duration and convexity of assets and liabilities or generating earnings from taking imperfectly matched positions. —Lending: The origination, servicing, and investment in or distribution of credit risk. —Deposits: Gathering and investing deposit liabilities. —Fee income: The provision for fees of non-balance-sheet services, ranging from payments to investment management. The profitability of these activities depends to a large extent on a number of structural factors beyond banks’ direct control: the regulatory regime, the level of competition, and the macroeconomic environment. For the treasury, or balancesheet, elements of the business, the most important factors are the shape of the

02-0404-1 chap2.indd 13

7/12/10 6:00 PM

14

alan m c intyre and michael zeltkevic

Figure 2-1. Rate of Return, U.S. Commercial Banking Industry, 1936–2008 Post–Great Depression recovery

14

Economic turmoil: stagflation

S&L crisis

Golden decade

Financial crisis

Average ROE 11.1 10.1

12 10 8 6 4 2 0 –2 –4 –6

1936 1942 1948 1954 1960 1966 1972 1978 1984 1990 1996 2002 2008 Source: FDIC (various years b).

yield curve, the interest rate trend, and the rate of GDP growth. The history of the industry is to a large extent the history of these structural factors. In this look back, we focus on the extraordinary confluence of favorable conditions during the golden decade and on their undoing before and during the financial crisis (figure 2-1).

The First Three Eras, 1945–92 The last trauma of similar magnitude to the current crisis to affect the banking industry was the banking crisis of the early 1930s, which led to the creation of the Federal Deposit Insurance Corporation (FDIC), the Glass-Steagall Act separating commercial and investment banking, and the government-sponsored mortgage enterprises. Following this period the industry settled down to nearly forty years of relative stability. Its business model was simple: conservative underwriting and low losses (an average of ten basis points of total assets), low interest rate spreads (on average

02-0404-1 chap2.indd 14

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

15

2.1 percent of assets), and limited fee income (0.6 percent of assets).3 While this model led to low income levels, revenue was aligned with a simple low-cost business model (net interest expense was 1.6 percent of assets) to generate return on assets in the 0.6 percent range. Apart from the bump in returns that came from the drop in equity levels (from 13 to 8 percent of assets during the late 1930s and early 1940s), it was also a period of stable if not spectacular shareholder returns (8–10 percent). This stability reflects to a large extent the stable regulatory and competitive macroeconomic environment in which the industry was operating. While the 1970s were challenging economic times for much of the U.S. economy, the economic turmoil actually benefited the commercial banking industry. With spiking inflation and rising interest rates, banks had far more scope to manage interest rate margins than they had in a more stable macroeconomic environment. As a result, net interest margin expanded to an average of 3.2 percent of total assets during this period. Although fee income remained low (0.7 percent of assets) and operating costs grew (to 2.5 percent of assets), this higher margin combined with continued low losses (twenty basis points) to deliver an industry average return on equity of 12 percent. At the end of September 1981 banks faced an unusual and, at the time, challenging macroeconomic environment. Ten-year U.S. Treasury bond yields peaked at 15.8 percent. Struggling to quell the inflation of the 1970s, Paul Â�Volcker’s Federal Reserve inverted the yield curve with three-year Treasury bonds yielding 16.5 percent and a Federal Reserve funds rate peaking (in May 1981) at 20 percent, generating a negative interest rate gap for commercial banks with short liabilities and long assets. The general economic outlook was bleak, and the ensuing recession saw double-digit unemployment and a sharp and significant contraction in the economy. These were difficult times for banks and nonfinancial corporations alike. While the aggressive action of the Federal Reserve Bank may have helped bring on the recession of the early 1980s, it also set the stage for a long-term boom in commercial banking. The taming of inflation stabilized and then strengthened the value of the dollar and set in motion a long-run fall in interest rates, which bottomed out (temporarily) with a ten-year Treasury yield of 3.1 percent in June 2003. During the period 1980–92 the underlying profitability of commercial banking improved. Declining short-term interest rates created an environment in which banks could easily profit from their natural maturity mismatch of assets 3. We calculate net interest margin on the basis of total industry assets rather than on earnings assets, which are commonly used when individual bank margins are reported. All performance ratios are derived from FDIC historical profit and loss and balance-sheet data.

02-0404-1 chap2.indd 15

7/12/10 6:00 PM

16

alan m c intyre and michael zeltkevic

Figure 2-2. Pretax Net Operating Income, Commercial Banking, as Share of All U.S. Corporations, 1934–2006 Percent 16 14 12.3

12 10

5.7

8 6 4 2 2006

2002

1998

1994

1990

1986

1982

1978

1974

1970

1966

1962

1958

1954

1950

1946

1942

1938

1934

0

Source: FDIC (various years a); Thomson Reuters.

and liabilities: that is, funding long-term assets with short-term liabilities. Banks also began to change the way they priced their services, adding a wider range of fees to their traditional balance-sheet-related interest rate spread. However, industry performance was undermined by the losses of the S&L crisis. In 1987 provisions peaked at 1.3 percent of assets, a level not to be visited again until 2009–10. It was only after the savings and loan crisis and the recession of 1991– 92 had passed that the golden decade of commercial banking truly arrived.

The Golden Decade, 1993–2003 For financial firms, and especially banks, 1993–2003 was a period of unprecedented returns. Average return on equity over the decade was 14 percent, with spikes closer to 16 percent (compare this to the long-run industry average of 10–11 percent). This was also a period in which commercial banking profits decoupled from the rest of the economy. For sixty years commercial banking had accounted for around 6 percent of total corporate profits. In contrast, commercial banks now accounted for more than 12 percent of profits, spiking closer to 18 percent in the early part of the decade.

02-0404-1 chap2.indd 16

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

17

Figure 2-3. Bank Profit Drivers in the Golden Decade, 1993–2003 Net long Treasury positions

Steep yield curve

High Treasury profits

Falling rates

Strong home price appreciation

Sustained high employment

Increased second market liquidity

Rapidly increasing capacity to borrow (demand)

Rapidly increasing capacity to lend (supply)

Very high asset origination volumes and balance growth

Very low credit losses

High lending profits and operating leverage

Strong deposit growth

High barriers to entry into deposit business

Depressed competitive intensity sustains market share and pricing power of incumbent banks

High profits and operating leverage in branch-based deposit gathering

High banking profits

Source: Oliver Wyman.

This performance was the result of improved profitability (return on assets of 1.2 percent and total revenue of 5.8 percent of assets), not the result of increased leverage. The notorious high leverage of the pre-crisis period was characteristic of investment banks, not commercial banks. On the contrary, commercial banks’ equity ratios actually increased from the lows reached during the S&L crisis. The composition over the decade shifted somewhat from tangible common equity toward other types of capital, but these levels were not out of line with the longrun history of the industry. A case can be made that if some of the larger commercial banks such as Bank of America and Citigroup had been forced to bring onto their balance sheet all of their structured investment vehicles and other off-balance-sheet vehicles—which they ultimately had to stand behind—then the effective leverage of commercial banking would have been much higher. However, for the typical regional bank, returns during the golden decade were a result of increased profitability, not simply of increased leverage.

02-0404-1 chap2.indd 17

7/12/10 6:00 PM

18

alan m c intyre and michael zeltkevic

Figure 2-4. Treasury Rates, Six-Month Moving Average, 1982–2009 Percent 14 12 2-year

10

10-year

8 6 4

3-month

2

0 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Source: Thomson Reuters.

This increased profitability was caused by an extraordinary confluence of positive structural factors. Deregulation, innovation (especially the rise of asset securitization, which decoupled origination volumes from the need for balancesheet growth), and persistently high barriers to entry in the extremely profitable business of retail deposit gathering were among the most important. These industry-specific factors combined with a generally positive economic environment to drive banking profits ever higher. In addition the golden decade also saw sustained growth in the fee income associated with deposit products. Between 1982 and 2003 the yield curve changed shape many times; although the overall trend was falling interest rates, the typical curve remained upward sloping (figure 2-4). Long-term rates (ten-year) were consistently around 200 basis points higher than short-term rates (overnight or thirty-day money). Banks took advantage of this difference to generate “gap” earnings, funding long-term lending with short-term borrowing. This strategy generated significant additional margin for banks from the early 1980s until about 2003, when the Federal Reserve funds rate hit 1 percent. Interest rates fell dramatically and more or less continuously from the early 1980s until 2003, a trend with a positive effect on asset growth. Even as household debt increased by almost 300 percent, the cost of servicing that debt (as a fraction of income) barely moved because of falling interest rates (figure 2-5).

02-0404-1 chap2.indd 18

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

19

Figure 2-5. Consumer and Real Estate Debt, 1987–2005 U.S. consumer debt growth

U.S. residential real estate growth

Percent 4

4 Consumer debt

3 2

Household debt service

1 0

Mortgage outstandings

3 2

Home price index

1 10-year Treasury 1987 1990 1993 1996 1999 2002 2005

0

30-year fixed mortgage 1987 1990 1993 1996 1999 2002 2005

Source: Federal Reserve (2010); Thomson Reuters; Federal Reserve (various years); Mortgage Bankers’ Association (various years).

Consequently, consumers and businesses could increase debt levels much more quickly than their incomes grew. Fee and spread revenues associated with originating and holding lending assets boomed, with total fee income rising to 2.2 percent of assets during the golden decade. Funding options also expanded as credit “liquefied” and securitization markets grew. And after peaking in the early 1990s, credit losses declined (to nearly zero in some asset classes). This decline inflated margins for holders of risk and abetted the growth of securitization. In addition to falling interest rates, at least three major innovations in lending helped to reduce the cost and extend the reach of consumer finance during the golden decade: —The proliferation of risk-based pricing, which increased the range of consumer credit that a bank was willing to take on. —The rise of asset-backed securities and the disaggregation of parts of the lending activity chain (originating, servicing, and holding), which allowed the industry to focus on asset origination relatively unconstrained by balance-sheet capacity. —The introduction of home equity lending, which allowed consumers to tap into real estate equity to fund current expenditure. The exceptional interest rate environment, combined with these product and funding innovations, resulted in two significant trends. First, banks were motivated to push growth in real estate lending by tapping increasingly risky borrower segments. Second, lenders enjoyed a healthy demand for loans. Even as monoline

02-0404-1 chap2.indd 19

7/12/10 6:00 PM

20

alan m c intyre and michael zeltkevic

Figure 2-6. Commercial Banks, Savings Deposit Growth and Overall Macroeconomic Performance, 1970–2010 U.S. Treasury note Recession: –1.6% real growth

Real GDP growth (%) Recession: –2.9% real growth

Exception: mild recession

Recession: –7.0% real growth

Flat growth in recent quarters

Unemployment

12

8

10

6

8

Real GDP change

4

6

2

4

0

2 0

Time and savings deposits 1970

1975 3.5%

1980

1985 1.9%

1990

1995 –0.1%

2000

2005 5.8%

–2 2010 Deposit growth rate

Source: Thomson Reuters; FDIC (various years b; various years c).

credit card and real estate lenders dominated, there was still plenty of volume in the market to support the growth of commercial banks’ balance sheets. Deposits Deposit gathering was a major source of bank profits during the golden decade, benefiting from limited competition and strong deposit growth. Deposit return on equity across the industry was consistently above 30 percent, although the true value of deposit products is often obscured by poor internal transfer pricing.4 Deposit businesses benefited first and foremost from strong volume growth during this period (figure 2-6). Deposit growth also outpaced growth in the number of bank branches during much of this period. This resulted in a relatively fixed-cost base supporting ever increasing deposit volumes and revenues. Several trends contributed to this growth of balances: 4. Line-of-business profitability for deposits is reported by only a handful of banks, including Bank of America. However, confidential Oliver Wyman line-of-business profitability studies across the industry support this more general conclusion.

02-0404-1 chap2.indd 20

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

21

—Aging of the population. —Economic strength (high income growth and low unemployment). —Concentration of wealth leading to more high-balance accounts. The economics of the deposits business was also supported by limited competition, largely due to high switching costs, price opacity, and regulation. Until relatively recently, banks viewed their deposit accounts not as a profitable business but as a cost of doing business. They did not compete aggressively on rates (price), because unsophisticated transfer pricing regimes limited their ability to understand the effect of pricing change on the economics of the business. Likewise, consumers were limited in their ability to compare pricing. Hence branch proximity to the customer and local branch density remained a key driver of customer acquisition. From a regulatory standpoint, interstate banking laws made it difficult for scale players to emerge and penetrate new markets before the early 1990s. Rather than a national market for deposits and national competitors, many local, highly concentrated markets dominated the industry. In theory, the regulatory barriers should have come crashing down with the dismantling of interstate banking regulations.5 In fact, deregulation was the impetus for few shifts in local market dynamics. Instead of focusing on de novo market entry, large banks relied on mergers and acquisitions (figure 2-7). The result was an increase in local market concentration, as the footprints of newly conjoined entities overlapped. Fee Income It is ironic that after “free checking” was introduced in 1994 deposit fees rose from $16 billion a year to close to $40 billion a year in 2007, with $29 billion related to overdraft and nonsufficient fund (NSF) fees.6 Before the mid-1990s noninterest income on deposit accounts was earned through monthly maintenance fees and minimum-balance fees, which were standard on most checking accounts. However, starting with Washington Mutual, the mid- to late 1990s saw a competitive shift toward checking accounts with no minimum balances or monthly fees. These accounts quickly became part of the standard offering of most commercial banks. This move coincided with a dramatic rise in debit card transactions, as consumers shifted to using cards rather than cash for small-ticket items (figure 2-8).

5. The Riegle-Neal Interstate Banking and Branching Act of 1984 eliminated interstate banking regulations over a number of years, concluding with the allowance of interstate mergers in 1997. 6. FDIC (2008).

02-0404-1 chap2.indd 21

7/12/10 6:00 PM

22

alan m c intyre and michael zeltkevic

Figure 2-7. Commercial Banks, Herfindahl-Hirschman Index and Mergers, 1994–2006

1,100

M&A at very high levels, contributing to early rise in HHI

Bank mergers per year

M&A at lower levels added de novo activity, contributing to lower HHI

1,000

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

900

700 600 500 400 300 200 100 0

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Depost-weighted “national” HHI

Source: FDIC (various years b). a. The Herfindahl-Hirschman index (HHI) is defined as the sum of the square of market shares times 100. The index is calculated at the level of the core-based statistical area (CBSA) for all “retail” branches. Resultant HHIs are weighted by amount of deposits in each CBSA to create a deposit-weighted “national” HHI. Mergers include commercial banks and savings banks.

Figure 2-8. Debit Card Transactions, 1998–2008 Percent 60 53

Compound annual growth rate

50 40

34

30

27

37

40

42

55

58

44

30

23

20 10 0

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Source: Nilson Report (2009).

02-0404-1 chap2.indd 22

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

23

Figure 2-9. Commercial Banks, Deposit Service Charges and Non-Interest-Bearing Deposit Insurance, 1994–2006 Growtha 240 220

Service charges on deposit accountsb

200 180 160 Non-interest-bearing depositsc

140 120 100

1994

1996

1998

2000

2002

2004

2006

Source: FDIC (various years b). a. Indexed to 100. b. Compound annual growth rate, 7.3 percent. c. Compound annual growth rate, 5.7 percent.

The growth in debit transactions generated a new source of income for banks through debit interchange fees, but the lack of real-time balance information also increased the frequency with which customers incurred overdraft and insufficient funds fees. These fees, which can be as much as $40 per incident, have come to account for close to 75 percent of all deposit fees, which has allowed banks to cover the cost of servicing low-balance checking accounts without resorting to monthly maintenance fees.7 The rise of deposit fee income was also associated with a change in banks’ philosophy toward NSF incidents. Historically, the NSF fee was meant to discourage writing checks without sufficient funds in the account. But as deposit fee income grew in importance, banks began to frame penalty fees as fees for service to allow customers to avoid the embarrassment of being unable to pay at the point of sale or being late with bill payments. The result was that, during the golden decade, service charges on deposit accounts grew more quickly than noninterest-bearing deposit balances (figure 2-9). These charges boosted overall bank profitability. 7. Ibid.; Bretton Woods (2008).

02-0404-1 chap2.indd 23

7/12/10 6:00 PM

24

alan m c intyre and michael zeltkevic

As deposit fee income shifted toward insufficient-fund transactions, it also became more concentrated among a small group of customers with exceptionally high NSF transactions. By 2007 only 5 percent of transaction accounts generated 68 percent of this income, with 74 percent of checking accounts having no NSF fees at all.8 By the end of the golden decade this skew in fee income essentially created two models of transaction account profitability for the industry. For the ~15 percent of customers with more than five NSF fees a year, the fees associated with these incidents were enough to make the accounts profitable. At the other end of the spectrum, accounts with average balances of more than ~$3,000 were profitable based on net interest margin alone. However, more than 50 percent of accounts fell outside these two groups and were unprofitable, leading to material cross-subsidization between high-balance/high-NSF customers and low-balance/ low-NSF customers.9 As commercial banks came to rely more and more on rising deposit fees from a small minority of their customers, their profitability rose, but they found themselves in a situation in which their net interest income was highly concentrated and therefore vulnerable to changes in either customer behavior or the regulation of their deposits.

The Financial Crisis, 2004–09 When discussing the 2007–09 banking crisis, the popular press largely focuses on events that took place after the collapse of the two Bear Sterns hedge funds in August 2007. Attention also focuses on credit losses and the severe contraction of both retail and wholesale lending markets. However, the problems faced by commercial banks are not confined to the asset businesses, and signs of trouble arguably appeared many years earlier. Well before the current crisis came to full fruition, multiple forces were already depressing the returns of the commercial banking industry. In the discussion that follows we review significant changes in the Treasury and in the lending and deposit businesses between 2004 and 2007. Treasury After a period of sustained monetary stimulation following the bursting of the dot.com bubble, in the summer of 2004 the Federal Reserve began to tighten interest rates, leading to a rise in short-term rates and, briefly, an inverted yield curve (figure 2-10). This had two immediate effects. First, because of the flattening curve, banks were prevented from making new “gap” investments. Second, 8. FDIC (2009). 9. Oliver Wyman analysis based on FDIC data.

02-0404-1 chap2.indd 24

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

25

Figure 2-10. Commercial Banks, Three-Year Swap Rate, Three-Month LIBOR, 2004–07 Percent

2.0 1.0 0.0 –1.0 –2.0

2004

2005

2006

2007

Source: Thomson Reuters.

because of the rise in rates, banks had to pay more for short-term liabilities without being able to reprice their assets. The result was that, for only the second time in twenty years, the cumulative return of funding long-term assets with short-term liabilities turned negative. The lucrative Treasury “carry trade” was eliminated. Lending Changes in the interest rate environment also had a significant impact on lending businesses. When rates started to tick up, loan originators were no longer able to sell lower payments to borrowers based on interest rate declines. Since credit balances were no longer becoming cheaper for consumers, lenders were forced into expanding the definition of acceptable credit in order to maintain origination volume. By 2005 originators had significantly grown their appetite for lending to less creditworthy borrowers and had also expanded their suite of “innovative” mortgage products, such as adjustable-rate mortgages with their subprime rates, pay options, and negative amortizations. Lax credit underwriting was enabled and encouraged by large securitization shops on Wall Street, which had built businesses based on moving credit off originators’ balance sheets and into the capital markets. The result was that

02-0404-1 chap2.indd 25

7/12/10 6:00 PM

26

alan m c intyre and michael zeltkevic

Figure 2-11. Credit Liabilities and Mortgage Originations, Select Years, 2000s U.S. credit $ billion

Subprime and Alt-A mortgage originations $ billion 1,000

20,000

800

15,000

Household liabilities

10,000

Percenta Subprime Alt-A

600 400

5,000 Nonfinancial business liabilities

0 2001 2002 2003 2004 2005 2006 2007 2008 2009

200 0

2001 2002 2003 2004 2005 2006

35 30 25 20 15 10 5 0

Source: Federal Reserve (various years); Inside Mortgage Finance, various publications; FDIC (various years a). a. Percent of total originations.

credit quality became secondary to sustaining the fees associated with origination, servicing, and securities manufacturing. The industry continued to support growth in consumer credit during this period of rising rates, but over a third of the mortgage originations were in the subprime and alternative-A sectors, which together accounted for over $1 trillion of originations in 2005 and 2006 (figure 2-11). Deposits The lack of competition that had protected the banks’ deposit businesses started to come under threat in the early part of the 2000s. In particular, pricing transparency improved as consumers, previously content to simply bank at the closest branch, became savvy about finding the best value. The rise of the Internet and sites such as Bankrate.com enabled consumers’ price comparisons. As a result, the relationship between bank branches’ local market share and deposit share began to break down. As consumers became better able to evaluate the trade-offs among convenience, service, and price, they considered a broader array of competitors for their deposit banking. Nonbanks such as GMAC, de novo banks such as Commerce, and foreign banks such as ING Direct began to gain ground. The top thirty nonbanks and foreign banks achieved compound annual growth rates on deposits of 96 percent and 21 percent, respectively, between 2001 and 2007. These rates compare favorably to the 8 percent rates achieved by the top thirty U.S. banks

02-0404-1 chap2.indd 26

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

27

during the same period.10 In addition, since about 2007 (although more than a decade after the regulations changed), de novo (or “buy and build”) market entry strategies at long last arrived.

The Financial Crisis after 2007 The crisis of 2008–09 has been much written about. For this discussion the most important conclusion to be drawn from the events of those years is that commercial banking, despite its problems and the need for government support, has turned out to be the most robust business model in financial services, as evidenced by the fact that: —All large thrifts, with their mortgage focus, have disappeared, including Washington Mutual and IndyMac. —Asset monolines have either chosen to become commercial banks (COF) or were forced to become bank holding companies (GMAC, AXP, DFS, CIT). —Independent broker/dealers have either disappeared (BS, LEH), sold themselves (ML), or acquired bank holding company charters (GS, MS). In many ways, the crisis demonstrates not the weakness of the traditional commercial banking business model but the extreme fragility of the shadow, off-balance-sheet, system funded with short-term liabilities that thrived and grew after the 1990s. Once again, during times of severe distress only models with strong government support were viable, and the commercial banking model, with its deposit funding and FDIC backing, has proven to be more robust than the alternatives. Even though the crisis may ultimately result in the failure of up to a thousand FDIC-insured institutions, the commercial banking sector as a whole will have survived. The injection of government capital via the Troubled Asset Relief Program (TARP) program, the provision of multiple liquidity guarantees to ensure access to funding, and the confidence-building exercise of the Supervisory Capital Assessment Program’s stress test in the spring of 2008 all helped shepherd the industry through the period without the systemic collapses that characterized the banking crisis of the early 1930s. Significant losses still must work their way through the banking system—the second quarter of 2009, for instance, was a low point in bank profitability.11 However, the questions have now shifted from whether the industry will survive in its current form to what the prospects are for the recovery of profitability and shareholder returns. 10. Oliver Wyman analysis of FDIC and Federal Reserve deposit statistics. 11. SNL Financial; FDIC data analysis.

02-0404-1 chap2.indd 27

7/12/10 6:00 PM

28

alan m c intyre and michael zeltkevic

The Uncertain Future of Commercial Banking The financial crisis and its aftermath could see up to a thousand FDIC-insured banks fail. Although this would be a severe blow to the industry, it does not represent a fundamental industry restructuring, and the failure levels are likely to be below what was seen during the height of the S&L crisis in the early 1990s.12 The banks that will survive are now asking, What comes next? Will the industry again see the attractive returns of the golden decade? Or will it be condemned to utility returns and a reversion to the simpler, less risky, and less profitable business model that characterized it for forty years after the Great Depression? As mentioned, the fortunes of the commercial banking industry depend on a number of structural factors. How the industry fares over the next several years will largely depend on how these factors evolve. In this section we consider the environment in which commercial banks will likely operate once the worst of the current credit crisis has passed through the banks’ balance sheets and profit and loss statements. Our intent is not to assess the short-term prospects for the industry, which will continue to be dominated by credit losses. Rather, our focus is on the period when, although the sins of the recent past have largely been recognized, the industry is still subject to the postcrisis pressures on profitability; we expect this period to run from late 2011 to 2014–15. We do not try to predict the future. Rather, we consider three plausible scenarios: base, benign, and malign. These scenarios are defined by three environmental drivers: general macroeconomics, government regulation, and market competition. The scenarios represent a spectrum of likely outcomes, ranging from what a reasonable optimist might hope for to what a reasonable pessimist might fear. There is also no a priori high correlation among the three environmental drivers. It is plausible for example to posit a future (as we do in the malign scenario) in which heavy-handed government regulation combines with challenging macroeconomics to create strong downward pressure on industry profitability. For each scenario we estimate the effect of the environmental factors on a number of input variables and model their impact on industry performance. These input and output variables are: —Net interest margin (NIM), which affects the pretax return on assets. —Non-interest income (NII), which affects the operating margin. —Non-interest expense (NIE), which affects the efficiency ratio. —Provisions, which affect the pretax return on tangible equity (ROTE). —Tangible common equity (TCE), which affects post-tax ROTE. 12. The S&L crisis saw the narrow failure of 747 thrifts, but between 1986 and 1993 the FDIC reported 1,453 failures of insured institutions. FDIC (2009).

02-0404-1 chap2.indd 28

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

29

The inputs and outputs are industry averages. There is, of course, variation around these averages, with some banks doing better than others. This variation is not entirely idiosyncratic. The way the environment evolves favors or harms some banks more than others. We consider such disparate impacts below, where we consider the likely characteristics of the industry’s high performers.

The Changing Environment for U.S. Commercial Banking The environment for commercial banking can be changed, as mentioned, by any of three factors: general macroeconomics, government regulation, and market competition. We analyze these factors below. General Macroeconomics The macroeconomic environment is a key determinant of banking profitability. This environment is shaped by three principal drivers: —The height and shape of the yield curve and its impact on Treasury profits. —Asset performance, which is dependent on economic activity, income, and€unemployment. —Asset and liability growth rates, which are dependent on the aggregate growth of the economy and level of savings. The most uncertain macroeconomic environment is that in the immediate future (2010–15). While the economy appears to have officially exited the recession sometime in the second half of 2009, the character, strength, and timing of the economic recovery remains a subject of much debate. At the core of this debate is which of two opposing opinions will prove correct. On one side are those who think the United States is heading for a Japanese-style period of low growth, low inflation, and low interest rates, driven by continued personal and corporate deleveraging. On the other side are those who think that rapidly increasing and ultimately unsustainable federal deficits and the expansion of the monetary base to prop up the financial system will lead to faster growth but also higher inflation, higher interest rates, and possibly an acute dollar crisis, as foreign creditors lose faith in the creditworthiness of the U.S. government. In our scenarios we try to capture a realistic range of outcomes for these macroeconomic drivers, anchoring our speculations in how the current bond market is pricing in expectations on inflation and interest rates. For example, the spread between nominal Treasury yields and index-linked TIPS (Treasury inflationprotected securities), which protect against inflation, indicates muted inflation in 2010 and beyond. The case can be made that traditional inflation-warning indicators are distorted because, with the Federal Reserve lending money at a rate close to zero, banks have developed a lucrative carry trade by simply buying

02-0404-1 chap2.indd 29

7/12/10 6:00 PM

30

alan m c intyre and michael zeltkevic

Treasuries, thus keeping their yield artificially low. This may be true, but rather than second-guessing the bond market, we choose to simply interpret what the market is telling us as the basis for our macroeconomic factors. Government Regulation The nature of the future regulatory landscape is becoming clearer as the legislation passed in both houses of Congress is reconciled and as the Federal Reserve and other regulators move from preliminary findings to final rulings on a range of topics. For products such as credit cards, legislation has already passed, and its impact is now being felt in the industry. However, most of the regulations with the biggest impact on the commercial banking industry are still in the sausagemaking phase of the legislative process. From the macroeconomic architecture of the regulatory agencies to the microeconomic specifics of calculating and applying overdraft fees, the proposals are many and varied. Some of these proposals, such as the rollback of national regulatory preemption for consumer lending products, could have significant impacts on the industry, since the issues could be determined state by state. Whatever specifics emerge from the legislative process, a fair assumption seems to be that the impact on the economics of commercial banking will be negative, and this is what we have built into the scenarios. Among the likely detrimental impacts on profitability are higher capital levels, lower fee income as a result of both overdraft regulation and debit interchange restrictions, and a greater cost of compliance. However regulation is also likely to create a level playing field that will be beneficial to commercial banks by eradicating the shadow banking sector that thrived during the golden decade. With higher capital levels, regulatory minimums are likely to increase and the quality of capital will also likely improve. Changing capital levels will be a bigger issue for the largest institutions, as they will not only be subject to systemic risk oversight (the concept of tier 1 financial holding companies) but will also be hit by narrower changes, such as having to hold more trading book capital. Regulatory minimums will likely increase, and de facto, market-imposed minimum capital levels are also likely to increase as investors take a harder line with institutions regarding an equity buffer above and beyond regulatory minimums. A variety of measures seem likely to restrict fee income. In the cases of overdrafts and deposit fee income, this is a $25 billion to $40 billion revenue stream, depending on what is included. In 2006, at the peak of bank profitability, deposit fees accounted for 17 percent of pretax net income.13 Given current changes being discussed, $10 billion to $15 billion of this revenue stream could be at risk. 13. FDIC (2008).

02-0404-1 chap2.indd 30

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

31

Some fee changes will be adopted voluntarily, as exemplified by those banks that have already moved to “opt in” overdrafts. Other changes, such as the Federal Reserve’s regulation E and debit interchange fees, are being imposed on banks. Some of these changes, such as regulation E, will undoubtedly lower fee income for banks. However, other contributors to lost fee income are likely to be at least partially replaced by the introduction of other fees (such as a return to monthly account fees for accounts with low monthly balances). The behavior of customers in light of these changes is also uncertain. Current deposit service fees are highly concentrated among 10–20 percent of banks’ customers.14 These customers certainly pay high fees, but they also benefit from the service. For example, having the confidence that you will be able to pay by debit card at a Wal-Mart checkout without having your card rejected has value to many customers, and with debit card payments rising at more than 10 percent a year, the use of this payment protection is likely to increase further.15 While the intent of the current legislation may be to restrict banks’ fee income, what is unclear (even in an opt-in environment) is whether customers will be willing to trade the certainty of lower fees for the certainty of payment that the current system offers. Current anecdotal evidence suggests many customers are choosing payment certainty with opt-in overdraft rates at 70.80 percent at some institutions. Complying with all the additional regulation, and lobbying authorities for favorable decisions, will not be cost-free. For example, the banking industry has already spent more than $220 million on legislative lobbying in 2009, and current estimates are of 2,000 to 3,000 lobbyists trying to influence the final stages of the financial reform process.16 The silver lining is that the growing regulatory burden should benefit incumbents by creating barriers to entry for new competitors; however, the overall impact on the industry will undoubtedly be negative. Given that the compliance burden is a largely fixed cost, its growth should also encourage further industry consolidation. However, the impact of increased regulation on the industry will not be homogeneous. The creation of tier 1 financial holding companies may create diseconomies of scale for the largest institutions, as they are forced to comply with a bespoke range of supervisory and regulatory requests, while smaller institutions may benefit from their exemption from the Foreign Corrupt Practices Act. 14. Ibid. 15. Nilson Report (2009). 16. Stephen Labaton, “Lobbyists Mass to Try to Shape Financial Reform,” New York Times, October 14, 2009.

02-0404-1 chap2.indd 31

7/12/10 6:00 PM

32

alan m c intyre and michael zeltkevic

Where the break points in the costs of compliance will fall is unclear, as the line is being drawn at different levels for different regulations. For example, the cutoffs for the Supervisory Capital Assessment Program’s stress test and the more recent Federal Reserve compensation review have been different. The waters have been further muddied by the proposed Federal Deposit Insurance Corporation (FDIC) resolution fee for systemically important institutions. This fee would be applied to all institutions with greater then $10 billion in assets, which at the moment means around 120 banks (the same level at which the proposed small bank exemption from the Foreign Corrupt Practices Act would kick in). As these various regulatory threads resolve themselves into a new regulatory landscape, the differential cost impact on the banking industry will become clearer. If the regulations pass as currently drafted, we may be entering a period in which, from a compliance cost standpoint, small to midsized regional banks with the scale to spread fixed compliance costs but still small enough to fly under the federal regulatory radar may be advantaged. Before the crisis, commercial banks had to compete with financial players that, in some cases, acted like banks but were regulated differently and were not always subject to the same restrictions as commercial banks (broker-dealers, consumer finance companies, finance arms of industrials, and so on). These shadow banks were able to tap wholesale funding at attractive rates but also to lever their capital far more than traditional commercial banks. The crisis has leveled the playing field by either eliminating many of these competitors or bringing them under the same regulatory umbrella as commercial banks. The market also leveled the playing field by simply denying them funding during the liquidity crisis of late 2008 and early 2009. For those that survived, some may have advantages in terms of operating model and differential expertise in their chosen markets, but what they won’t have will be an unfair advantage in capital levels, and in many cases there may still be a severe disadvantage in terms of funding costs (as demonstrated by the difficulty CIT had in trying to restructure its debt before entering chapter 11 bankruptcy, despite having converted to a bank holding company). The draft Basel III liquidity and funding rules currently under discussion will also place a premium on deposit funding and may severely restrict the asset growth of wholesale funded institutions, even if that wholesale funding is available at attractive rates. Competition Both macroeconomic conditions and the evolution of the regulatory environment are likely to be negatives for bank profitability through 2015. In contrast, we expect a balance of positives and negatives to arise from the following three competitive developments.

02-0404-1 chap2.indd 32

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

33

—More rational competition. Given the increased concentration of the industry, the demise of some of the more aggressive competitors, and the withdrawal of certain nonbank players, we expect a less aggressive competitive environment with pricing that will support better margins. While credit has contracted overall, spreads on the loans that are being made have widened considerably in both retail and wholesale asset classes, and the power relationship between borrowers and lenders has shifted dramatically back in favor of lenders. —Potentially lower deposit profitability. The profitability of the deposit business is largely dependent on macroeconomic factors and alternative funding mechanisms. The intense competition for deposits in the fourth quarter of 2008, at the peak of the liquidity crunch, abated as the Federal Reserve added liquidity into the system. At the same time, if interest rates stay low for an extended period, deposit margins, even on noninterest-bearing accounts, will remain low. However, the liquidity programs put in place in late 2008 are already beginning to be withdrawn, and more focus is being put on traditional wholesale funding mechanisms. As the true cost of that borrowing becomes apparent (as less robust institutions access the funding markets), the true value of low-cost deposits may again increase. The immediate future is also likely to see far more relationship banking, which could in turn result in lower deposit margins as banks are forced to give up marginal revenue to retain valued customers. —Reduced fee income. We could see a potentially value-destroying race to the bottom to respond to legislative and consumer pressure around deposit fees and opt-in overdrafts. However, the beginnings of that industry realignment on deposit fee income suggest that we are more likely to see a return to an era when overall deposit product pricing was more explicit and costs of the transaction account infrastructure were spread across customers rather than concentrated on the 10–20 percent of customers who account for 80 percent of current deposit fee income. Any behavioral effect of such clarity is uncertain; it will partly be determined by banks’ willingness to market and price overdraft protection products.

Scenarios Analysis The environmental factors discussed above could interact in a myriad of combinations to produce a full spectrum of profit outcomes for the U.S. commercial banking industry. Here we consider three plausible scenarios that represent particular combinations of macroeconomic, regulatory, and competitive futures: —A base scenario (table 2-1), which lies between the extremes (of benign and malign) described below. This scenario represents our best estimate of where the industry may be heading in the years leading up to 2015. This outcome, though unspectacular, is broadly positive for the industry as a whole and is predicated

02-0404-1 chap2.indd 33

7/12/10 6:00 PM

34

alan m c intyre and michael zeltkevic

Table 2-1.╇ Baseline Scenario Regulation

Scenario

Capital

Minimum capital levels are increased somewhat in Â�aggregate, though more so for systemically important Â�institutions. Consumer protection—lending Some restrictions on sales of certain complex and high margin products (largely drawn from the various CFPA proposals). Consumer protection—deposits Limited opt-in overdraft legislation is passed with the remainder of deposit fee income being subject to optout provisions. Macroeconomics Interest rates Unemployment Inflation GDP growth Savings rate Competition and markets Wholesale lending market Secondary market

Deposit competition Consolidation

Second-order effects Credit losses Asset growth Deposit growth Bank failures

02-0404-1 chap2.indd 34

Relatively steep curve with 10-year rates around 4%. Elevated (8–9%). Subdued (1–2%). U-shaped recovery with 2–3% growth thereafter. 7–8%.

Volumes return to levels seen early in the decade though risk-adjusted margins remain high. Private mortgage securitizations return as a funding mechanism but only for the jumbo segments; card-securitization market returns to normal. Steady increase in competition from nonbranch banks, reflecting challenging wholesale funding conditions. Consolidation from a combination of large banks buying small community banks and FDIC-brokered deals for failed institutions (but limited merger of equals among larger banks).

Defaults remain elevated though recoveries improve as collateral prices stabilize. Total outstanding credit remains flat. Deposits grow modestly faster than GDP. Bank failures peak in 2010 but continue at an elevated level in 2011 and 2012. More than 500 failures.

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

Regulation Input metric (% of assets) NIM NII

NIE

Provisions TCE

35

Scenario Reversion to 2008-like interest rate environment; steep yield curve (reference: 2008, 2.9%). Restricted secondary asset market plus modest consumer protection activity returns fee levels to those seen in the mid-1990s (reference: 1995, 1.9%). Costs do not rise above pace of inflation, but asset volumes stagnate, returning NIEs to precrisis levels (reference: 2005, 3.1%). Provisions similar to the 2007 experience (reference: 2007, 0.6%). Minimum capital levels are elevated, though not substantially (reference: 1994–2003, 7.1%).

Outcome metric

Baseline

Pretax ROA Operating margin Efficiency ratio Pretax ROTE After-tax ROTE

â•⁄ 1.1% 23.0% 65.0% 15.0% 10.0%

on a steady but somewhat jobless recovery and government regulation that satisfies some of the political imperatives without having too negative an impact on banking profitability. —A benign scenario, in which the impact of regulation is muted and positive macroeconomic and competitive forces return the industry to performance levels similar to the those of the golden decade, although with different dynamics (table 2-2). This scenario is driven by the successful reflation of the economy by the Federal Reserve without the precipitation of either an inflationary spiral or a dollar crisis. —A malign scenario, in which the impact of regulation is more pronounced and the macroeconomic situation returns to the point at which the industry would only break even (table 2-3). This is our deflationary, or Japanese-style, scenario. We set aside potential interventions by the government to ensure higher levels of bank profitability, but we do believe that this scenario, if it were to come to pass, would stimulate widespread and government-supported industry

02-0404-1 chap2.indd 35

7/12/10 6:00 PM

36

alan m c intyre and michael zeltkevic

Table 2-2.╇ Benign Scenario Regulation

Scenario

Capital Capital requirements regime stays largely as is. Consumer protection—lending Limited additional regulation beyond what has already been passed (primarily enhanced mortgage disclosures, card rate and fee limitations, and restrictions on the most problematic mortgage products). Consumer protection—deposits Regulations remain unchanged, with voluntary industry action on overdraft fees sufficient and the associated reduction in fees being quickly replaced by broader account fees. Macroeconomics Interest rates Employment Inflation GDP growth Savings rate Competition and market Wholesale lending market Secondary market

Deposit competition Consolidation

Second-order effects Credit losses Asset growth Deposit growth Bank failures

02-0404-1 chap2.indd 36

Steep curve, 10-year rate around 5%. 6–8% unemployment. Positive at 2–3%. U-shaped recovery with ~3% growth thereafter. 5–7%.

Availability of credit gets close to mid-decade levels, but risk-adjusted pricing remains attractive. Some near-prime private MBS returns along with the jumbo market, but subprime remains largely dormant; ABS (including auto and student lending) return to �normal. Deposit share of nonbranch banks quickly stabilizes given availability of attractive wholesale funding. Some merger of equals among regional banks in addition to roll-up and FDIC-brokered deals for failed �institutions.

Losses return to levels of a normal business cycle. Credit expands in step with GDP growth. Deposits grow modestly faster than GDP. Bank failures peak by mid-2010 and then drop quickly, with 300–500 failures.

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

Regulation Input metric (% of assets) NIM

NII NIE

Provisions TCE

37

Scenario Late-1990s interest rate environment; good competitive environment in the lending space offset by a limited ability to take on risk (references: 1998, 3.5%; 2008, 2.7%). Stable fee environment along with improved secondary market allows for return to precrisis levels (reference: 2006, 2.2%). Modest growth in assets coupled with a slimmed down cost base keep NIE in check; costs rise modestly from crisis levels (reference: 2008, 3.0%). Provisions return to 2007 levels (reference: 2007, 0.5%). Capital holes are filled, but levels are not boosted (reference: 2006, 6.8%).

Outcome metric

Benign

ROA Operating margin Efficiency ratio Pretax ROTE After-tax ROTE

â•⁄ 1.8% 34.0% 57.0% 26.0% 17.0%

restructuring to prevent the exit of capital and the negative knock-on effects on overall economic growth.

The Distribution of Returns: Who Will the Winners Be? Our base scenario has the commercial banking industry realizing an average return on equity in the 10–11 percent range once current credit issues have worked their way through profit and loss statements and balance sheets. This scenario has a wide confidence interval around it, given the uncertainty of key macroeconomic and regulatory drivers of profitability. However, even if our base scenario turns out to be a predictor of the industry’s economic performance, we expect wide variation in the performance of individual institutions. This discussion identifies the factors that will drive that variation. Despite often being seen as a homogeneous industry, with its economics primarily driven by macroeconomic and regulatory factors, commercial banking has always shown a variation in performance around the mean. Further, during the

02-0404-1 chap2.indd 37

7/12/10 6:00 PM

38

alan m c intyre and michael zeltkevic

Table 2-3.â•⁄ Malign Scenario Regulation Capital levels Consumer protection—lending

Consumer protection—deposits

Macroeconomics Interest rates Employment Inflation GDP growth Savings rate Competition and markets Wholesale lending market Secondary market Deposit competition Consolidation

Second-order effects Credit losses

Asset growth Deposit growth Bank failures

02-0404-1 chap2.indd 38

Scenario Minimum capital levels are increased substantially for all banking institutions. Full (or nearly full) implementation of the CFPA as originally proposed with significant oversight responsibility delegated to the new agency; national preemption in lending regulation is removed, which increases compliance costs. Broad opt-in overdraft legislation or a bill with wideranging and explicit overdraft fee restrictions is passed.

Flat curve with the 10-year rate at 3% or lower. Unemployment above 9%. < 0–1%. Double-dip recession in 2010 with ~1% growth Â�thereafter. 8–10%.

Market remains tight with little improvement from today’s levels of credit availability. GSEs are only game in town for mortgages; card securitizations continue but with higher spreads. Limited deposit competition given continued expansionary monetary policy from Federal Reserve. Consolidation only results from bank failures, which peak in 2011 but continue through 2012.

Losses remain high for several years as prime consumer credit and commercial real estate defaults peak later than expected. Credit continues to contract as new volumes do not exceed maturing loans and defaults. Deposits grow in line with GDP. Bank failures do not peak until 2011; 600–1,000 bank failures.

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

Regulation Input metric (% of assets) NIM

NII

NIE

Provisions

TCE Outcome metric Pretax ROA Operating margin Efficiency ratio Pretax ROTE After-tax ROTE

39

Scenario Late 1970s interest rate environment; good competitive environment in the lending space offset by a severely limited ability to take on risk (references: 1978, 2.8%; 2008, 2.7%). Restricted secondary market for assets plus substantial consumer protection regulation reduces fee income significantly (reference: 2008, 1.6%). High compliance costs and low asset growth makes NIE difficult to control, though cost cutting serves as a counterbalance (reference: 1992, 3.8%). Losses fall significantly from peak levels but are still on par with experience during 2001–02 recession (references: 2008, 1.3%; 2001–02, 0.7%). Minimum capital levels are elevated substantially (references: 2002, 7.4%; 1993, 7.5%). Negative â•⁄ 0.1% â•⁄ 2.4% 85.0% â•⁄ 1.3% â•⁄ 0.9%

crisis not only has that mean return fallen, but the variation has increased (figure 2-12). So what will characterize the institutions that outperform and generate returns in the 15–20 percent range (rather than clustering around or below the industry average)? Two macro factors are critical, institutional positioning and management execution.

Institutional Positioning Institutional positioning is a combination of the lines of business in which a bank chooses to compete (and within them the customer or product segments it chooses to target) and the geographic markets in which it does so. From a geographic perspective, not all commercial banking markets are created equal. Demographic and competitive factors interact to create wide variations in fundamental market attractiveness; metropolitan statistical areas can be ranked

02-0404-1 chap2.indd 39

7/12/10 6:00 PM

40

alan m c intyre and michael zeltkevic

Figure 2-12. Distribution, Return on Average Equity, Publicly Traded Commercial and Thrift Banks, 1997–98 and 2007–08 Percent

1997–98

25 20 15 10

2007–08

5 0

< –10

–5

0

5

10

15

20

25

Source: FDIC (2009).

according to their fundamental attractiveness as a deposit market based on a combination of balances, local pricing, and costs (figure 2-13). The second positioning factor derives from business mix and portfolio composition. The sources of such balance-sheet income are many and varied: on average, noninterest income accounts for 41 percent of total revenue for these institutions, but variation is apparent not only in the level of that income but also in its composition and source (figure 2-14). These business mixes will be more or less attractive depending on the combination of macroeconomic, regulatory, and competitive forces. The average economics of the commercial banking industry is hard to predict, given uncertainty about macroeconomic, competitive, and regulatory drivers. When the industry is segmented along geographic and line-of-business dimensions, that difficulty increases. What is clearly true is that certain geographies will continue to be differentially affected by macroeconomic drivers, and that these drivers will affect the quality and the volume of both assets and deposits. Macroeconomic and regulatory pressures will also have differential impacts on individual institutions according to their business mix. In 2009 banks with vanilla capital market trading businesses benefited from wide spreads and more attractive competitive dynamics. Institutions like Wells Fargo and Bank of America also benefited from the growth in fee income associated with mortgage originations in a far less competitive market. In the second quarter of 2009 these two

02-0404-1 chap2.indd 40

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

41

Figure 2-13. Metropolitan Statistical Areas (MSAs), Market Attractiveness Index MSA quality score Honolulu

300 250 200

Las Vegas

150

Nashville

Kansas City

100

Cleveland

50 0 Source: Nilson Report (2009).

Figure 2-14. Noninterest Income Revenue, Contribution and Mix, Ten Regional Commercial Banks, 2007–08 Noninterest income % of total revenue Asset management middle-market services Payments corporate trust 52

50

51 47

40 62

30

57

90

National mortgage, private Asset banking, Payments mgmt retail brokerage 45

44

Key sources of “other” fee income

Insurance 41

42

34 57

57

55

66

2

47

31

45 50

20 10 0

39

43

14 3

20

25

4 14

10

55

Regional PNC US Bank TCF bank average Deposit service charges

14 6

26

23

18

Fifth Third

Key

Mortgage banking

21

18 8

5 7

4

19

22

28

52 23 48 26

Wells BB&T SunTrust RBS- Comerica Fargo Citizens Trust and Investments

Other

Source: Thomson Reuters; SEC filings; Oliver Wyman analysis.

02-0404-1 chap2.indd 41

7/12/10 6:00 PM

42

alan m c intyre and michael zeltkevic

institutions accounted for 44 percent of all mortgage originations, up from 29 percent a year earlier.

Management Execution In this complex environment, with many countervailing forces affecting the industry, there is no single way to succeed and outperform. However, high performance is likely to be underpinned not only by good institutional positioning but also by outstanding management execution. Such management execution is likely to be obvious in several areas: risk management, targeted asset growth, deposit focus and competence, cost control, advantaged mergers and acquisitions, human capital, and innovation. Risk Management The current crisis was precipitated by a misunderstanding and mismanagement of various types of risk, primarily the credit risk of subprime mortgages and the liquidity risk associated with wholesale funding models. However, in the golden decade banks faced what appeared to be a low-risk environment. High deposit fees, consistent Treasury profits, high growth rates of assets and deposits, and the belief that many loan classes were virtually risk free meant that most commercial banks saw risk as muted and risk management as somewhat of a commodity that relied heavily on vended external models rather than internal capabilities. In the future, with a tougher interest environment, lower fee income, low asset growth, and the other economic challenges, returns over the cost of capital are going to require both prudent risk taking and risk management and, most important, the ability to convince regulators of that prudence. It is important not to mistake current low-cost government funding and the positive interest gap as anything but temporary. The initial pre-crisis compression of the net interest margin, immediately following the end of the golden decade in 2004, arose from exactly this type of interest rate situation, in which commercial banks with predominantly short-term, rate-sensitive funding found their profitability squeezed by rising rates. While the current rate environment offers shortterm profit opportunities, these opportunities will disappear once rates begin to rise, and banks will position their balance sheets accordingly. In general, high performers will exhibit agile and forward-looking balance-sheet management, which will allow them to protect and enhance their net interest margin in what is likely to be a sustained low-rate (but rising-rate) environment. While effective balance-sheet and Treasury management will be a common characteristic among high performers, the ability to grow earnings will also be dependent on the ability to grow both sides of the balance sheet.

02-0404-1 chap2.indd 42

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

43

Targeted Asset Growth Asset growth has not been on the agenda for most commercial banks since mid-2008. Total banking assets grew by 4 percent between September 2008 and October 2009, but total loans and leases fell by 5 percent, and commercial and industrial loans fell by 13 percent.17 Instead of lending, banks have been paying off debt, rebuilding securities portfolios, and holding more cash on their balance sheets (more than 10 percent of total assets in October 2009). Balances of the industry as a whole continued to grow in 2009, even though many individual institutions saw their balances shrink, as they wrote off or attempted to sell existing assets while hoarding available capital in anticipation of further losses. The banks’ incentive to reduce lending was matched by voluntary customer deleveraging, as both corporations and individuals sought to repair their own balance sheets. Total consumer debt outstanding fell steadily after February 2009 and continues to fall by an annualized rate of around 6 percent.18 The quarterly Federal Reserve loan officer survey also indicates that demand for corporate and individual loans has fallen every quarter since 2006 (although utilization of existing lines did rise in late 2008 and early 2009). These institution-specific, industrywide, and customer-driven trends bolstered bank capital ratios and signaled a healthy deleveraging after a period of excessive asset growth, but in the long run shrinking loan books will make it hard to be an industry high performer. In a persistently low-rate environment like the one in our base scenario, the ability to originate and effectively price assets will be a key contributor to high performance. However, until the economic recovery picks up steam, loan demand is likely to remain anemic. For those institutions with balance sheets capable of absorbing asset growth, the key question has been and will continue to be where to lend to grow their balance sheets (or just counteract recent shrinkage). With many of the specialist originators and finance companies like CIT either gone or severely constrained in their ability to lend, commercial banks (and in particular small and midsized regional banks) should have access to a far broader range of asset growth opportunities than in the golden decade, and those lending opportunities should have better risk and return characteristics. However, the competition among banks for those assets will also increase as they fight for shares of what may be a static or even declining total credit base. When commercial banks start ramping up lending again, where might the best opportunities lie? A priority is likely to be the reintegration of the overall 17. Federal Reserve (2009a). 18. Ibid.

02-0404-1 chap2.indd 43

7/12/10 6:00 PM

44

alan m c intyre and michael zeltkevic

bank balance sheet at the individual customer level. By prioritizing lending to the most profitable customer segments and relationships, particularly in the small business and middle-market segments, credit will once again be used to solidify deposit and transaction relationships. This in turn will strengthen deposit economics in a rising-rate environment. We think that high performers will therefore focus on small businesses and middle-market corporations and that they will operate on both sides of the balance sheet, with the asset side probably offering the most attractive returns, given the sustained low interest rate environment and the pricing power that has returned to the banks. Obviously as capital flows back toward this market, the current wide margins will contract somewhat, but we believe they will remain attractive through 2015. This environment may also favor commercial banks, which maintained or can quickly build consumer asset origination capabilities rather than ceding them to specialist competitors. High performers will be capable of originating, underwriting, and pricing consumer assets from their current customers to benefit from attractive product and relationship economics. Indeed, many regional banks are interested in reentering the credit card industry, which they exited during the 1990s. Effectively executing such a relationship lending strategy in both commercial and consumer arenas will require a robust, comprehensive, and forward-looking view of customer value, a view still lacking in many commercial banks. Some of the support for this strategy will come from external information providers and credit bureaus, which are already retooling to provide a far more rounded view of the financial situation of consumers and small businesses than they did before the crisis. However, the primary differentiation will be the ability of the banks themselves to make better use of their own internal proprietary information on their customers to rethink their approach to credit underwriting. At least for the foreseeable future, the balance of power has shifted back toward banks in asset products. Banks that use that power wisely, to create and protect broader customer relationships (rather than simply rationing credit for the highest return), will be more likely to outperform the industry. Deposit Focus and Competence It is easy to conclude that, in a low-interest-rate environment, deposits are less attractive than in a high-interest-rate environment and that the focus will shift to the asset side of the balance sheet. Many commercial banks, in only a few months, moved from gathering deposits at any cost to realizing that they cost too much. Behind this conclusion were four factors: —Government-insured debt programs made banks less concerned about liquidity and funding.

02-0404-1 chap2.indd 44

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

45

—The reemergence of alternative wholesale funding and continued low interest rates squeezed the margin on deposits. —Increased FDIC assessments raised the all-in cost of deposits by twenty to thirty basis points.19 —The lower fee income associated with transaction accounts made these accounts less attractive than other funding sources. During most of 2008 and 2009 deposit funding was the key factor of production in commercial banking. In the golden decade, deposits were a largely unacknowledged source of earnings, but as other sources of funding dried up, deposits became a prerequisite for financial stability. However, as pressure from short-term liquidity and funding eased, the underlying economic challenges in the deposits business have become more apparent. Despite these challenges, we still think that maximizing high-quality deposit growth will be a strategic priority for most commercial banks. In a rising-rate environment, low-cost and rate-insensitive deposits have a material impact on the net interest margin. From a Treasury standpoint, as interest rates rise assets will reprice upward, widening the spread against a low-cost deposit base—or at least mitigating the impact of rising liability costs from other sources. This is in stark contrast to institutions primarily funded by market liabilities or interest-rate-sensitive time deposits, whereby rising rates tend to compress margins regardless of the shape of the yield curve. While zero-interest transaction accounts dominated the deposit liability mix of commercial banks in the 1960s, these deposits now account for less than 10 percent of total deposits (figure 2-15). Hence competition for them is likely to be intense. Competition will also be increased by low growth in deposits. While the savings rate is likely to remain at 7–8 percent, this is driven by debt redemption and will not result in material deposit growth, particularly if equity markets continue to rise. Instead, deposit growth will continue to be low or even negative in real terms, given the evidence from prior recessions. However, our base scenario does not feature sharply rising rates but rather a protracted steep curve, with long-term rates not rising above 4 percent. This interest rate environment will favor transaction accounts as a source of profitability, but the impact will be muted compared to ten-year rates going to 6 percent. In this environment deposits may be less attractive than alternative wholesale funding, at least from the perspective of the net interest margin, as evidenced by the Goldman Sachs decision not to enter the deposits business despite having the opportunity to do so under its bank holding company charter. However, the 19. Morgan Stanley (2009).

02-0404-1 chap2.indd 45

7/12/10 6:00 PM

46

alan m c intyre and michael zeltkevic

Figure 2-15. Commercial Banks, Deposit Liabilities, 1991–2007 Percent 90 Time

80 70 60 50 40

Savings

30 20 Demand

0

1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

10

Source: Federal Reserve (2009a).

government guarantee of bank debt is not permanent, and the cost of issuance is already rising. This fact will level the playing field somewhat with respect to the attractiveness of deposit funding. In this environment the core role of transaction deposit accounts will be to broaden customer relationships and provide a customer base for fee income and asset cross-selling; the transaction account will also provide differentiated behavioral and competitive information to better underwrite asset products. Asset businesses may continue to be more attractive in the short term, but without a wellrun and effective deposit franchise, many institutions will not be able to access those assets in an advantaged way. The differentiated management of the deposits business, either for net interest margin or for customer acquisition, will require four initiatives at most banks: —Rethink funds transfer pricing to better incorporate the true value of liquidity and use that insight to drive pricing and customer management decisions. In a sustained low-rate environment, this will allow banks to recognize the relative attractiveness of deposit funding over wholesale funds and to adjust their product and distribution strategies accordingly. If the proposed Basel III funding rules are enacted, there will also need to be a far more explicit link between customer deposits and the assets that they are able to fund.

02-0404-1 chap2.indd 46

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

47

—Increase analytical spending on deposits to better understand and retain current customers, especially those customers whose product preferences and behavioral characteristics provide the best chance of generating attractive account-level economics. —Reassess the viability of each physical branch location in light of the new deposit economics. —Understand how product design, value packaging (the incorporation of nonrate features into deposit products), and segment focus can help capture and maintain share in target markets, particularly in light of the fee-income pressure in current legislation. While product design and packaging are important tactics, segment focus will likely be what defines those institutions with the best deposit economics (25 percent of U.S. households account for 75 percent of deposit profitability).20 The winning commercial bank deposit models throughout the 2000s that were not rate based (such as Signature, Commerce, and Pinnacle) focused on creating business models that are attractive to this 25 percent of customers by emphasizing a customer experience that is markedly superior to typical mass market offerings. These banks were rewarded with four to ten times more deposits per branch than deposits in large regional banks, which have lost market share as a consequence.21 The commercial banks that win the deposit battle will learn from the success of these targeted business models and will find ways to replicate them. These target customer bases for deposits will also provide some protection against changes in deposit fee structures, as higher fees tend to be associated with lower average balance accounts and a socioeconomic demographic that is more mass market. Cost Control The impact of cost efficiency on performance is not a straightforward issue of economies of scale and of being the low-cost producer (figure 2-16). In fact there is an argument to be made that diseconomies of scale will be associated with being a first-tier financial institution under the proposed regulatory reforms. Both the cost of compliance and the regulatory capital requirements may well be higher than for second-tier regional banks. But even in a more benign regulatory environment, such as prevailed in 2006, only mild scale economies are apparent. This outcome can be traced to the heterogeneity of the industry. A regional bank with a trust and investment management business may have a significantly 20. Federal Reserve (2009b); Oliver Wyman analysis. 21. FDIC (various years b).

02-0404-1 chap2.indd 47

7/12/10 6:00 PM

48

alan m c intyre and michael zeltkevic

Figure 2-16. Commercial Banks, Efficiency Ratio, by Return on Average Equity a Percent

Q2—2006

50 30 10 –10 –30 –50 –10

10

30

50

70

90

110

130

150

Efficiency ratio Source: SNL Financial; Oliver Wyman. a. R 2 = 29 percent.

higher cost-to-income ratio than a more balance-sheet-oriented bank, but it delivers the same investor returns because of its lower capital requirements. Moreover, unlike many other international banking markets, the United States has outsourced scale economies to specialist vendors. With a range of off-theshelf packages for entire back-office operations, the advantage of being a scale originator of transactions is minimized. Rather than being a true determinant of high performance, good cost control is likely to become more of a hygiene factor to achieve even average industry returns. The only scenario in which this may not be true is in the malign scenario, in which break-even economics for the industry as a whole may make good cost control a true differentiator of high performance and a compelling reason for being an acquirer rather than an acquired. One exception even under the base scenario is likely to be the ability of institutions to increase the productivity of the largest cost item in most commercial banks, namely, the branch network. In the golden decade, many branch networks could afford to resemble planned economies, with micromanagement of resource allocation from the center, small and opaque rewards for superior front-line performance, and a consequent lack of ownership of the results at the local level.

02-0404-1 chap2.indd 48

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

49

Commercial banks that perform best over the next cycle are likely to rethink how they measure the value of each local market, how they set targets, and how they reward high performance to better unlock local entrepreneurship and kick-start productivity improvements. In short, they will begin to think more like retailers. Advantaged Mergers and Acquisitions As the banking crisis unwinds cumulative bank failures are likely to be in the range of 500–1,000 institutions in the period 2009–12 (compared to 1,100 during the peak of the savings and loan crisis of 1986–90).22 More than 400 institutions are on the FDIC watch list, representing over $300 billion in assets. At least another 600 have Texas ratios that indicate levels of distress that could lead to short-term failure.23 Accrual losses in such asset classes as traditional commercial and industrial loans, commercial real estate, and nonmortgage consumer lending will likely peak in 2010, and many commercial banks are still rightly focused on survival. However, many others now have sufficient capital to both absorb prospective losses and simultaneously reshape their businesses. This is already happening through opportunistic acquisitions, as demonstrated by JPMC, Bank of America, PNC, Wells Fargo, BB&T, and others. With many commercial banks trading at or below book value, and with regulators willing to enter into unprecedented loss-sharing agreements on the asset side of the balance sheet, there will be attractive opportunities for consolidation. Those who can strike good deals and then adopt a superior business model are likely to outperform the industry. Human Capital One of the defining elements of this banking crisis is the bright light it shines on compensation practices in the financial services industry. While the focus has mostly been on the investment banking industry and the question of the link between compensation and true shareholder value added, the commercial banking industry has not escaped attention in this regard. The TARP capital for commercial banks came with restrictions on executive compensation. The Federal Reserve has also launched a wide horizontal compensation review of large commercial banks to better assess whether executive compensation is truly risk adjusted and aligned with long-run value creation rather than short-term nominal profitability. 22. FDIC (various years a). 23. The Texas ratio is calculated by dividing the value of the lender’s nonperforming assets (nonperforming loans + real estate owned) by the sum of its tangible common equity capital and loan loss reserves.

02-0404-1 chap2.indd 49

7/12/10 6:00 PM

50

alan m c intyre and michael zeltkevic

One consequence of this focus on compensation is that the commercial banking industry may struggle to attract and retain the best human capital. If, as our baseline suggests, we are looking at medium term returns barely above the cost of capital, there is clearly a risk that the best talent in the industry will look to deploy that talent in less-regulated parts of the financial services industry, with lower levels of public scrutiny and higher personal returns to competence. The danger for commercial banks in accepting the view that they are industry utilities is that they will attract human capital that is more bureaucratic than value creating, and mediocre returns will then become a self-fulfilling prophecy. A challenge for those institutions that aspire to outperform is therefore to create attractive positions with enough financial and nonfinancial compensation to attract and retain the best talent while also ensuring that compensation is truly aligned with medium-term risk-adjusted value creation. As the lengthy, and ultimately fruitless, search for an external successor to Ken Lewis at Bank of America demonstrated, this will not be a straightforward task. Innovation With traditional sources of profitability under pressure from both macroeconomics and government regulation, the best-managed banks will look at the needs of their customers, at their assets and capabilities, and at the competitive environment to identify opportunities for new and profitable businesses. There is obviously a strong link between such innovation and good management: without the right talent to drive it, innovation will not come easily. Compared to most industries, financial services has not been a hotbed of innovation. While you can point to Bank of America’s introduction of the credit card, the creation of money-market mutual funds, and a small number of other industry-changing innovations, the income streams of most commercial banks are remarkably similar to those of the 1970s and are pretty much static in comparison to, say, the computer industry. Where innovation has happened—for example in asset products and funding mechanisms (pay option subprime mortgages and structured credit securities)—many of those involved turned out to be major contributors to the current financial crisis. So what might profitable innovation in commercial banking look like in the years up to 2015? Just as the sources of the current crisis were unexpected for many in the industry, the sources of profitable innovation are also not obvious. One hypothesis is that the industry will make better use of its centrality to all commercial activity and find ways to monetize the huge amount of information that flows through commercial banks on a daily basis. Just as iTunes and Amazon tap into customer information streams to customize their services and shape their

02-0404-1 chap2.indd 50

7/12/10 6:00 PM

the uncertain future of u.s. commercial banking

51

offerings to their customers, might banks also be able to use the vast behavioral information they are privy to to improve not only financial but also nonfinancial product offerings? Another possible source of innovation is in retirement services. At the moment the primary needs of U.S. retirees (retirement income, medical coverage, intergenerational wealth transfer, long-term care) tend to be served by a variety of products and services. Might there be an opportunity for banks to take the lead in bundling products to better serve these important and interconnected needs? Finally, are there opportunities to simply improve the basic products and services of commercial banking in innovative and profitable ways? The atomization of the mortgage value chain across originators, servicers, packagers, credit enhancers, securitizers, and ultimate investors constrains cross-product innovation such as offset mortgages, in which savings balances reduce the interest payment on longrun debt by offsetting part of the principal balance. In this turbulent time for the industry, it may be that there is now space for product innovations to gain ground. So despite the profitability of the commercial banking industry returning to long-run average levels, there will be still be opportunities for individual institutions to outperform. Such performance will require good positioning in the right businesses and geographies plus outstanding management of risk, balance sheets, operations, human capital, and innovation. To generate returns in the 15–20 percent range, disciplined and smart execution of a traditional commercial banking business model will be needed—a capability we hope will be more evident in the commercial banking industry of the future than it was in the late 2000s.

References Bretton Woods. 2008. “Overdraft Study.” FDIC. 2008. “Study of Bank Overdraft Programs.” ———. 2009. “Study of Bank Overdraft Programs.” ———. Various years (a). “FDIC Failed Bank List.” ———. Various years (b). “Historical Statistics on Banking.” ———. Various years (c). “Statistics at a Glance.” Federal Reserve. 2007. “Survey of Consumer Finances.” ———. 2009a. “Assets and Liabilities of U.S. Commercial Banks.” News Release H8. ———. 2009b. “Survey of Consumer Finances.” ———. 2010. “Consumer Credit Statistical Release.” Statistical Release G19. ———. Various years. “Flow of Funds, Mortgage Bankers Association.” Morgan Stanley. 2009. “Brokerage Report.” Mortgage Bankers. Various years. “Mortgage Bankers Performance Report.” Nilson Report. 2009. Issue 924, April. Thomson Reuters. Various years. “Datastream.”

02-0404-1 chap2.indd 51

7/12/10 6:00 PM

02-0404-1 chap2.indd 52

7/12/10 6:00 PM

kei kodachi tetsuya kamiyama

3

Regulatory Changes and Investment Banking: Seven Questions

W

ork is currently going on under the auspices of the Group of Twenty (G-20) to reform the international financial system to prevent a recurrence of the present global financial crisis. The G-20 identifies the root causes of the crisis and voices its concern about financial sector practices as follows: “During a period of strong global growth, growing capital flows, and prolonged stability earlier this decade, market participants sought higher yields without an adequate appreciation of the risks and failed to exercise proper due diligence. At the same time, weak underwriting standards, unsound risk management practices, increasingly complex and opaque financial products, and consequent excessive leverage combined to create vulnerabilities in the system.”1 The G-20 has set the goal of learning lessons from the crisis while at the same time dealing with a host of issues. These include improving the regulation and oversight of systemically important institutions, enhancing macro–prudential policy, improving international cooperation among supervisors while harmonizing bankruptcy frameworks for financial institutions, mitigating against procyclicality in regulatory policy, and reviewing compensation practices as they relate to incentives. The group is also drawing up a list of reforms to be implemented in

1. Group of Twenty (2008).

53

03-0404-1 chap3.indd 53

7/12/10 6:01 PM

54

kei kodachi and tetsuya kamiyama

the near future. Thus investment banking, if the G-20 reforms are adopted, will be much influenced by regulatory reforms.2

Questioning Some G-20 Reforms Although we agree with the aim of the G-20—preventing a recurrence of the present crisis—some of our views on the direction the reforms should take differ from those of the G-20. In particular, we question the need for the following eight reforms relating to investment banking. —Is increasing regulatory capital an effective means of preventing another crisis? —Is there any need to increase capital requirements for trading books? —Is the simple leverage ratio requirement appropriate? —Does the market for structured products need more rules? —Are the rules governing over-the-counter derivatives appropriate? —Are the rules governing short selling appropriate? —Do hedge funds need to be regulated?

Is Increasing Regulatory Capital an Effective Means of Preventing Another Crisis? The present global financial crisis stemmed from the subprime mortgage crisis in the United States. Banks raised money from foreign sovereign wealth funds and other private sector sources during the initial phase of the crisis but were later obliged to accept injections of public money as the crisis deepened. Before the crisis many banks met minimum capital requirements but later were short of capital. It therefore seems that increasing the banks’ regulatory capital has become the most important item on the G-20’s agenda of financial reform. However, we ask, will a higher level of regulatory capital prevent another crisis? We consider this question from several perspectives: maturity transformation and market liquidity, turmoil in liquidity markets, asset price effects, fire-sale externality, and capital buffers. Maturity Transformation and Market Liquidity The present financial crisis was probably caused not only by an impairment of bank assets both on and off balance sheets but also by a depletion of market liquidity on a global scale (namely market-oriented systemic risk). Examples of market-oriented systemic risk include the stock market crash of 1987, when stock markets malfunctioned and share prices fell sharply; the 2. Unless indicated otherwise, bank should be understood to mean either a commercial bank or an investment bank.

03-0404-1 chap3.indd 54

7/12/10 6:01 PM

seven questions on financial reform

55

Figure 3-1. Market-Based Assets and Bank-Based Assets in the United States, 2007, Second Quarter $ trillion 16 ABS issuers 4.1

14 12

Broker-dealers 2.9

10

Credit unions 0.8 Saving institutions 1.9

Finance companies

8 6

GSE mortgage pools

4

Commercial banks 10.1

GSEs 3.2

2 0

Market-based assets

Bank-based assets

Source: Adrian and Shin (2009).

Long-Term Capital Management crisis of 1998; and the junk bond market turmoil of 1989–90, when interest rate spreads widened and liquidity declined.3 The reason attention has switched from bank runs to market gridlock as a cause of systemic risk is probably that, since 1990 or so, the importance of the market-based financial system (the role of markets in financial intermediation) has increased relative to that of the traditional banking system.4 As a matter of fact, U.S. flow-of-funds data show that market-based assets exceeded bank-based assets before the present crisis (figure 3-1).5 3. Hendricks, Kambhu, and Mosser (2007) deals with the differences between systemic risks to banking systems and systemic risks to market-oriented systems. 4. U.K. Financial Services Authority (2009b). This publication identifies the following as the characteristics of the financial model in recent years: growth of the financial sector relative to the rest of the economy; increasing leverage both on- and off-balance sheets; growth of shadow banking as a means of transforming maturity; misplaced reliance on sophisticated mathematics; and the procyclicality of ratings, triggers, margins, and “haircuts.” 5. According to Adrian and Shin (2009), market-based assets totaled more than $16 trillion, while bank-based assets were less than $13 trillion.

03-0404-1 chap3.indd 55

7/12/10 6:01 PM

56

kei kodachi and tetsuya kamiyama

In addition, maturity transformation (the traditional function of the banking system, whereby banks hold assets with longer maturities than their liabilities) has increased outside the traditional banking system. Maturity transformation outside the banking system that offered banking similar opportunities includes that performed by off-balance-sheet instruments, such as structured investment vehicles and conduits; the use by investment banks of short-term liabilities (especially the rapid development of the overnight repurchase (repo) market) to fund long-term assets; and the understanding that U.S. mutual funds would not fall below $1 per share.6 As a result, managers of structured investment vehicles, who had issued assetbacked commercial paper (ABCP) to fund these products, found that liquidity dried up as ABCP issuance declined (since the summer of 2007, when investors realized the risks attached to subprime mortgages). These managers then had no option but to depend on liquidity enhancement by their sponsor banks. At the same time, managers of money-market mutual funds who had invested in ABCP found their own liquidity drying up as a result of the collapse of ABCP market. Repo markets also found themselves in difficulty as a result of larger “haircuts” (the discount-to-asset value that determines how much a bank can borrow). In other words, the maturity mismatch created between long-term assets and short-term liabilities by maturity transformation entailed a risk that liquidity would decline. The present crisis is characterized by the close correlation between banks’ funding liquidity and market liquidity. This correlation makes the present crisis different from past crises. What appears to have happened is that banks that raised funds on the markets found themselves in difficulty when market liquidity dried up and that lenders anxious about the solvency of these banks rushed to secure their cash, causing a systemic liquidity crisis. Turmoil in Liquidity Markets The focus of interest regarding the relationship between market liquidity and banks’ funding liquidity has been the repo market. Although there are no official figures on the size of this market, it is estimated that between 2002 and the end of 2007 the U.S. repo market doubled in size, to about $10 trillion, roughly equivalent to the total assets of the U.S. banking sector.7 One of the reasons the repo market grew so rapidly was that banks funded their positions largely by means of overnight repo and other securities lendings.8 6. U.K. Financial Services Authority (2009). 7. Gorton and Metrick (2009). 8. According to Lehman Brothers’ balance sheet, November 2005, a quarter of the company’s assets consisted of short positions, long-term debt accounted for only a small proportion of Â�borrowing, and Â�shareholders’ equity was 4 percent of assets. The rest was financed by repurchases and other securities-backed loans.

03-0404-1 chap3.indd 56

7/12/10 6:01 PM

seven questions on financial reform

57

Table 3-1.╇ Typical Haircut, or Initial Margin, 2007 and 2008 Percent Asset U.S. Treasuries Investment-grade bonds High-yield bonds Equities Investment-grade corporate CDSs Senior leveraged loans Mezzanine leveraged loans ABS-CDOs AAA ABS-CDOs AA ABS-CDOs A ABS-CDOs BBB ABS-CDOs equity CLOs AA Prime MBSs ABSs

April 2007

April 2008

0.25 0–3 10–15 15 1 10–12 18–25 2–4 4–7 8–15 10–20 50 4 2–4 3–5

3.0 8–12 25–40 20 5 15–20 35+ 95a 95a 95a 95a 100a 10–20 10–20 50–60

Source: IMF (2008). “Financial Stress and Deleveraging,” Global Financial Stability Report. a. Theoretical haircuts as CDOs are no longer accepted as collateral.

Haircuts before and after the financial crisis also differ significantly (table 3-1). For example, the haircut on asset-backed securities rose from a range of 3–5 percent to a range of 50–60 percent during the same period. This meant that banks could borrow only half or less of the market value on such securities. Using lower-grade securities as collateral made it difficult for banks to borrow by means of repo. Haircuts on repurchases after market liquidity drastically declined caused a market panic.9 Since increasing a haircut reduced the amount that could be borrowed by a corresponding amount, the only possible responses were to borrow more money, raise capital, or sell assets. In the autumn of 2007 the response was still to raise capital. However, as it became more difficult to either raise capital or borrow money, banks had no alternative but to sell assets. As the price of these assets declined, so did their value as collateral. This caused the price of other assets to decline and led to what appears to have been an insolvent financial system. 9. As the repurchase market is estimated at between $8 trillion and $10 trillion, a 20 percent increase in the average haircut would force the banks to raise another $1.6 trillion to $2.0 trillion. See Gorton and Metrick (2009).

03-0404-1 chap3.indd 57

7/12/10 6:01 PM

58

kei kodachi and tetsuya kamiyama

Another problem in markets is that of rehypothecation. The U.S. Securities Exchange Act of 1934 prevents a broker-dealer from using its customers’ securities to finance its proprietary activities; the broker-dealer may use, or rehypothecate, an amount up to 140 percent of the customer’s debit balance.10 By contrast, there are no limits on rehypothecation. It is said that the reason that hedge funds opted for funding in Europe is that leverage is not capped. In the United Kingdom, prime brokers can rehypothecate their customers’ assets along with their own as collateral for securities, including repurchase, and their customers’ securities and cash in prime brokerage accounts can be commingled with the prime broker’s asset. If a prime broker in the United Kingdom becomes insolvent, the client may no longer secure creditors. This is not the case in United States, where the Securities Investor Protection Act protects investors. Since the bankruptcy of Lehman Brothers rehypothecation has been declining, as prime brokers have been required to offer more cash collateral in place of securities. This significant reduction in rehypothecation may lead to a systemic shortage of collateral, which could cause a funding problem for prime brokers. Asset Price Effects and Margin Spirals In financial systems in which balance sheets are valued at market, a rise in asset prices as a result of an increase in demand boosts the balance sheets of banks (and leveraged firms such as hedge funds). This in turn creates further demand for assets, and the resulting increase in asset prices further boosts balance sheets in a positive feedback effect.11 This is the result of procyclical leverage. In other words, the leverage boosts balance sheets when asset prices are rising. However, the opposite occurs when the business cycle is in a downturn. A decline in the price of the assets owned by banks depletes their capital cushion if the decline in their net worth is greater than the decline in the price of the assets. Fair value plays a decisive role in such cases by affecting asset prices. In a financial system in which prices are marked to market, defaults are propagated by changes in prices. If banks value their balance sheets at market, changes in asset prices will produce losses that will also affect banks and other financial institutions that have no exposure to those banks. Those losses will reduce the funding liquidity of many banks, which will trim their holdings of assets that threaten to produce losses should asset prices fall further. In particular, during a financial crisis spiraling losses lead to even greater changes in prices and, eventually, systemic risk.

10. See Rule 15c3-3 of the Securities and Exchange Act. Also see Singh and Aitken (2009). 11. Brunnermeier and others (2009).

03-0404-1 chap3.indd 58

7/12/10 6:01 PM

seven questions on financial reform

59

In other words, the procyclicality produced by fair-value accounting is a major factor in asset price effects. While fair-value accounting of financial products increases transparency and market discipline and even improves the marketability of illiquid assets, models that measure the fair value of nonmarketable assets can have latent costs in the form of high volatility in revenue, profits, and balancesheet positions as well as distorted data if they vary from one bank to another or are inaccurate.12 It has also been pointed out that the growing use of fair-value accounting of financial products has encouraged market participants to take risky positions in response to changes in asset prices.13 Furthermore, as we have seen in connection with the turmoil on repurchase markets, it can lead to a cycle of rising haircuts and margins. Rising haircuts and margins lead to deleveraging, which forces investors to sell, which leads to even greater haircuts and margins. During the present crisis, market prices seemed to deviate sharply from their fundamental values, and bank balance sheets were impaired as market liquidity dried up. This may have been compounded by rising haircuts on repurchase markets and negative synergies from impaired asset prices and reduced liquidity. Fire-Sale Externality and Regulatory Capital Normally, if a bank suffers a serious loss, it will comply with capital adequacy requirements by either trimming its assets or issuing more equity. When the market is under stress, however, weak demand for new equity is likely to oblige it to trim its assets.14 If one bank disposes of some of its assets, banks with similar assets may be induced by falling market prices to sell theirs. Such selling exerts pressure on the banks’ capital and obliges them to close their positions. In this way, one bank’s disposal of some of its assets can lead to other banks selling some of their assets in a process of systemic deleveraging. Such negative feedback is sometimes referred to as fire-sale externality.15 The externality would be more influential in the market-based financial system than in the traditional banking system because of the existence of asset price effects and margin spirals. As we saw in the case of the repurchase markets, fire-sale externality can cause short-term funding difficulties. Even if a bank has plenty of capital, it may have difficulty raising capital if it is relying on short-term funding and using securities as collateral, as any decline in the price of those securities may lead to an increase in the haircut or margin. As a result, a bank that sells some of its assets at 12. Taylor and Goodhart (2007). 13. Financial Stability Forum (2009). 14. A share issue may send the market a negative signal (for example, of further losses), so it is a difficult option. Companies under pressure of time may therefore have no alternative but to trim their assets. 15. Kashyap, Rajan, and Stein (2008).

03-0404-1 chap3.indd 59

7/12/10 6:01 PM

60

kei kodachi and tetsuya kamiyama

fair value may find that this puts pressure on the capital position of other banks whose financing depends on the fair value of their assets and leads to further sales of assets by the bank. The question is whether increasing regulatory capital as a result of this externality would be an effective means of preventing another market-oriented liquidity crisis. The U.S. Treasury has stated its desire to require banks to have enough capital to withstand systemic liquidity shocks that can cause a deleveraging spiral.16 However, we think fire-sale externality not only depends on the distinctive nature of banks’ assets and liabilities (such as the valuation of financial products, the mismatch between long-term assets and short-term liabilities, and their dependence on short-term market funding) but also is affected by factors outside the banking system (maturity transformation in markets, margins and haircuts, rehypothecation, and other practices in markets). We think solving appropriately the problems caused by these factors mitigates the necessity of strengthening bank capital. We wonder whether any attempt to incorporate these factors in bank capital adequacy requirements would not be fraught with difficulty and therefore require banks to incur excessive capital costs. Excessive regulatory capital, overwhelming banks’ economic capital, could be a disincentive for banks. Since market participants do not want banks to increase their relatively expensive regulatory capital when times are good, rules that require banks to increase their capital by more than the market wants are likely to be a strong incentive for banks to reduce capital charges where at all possible.17 Proposals that banks should pay a fee that would automatically commit another party to exchange debt securities for its equity at a time of stress (contingent capital) or take out insurance against a macroeconomic crisis are intended to avoid excessive capital costs.18 Raghuram Rajan, of the University of Chicago, has said that “ironically, faith in draconian regulation is strongest at the bottom of the cycle—when there is little need for participants to be regulated. By contrast, the misconception that markets will take care of themselves is most widespread at the top of the cycle—the point of maximum danger to the system. We need to acknowledge these differences and enact cycle-proof regulation.”19 Our view, however, is that rather than recapitalize banks to prevent the recurrence of market-oriented systemic risk, we need a financial system and infrastructure that will not produce negative feedback through balance-sheet-to-liquidity 16. U.S. Department of the Treasury (2009a). 17. Raghuram Rajan, “Cycle-Proof Regulation,” The Economist, April 11–17, 2009. 18. The Dodd-Frank bill includes provisions that require systemically important financial institutions to issue contingent capital instruments. 19. Rajan, “Cycle-Proof Regulation.”

03-0404-1 chap3.indd 60

7/12/10 6:01 PM

seven questions on financial reform

61

shock, that will mitigate overreliance on liquidity through marketability, and that will improve risk control (including appropriate initial margin, haircut and rehypothecation).

Is There Any Need to Increase Capital Requirements for Trading Books? On July 13, 2009, the Basel Committee on Banking Supervision (BCBS) approved a final package of measures to enhance the Basel II framework for market risk and set a deadline of the end of 2010 for its implementation.20 The main enhancements are the adoption of a stressed value-at-risk measure requirement, the adoption of the incremental risk charge, and revisions to charges on securitization exposure.21 Enhancing the Basel II Framework for Market Risk In view of the rapid developments in financial innovation and risk management that have taken place since Basel I was adopted following the Basel Accord of 1988, one of the aims of Basel II has been to keep pace with those developments. Under Basel I a risk asset approach, in which credit risks are weighted using a standard risk weight, was adopted, whereas under Basel II a more sensitive approach to measuring credit risk (for example, by means of internal ratings) has been tried, and new rules (including an internal, model-based approach) covering operational risks have been adopted. An internal, model-based approach has been in use for market risk ever since the Basel I agreement was revised to cover that in 1996. The framework requires banks to have regulatory capital based on the potential loss as calculated with a value-at-risk model using a ten-day holding period and 99th percentile, onetailed confidence interval. Although Basel II adopted rules covering operational risk and a more sophisticated approach to credit risk, it left the 1996 rules on market risk more or less unchanged. As a result of the financial crisis that began in mid-2007 many of the losses and most of the leverage occurred on trading books, and many banks found that their losses exceeded their minimum capital requirements under market risk rules. 22 The adoption of the stressed value-at-risk measure and the incremental risk charge and the revisions to charges on securitization exposure are therefore attempts to deal with some of the problems that arose during the latest financial 20. Basel Committee on Banking Supervision (2009). 21. The stressed value-at-risk measure is calculated using its average for the past sixty business days, including the base date. 22. Financial Stability Forum (2009).

03-0404-1 chap3.indd 61

7/12/10 6:01 PM

62

kei kodachi and tetsuya kamiyama

crisis. The stressed value-at-risk measure was adopted because the existing risk measure failed to recognize the full potential for losses during the crisis. A capital charge is additionally levied on stressed value at risk with model inputs calibrated to historical data from a continuous twelve-month period of significant financial stress relevant to the bank’s portfolio. As an example, for many portfolios, a twelve-month period relating to significant losses in 2007–08 would adequately reflect a period of such stress. The adoption of the incremental risk charge was intended to measure incremental risk including the default risks of trading book assets, for which only market risk had previously been measured, and to levy appropriate capital charges on these. Also, a new category (resecuritization exposure) was adopted to reflect the fact that complex resecuritized products such as asset-backed securities and collateralized debt obligations had been responsible for some of the losses on trading books and that risk weights were adjusted to align the treatment of banking books and trading books. Procyclicality of Value at Risk We have some reservations about the use of the stressed value-at-risk measure, because it is subject to a capital charge in addition to the charge on normal value at risk. Conceptually, the potential loss calculated using a simple value-at-risk measure should be a subset of the potential loss calculated using the stressed value-at-risk measure. Although the BCBS is confident that the adoption of a stressed value-at-risk measure will help to dampen the cyclicality of the regulatory capital framework, we wonder whether its explanation of the need for both is adequate.23 The value-at-risk measure itself has been questioned because of its procyclicality. It has been pointed out that it is subject to the vagaries of the economic cycle if the period of observation of the time series used is short term (for example, only twelve months) and likely to be highly procyclical.24 For example, market liquidity is generated if investors take a high risk because of the observed low volatility early on and the low risk that this suggests. However, if their later observation shows liquidity drying up and volatility rising, market confidence declines because of high volatility. It is well known that value at risk tends to increase during periods of stress.25 The value-at-risk time series for U.S. banks shows that, other than in the case of Merrill Lynch, value at risk has continued to rise despite deleveraging (figure 23. Basel Committee on Banking Supervision (2009). 24. Financial Stability Forum (2009). 25. Greenlaw and others (2008).

03-0404-1 chap3.indd 62

7/12/10 6:01 PM

seven questions on financial reform

63

Figure 3-2. Value at Risk, U.S. Banks, 2007–09 a $ million 400

Citigroup

350

JP Morgan Chase

300 250 Goldman Sachs

200

Bank of America

150 100

Morgan Stanley

50 0

Merrill Lynch

Q1

Q2

Q3 2007

Q4

Q1

Q2 Q3 2008

Q4

Q1

Q2 2009

Source: Nomura Institute of Capital Markets Research, based on corporate financial statements. a. The VaR confidence interval for Goldman Sachs, Morgan Stanley, and Merrill Lynch is 95%, while that for Citigroup, Bank of America, and JP Morgan Chase is 99%. With regard to the observation period, Goldman Sachs has weighted its historical data in order to reflect the most recent data more clearly. However, it has not said how it has done this. Morgan Stanley's observation period is four years, Bank of America's three years, and JP Morgan's one year. Neither Citigroup nor Merrill Lynch has said what its observation period is. Merrill Lynch has not published any VaR data since FY09 Q1. Bank of America is alone in using a three-month-average VaR.

3-2). We are therefore concerned that the enhancement of market-risk rules (in the form of stressed value at risk) may actually increase procyclicality. Because of the adoption of this measure, trading books will be subject to capital charges at least double their previous levels. The European Commission has proposed introducing the stressed value at risk, suggesting that adding a capital buffer based on stressed value at risk compared to ordinary value at risk would roughly double current trading book capital requirements.26 Also during periods of stress the amount of regulatory capital required as calculated using the ordinary value-at-risk measure would increase, and regulatory capital would have to be topped up by the amount required by the stressed value-at-risk measure. As a result, banks could be obliged to reduce their trading positions, and market liquidity could decline, actually increasing volatility. 26. European Commission (2009).

03-0404-1 chap3.indd 63

7/12/10 6:01 PM

64

kei kodachi and tetsuya kamiyama

In a report on procyclicality in the financial system, the Financial Stability Forum recommends that the BCBS reduce its reliance on value-at-risk-based capital charges.27 Similarly, the U.S. Treasury has called for the reliance on value at risk to be reduced because of its procyclicality.28 This debate on the disadvantages of relying on value at risk also suggests to us that simply adding a capital charge on stressed value at risk to the existing charge may not be a good idea and may even exacerbate procyclicality. Regulatory Arbitrage One of the reasons that banks lost large sums of money on their trading books during the financial crisis and that the market risk framework needed to be enhanced is because lower capital charges on trading books than on banking books induce banks to make extensive use of the former.29 It has always been assumed that trading books would be used to hold highly liquid financial products traded short term and that, in view of the nature of these assets, these accounts should be subject to lower capital charges than banking books, especially when these assets were highly rated. Banks therefore held high-risk complex assets in their trading books, which were highly rated but illiquid and opaque in risk profile, and they engaged in regulatory arbitrage between them and their banking books. This situation was known before the crisis began.30 The BCBS once considered imposing restrictions on the kind of assets that could be held in trading books but decided, after strong opposition from market participants, to define what could and could not be held in them and to leave oversight of actual trading to internal auditors. This is probably why the issue of regulatory arbitrage between the two kinds of accounts remained unresolved until the financial crisis began. Consideration of Liquidity Our impression of the enhancements to the market risk framework is that the BCBS wanted to simply increase capital charges on trading books in general in order to discourage regulatory arbitrage from banking books. However, the heart of the problem lay in the fact that banks were using their trading books to hold illiquid assets instead of the highly liquid assets traded short term that they were 27. Financial Stability Forum (2009). 28. U.S. Department of the Treasury (2009a). 29. Acharya and Schnabl (2009). 30. Basel Committee on Banking Supervision (2005).

03-0404-1 chap3.indd 64

7/12/10 6:01 PM

seven questions on financial reform

65

expected to hold. Also there was no mechanism for adequately capturing the credit risk of the assets held in trading books. As a result, financial products with a complicated structure that had been given a high credit rating by rating agencies but had an opaque risk profile were valued (and subject to a capital charge) only on the basis of their market risk.31 Our solution to this problem would be to decide on the basis of an asset’s liquidity whether or not a bank should be allowed to hold it in its trading book. If banks had been subject in 2005 to restrictions on the assets they could hold in their trading books, the losses on these accounts might have been lower, even if only slightly. Another point is that it is the job of supervisors to monitor regulatory arbitrage. If we take into account the principles of pillar 2 of Basel II, it is a problem that requires regulators to take appropriate regulatory action to remedy.32

Is the Simple Leverage Ratio Requirement Appropriate? One of the causes of the financial crisis was banks’ high leverage. Before the crisis U.S. investment banks had an average leverage ratio of about 30. Some European banks had even higher leverage ratios, although comparisons are complicated by accounting differences. It is well known that the leverage ratio of many commercial and investment banks began to rise in 2003. However, it has been pointed out that neither risk-weighted assets nor value-risk-based capital adequacy ratios accurately captured leverage or risks inherent in high leverage.33 As IMF (2008) observes, the increase in banks’ risk-weighted assets was much less than that in their total assets. This shows that risk-based capital adequacy rules fail to capture leverage. Capital Adequacy Rules and Leverage In order to avoid high leverage, consideration is being given to imposing restrictions directly on banks’ leverage. However, the high-risk complex assets that banks held in their trading books rather than in their banking books were measured only by the market value at risk and thus were excluded from banks’ risk-weighted assets, making regulatory arbitrage possible. Accounting rules also 31. For example, UBS, which used to trade super-senior collateralized debt obligation tranches, recognized the yield on these tranches only as a pure alpha in its internal value at risk and stress test, because they were AAA rated. Clementi and others (2009). 32. Pillar 2 reads, “Supervisors should review and evaluate banks’ internal capital adequacy assessments and strategies, as well as their ability to monitor and ensure their compliance with regulatory capital ratios. Supervisors should take appropriate action if they are not satisfied with the result of this process.” 33. U.K. Financial Services Authority (2009b).

03-0404-1 chap3.indd 65

7/12/10 6:01 PM

66

kei kodachi and tetsuya kamiyama

played a role. For example, some of the banks that used international financial reporting standards consolidated their exposure to the conduits on their balance sheets but treated them as off-balance-sheet items under capital adequacy rules.34 This exposure is not included in their risk-weighted assets, either. In other words, capital adequacy rules were badly designed in that they failed to capture either the regulatory arbitrage of the different capital charges on banks’ banking and trading books or banks’ exposure to such off-balance-sheet vehicles as structured investment vehicles and conduits. However, the realization that risk-based capital adequacy ratios fail to capture leverage has led regulators to consider adopting simple non-risk-based leverage rules. Quality of Leverage Attention on the excessive leverage that contributed to the financial crisis tends to focus on its quantitative aspects. However, we think that quality of leverage was also a problem. Banks and other financial institutions were overdependent on market funding as a result of their belief in liquidity through marketability. It has been pointed out that, before the crisis, banks believed it safe to hold to maturity long-term assets funded by short-term liabilities on the grounds that the assets could be sold rapidly in liquid markets if needed.35 However, while this seemed reasonable before the crisis, it was no longer reasonable after mid-2007, when banks simultaneously began to have to close out positions. On the repo market, before the crisis, banks borrowed mainly overnight (rolling over their positions), using a variety of low-quality securities (such as assetbacked securities or collateralized debt obligations) as collateral.36 As a result, when repo haircuts were increased as the market value of those securities fell and counterparty risk increased, banks found it increasingly difficult to borrow on that market. This experience illustrates the risk of relying excessively on market funding in periods of stress, when market liquidity dries up. In other words, to reduce such risks, banks need to pay attention to the quality of their leverage—by for example extending the maturity of their liabilities and reducing their maturity mismatches. It ceases to be simply a matter of quantitative leverage ratios. Not only that, regulating simply on the basis of leverage ratios has serious side effects. For example, if a bank increases its leverage by secure funding such as retail depositors, this may seem sensible in terms of the stability of the source. 34. Acharya and Schnabl (2009). 35. U.K. Financial Services Authority (2009b). 36. U.K. Financial Services Authority (2009b).

03-0404-1 chap3.indd 66

7/12/10 6:01 PM

seven questions on financial reform

67

However, regulating such types of funding simply by using leverage ratios would limit the size of the bank’s balance sheet. In Asian and emerging market economies, where capital markets are less developed than the banking sector, the banking sector tends to form the hub of the financial system, with traditional banking weighted heavily. Since simple leverage ratios take no account of the quality of leverage, their adoption could make it more difficult for banks whose funding is based on the relatively stable source of retail deposits to manage their balance sheets. In such cases, leverage rules could have the very opposite of their intended effect. In other words, we think it might be better to refrain from simplistic suggestions of a level playing field that take no account of the structure of a bank’s balance sheet. Also, if it were decided to adopt leverage ratios, it would in our view be better to do so as a measure of systemic risk in the financial system, that is, from a macro prudential view.

Does the Market for Structured Products Need More Rules? There is general agreement that one of the causes of the financial crisis was the rising delinquency rate on subprime mortgages and the losses that financial institutions were forced to take as the value of structured products incorporating these mortgages declined. The failures include the predatory lending practices of mortgage brokers; the lack of governance at credit rating institutions, as illustrated by their willingness to give high ratings to structured products without taking into consideration factors such as model errors and a lack of time-series data (or if they did take them into consideration, their failure to act on the results); the failure to foresee that credit diversification would not function during such a period of extreme stress (it may have been impossible to foresee this effect because of the simultaneous housing market crash); and the assumption by both lenders and borrowers that the price of the real estate on which subprime loans were granted would continue to rise. Once the subprime mortgage problem became apparent, the U.S. government started to increase its regulation and supervision of credit rating agencies and mortgage brokers in an attempt to deal with the aforementioned problems. The European Commission also drafted legislation to improve the regulation of credit rating agencies. The proposed measures (such as compulsory rotation of analysts and a prohibition on consulting about the design of structured products) address the conflicts of interest that face credit rating agencies. In both the United States and Europe four further rules have been proposed to give originators an incentive to maintain the quality of structured products. The

03-0404-1 chap3.indd 67

7/12/10 6:01 PM

68

kei kodachi and tetsuya kamiyama

first of these is a 5 percent risk-retention rule for originators of such products.37 The second is an enhanced disclosure requirement such as that on underlying assets. In Europe originators that fail to comply will be subject to risk weightings of 250–1,250 percent when their regulatory capital is computed. The third is the prospect in the United States of more stringent representation and warranty requirements under the Investor Protection Act of 2009. The fourth is a change in accounting standards to have off-balance-sheet entities transferred back to balance sheets. In the United States the notion of qualifying special-purpose entities will be abolished, and quantitative tests on the consolidation of variableinterest entities will be replaced by qualitative tests.38 We understand that the result of this will be that most of the off-balance-sheet vehicles that have been used to issue structured products will be moved back onto balance sheets.39 In Europe, too, International Accounting Standard 27, which was published in December 2008, was due to be amended by the end of 2009 so that entities that, in substance, control special-purpose entities are required to move them back onto their balance sheets. We see a number of possible problems if markets for structured products are made subject to the additional and more stringent regulations mentioned above. Under the risk retention rules due to be adopted in Europe and the United States, risk would be concentrated in financial institutions. We see this as impairing the risk diversification effect achieved by securitization. Even if risk retention rules do achieve the objective of giving originators an incentive to maintain the quality of structured products, this could also be achieved by more stringent representation and warranty requirements and stronger repurchase obligation provisions. This is because an originator would be legally and financially liable in the event of certain losses on structured products. As we have already seen, representation and warranty are dealt with by the Investor Protection Act of 2009, while progress on stronger repurchase obligation provisions has been made by requiring originators to disclose repurchases and by clarifying the repurchasing criteria of the American Securitization Forum.40 We therefore see little point in adopting risk retention rules even if they give originators an incentive to maintain the quality of structured products. 37. The Investor Protection Act of 2009 is in House-Senate conference as part of the Wall Street Reform and Consumer Protection Act, as of July 2010. The amendments to the EU’s “Capital Adequacy Directive” were passed by the European Parliament and the Council of Ministers in July 2009. 38. Financial Accounting Standards Board (2009). 39. Fitch Ratings (2009). 40. In July 2009 the American Securitization Forum invited public comment on its draft model, and it released the final version of the model in December 2009.

03-0404-1 chap3.indd 68

7/12/10 6:01 PM

seven questions on financial reform

69

As the United Kingdom’s “Turner Review” points out, U.S. and European financial institutions may already have been taking senior tranche positions in an acquire-and-arbitrage approach instead of an originate-to-distribute approach and may already have had exposure either in the form of credit default swaps or, indirectly, via structured investment vehicles.41 Commercial banks and brokerdealers owned not only a large proportion of subordinated collateralized debt obligations, for which there is unlikely to have been much demand from investors, but also a large proportion of AAA-rated securities, including super-senior tranches of collateralized debt obligation. This is another reason that we see little point in having risk retention rules. Also, risk retention rules and transferring securitization vehicles onto balance sheets both involve recognizing structured products that originated on financial institutions’ balance sheets and thus may lead to additional capital charges on originators. Two of the benefits of securitization to originators are diversification of credit risk and the ability to effectively recycle capital by selling assets or moving them off the balance sheet. However, risk retention rules and moving structured vehicles onto balance sheets would impair these benefits and affect financial institutions’ business incentives. The impact of tighter rules on securitization would affect not only financial institutions. Securitization has been a major source of credit in the United States, underpinning both mortgages and such consumer vehicles as auto loans and credit cards. The reduction in origination that would result from the adoption of risk retention rules or from moving structured vehicles onto balance sheets could affect 16 percent of non-government-sponsored mortgages and 25 percent of consumer loans. We therefore think that tightening regulation of the market for structured products would conflict with the aims of government initiatives such as the term asset-backed loan facility and the public-private investment program: namely, to move nonperforming assets off balance sheets and to revive the market for structured products. Perhaps we need to remind ourselves of the original aims of securitization: to diversify risk at a low capital cost and, by extension, to make it easier for endborrowers to borrow. In our view, risk retention rules and moving structured vehicles onto balance sheets make it more difficult to achieve these aims and are therefore undesirable. On the other hand, we recognize that the aim of the rules (namely, to give originators a greater incentive to maintain the quality of their structured products) is reasonable and that stronger provisions on matters such as

41. U.K. Financial Services Authority (2009b).

03-0404-1 chap3.indd 69

7/12/10 6:01 PM

70

kei kodachi and tetsuya kamiyama

disclosure, representation and warranty, and repurchases will therefore be needed to ensure the sound development of the market.

Are the Rules Proposed for Over-the-Counter Derivatives Appropriate? In the debate on the causes and consequences of the global financial crisis, the view has gained ground that over-the-counter derivatives—especially, credit default swaps (CDSs), which have mushroomed in recent years—have fueled market uncertainty. The reasons for this view include the interconnectedness of CDSs, counterparty risk, and lack of market transparency. Although protection sellers hedged their risks and their counterparties hedged theirs, such bilateral relationships became interwoven like a spider’s web as protection sellers and their counterparties were joined by naked CDS holders (that is, investors holding the CDSs but not the reference securities). As a result, transparency became an issue as it became impossible to tell who owned what and the motivation of the ownership. In addition, there was the counterparty risk that the protection seller might become insolvent. Then the problem of interconnectedness—if losses were propagated along the transaction chain or if the protection seller had to sell the underlying assets in order to meet his obligation to pay— could have a knock-on effect on the stock market.42 In practice there was also a problem of concentration risk, as in the case of AIG, when a single entity had a massive unhedged net exposure. It has also been said of CDSs that they sometimes distort normal market decisionmaking. For example, it has been reported that when General Motors was in debt reduction talks with its creditors many bondholders who owned CDSs referenced to its bonds preferred the company to go into chapter 11 bankruptcy (a credit event that would trigger protection) than to reduce its debt.43 Similarly, in May 2009 the U.S. Securities and Exchange Commission filed against a hedge fund (and the investment bank that provided the information) for buying CDSs after obtaining insider information about the reference bond.44 With regard to naked CDSs, it has been reported, for example, that, if their volume is low, even small transactions can widen the spread and, as a result, have a negative impact on the share price; and that a profit can be made by buying these instruments and selling the corresponding shares short if the their price rises and share price declines.45 The U.S. Treasury has therefore been considering tightening the regulation and supervision of over-the-counter derivatives such as CDSs and interest rate swaps 42. Sirri (2009). 43. Henny, Sender, “Credit Insurance Hampers Restructuring Plan,” Financial Times, May 12, 2009. 44. U.S. Securities and Exchange Commission (2009b). 45. U.S. Government Accountability Office (2009).

03-0404-1 chap3.indd 70

7/12/10 6:01 PM

seven questions on financial reform

71

(the Over-the-Counter Derivatives Markets Act of 2009 was proposed in August 2009). We think the following two aspects of the bill are particularly important. First, the bill proposes that trading in nonstandard derivatives should be subject to additional capital and margin requirements in order to enhance standardization and that trading in and settlement of standardized derivatives should be carried out on exchanges and clearinghouses.46 It also proposes the creation of a repository for recording and storing transaction data. Trading in derivatives has traditionally been mainly bilateral, and this has given rise to the aforementioned problem of interconnectedness. The bill therefore proposes that the clearing counterparty should, as far as possible, be a central counterparty in order to simplify collateral transaction, reduce counterparty risk, and increase settlement transparency and security. Similarly, the European Commission has proposed that over-the-counter derivatives be standardized, that the use of exchanges and clearinghouses be encouraged, and that a repository be established.47 In addition, at the end of July 2009, ten CDS dealers began trading via two central counterparties, ICE Clear Europe and Eurex Clearing. Second, the bill proposes that the Securities and Exchange Commission (SEC) and the U.S. Commodity Futures Trading Commission (CFTC) be given the power to impose position limits and require trading reports from “major swap participants,” because trading in over-the-counter derivatives performs or affects a significant price discovery function with respect to regulated markets. We recognize that these proposals are intended to address the possible impact of naked CDSs or positions with a similar economic impact (such as synthetic collateralized debt obligations) on the volatility of cash stocks. In July 2009, before the Treasury bill was presented, House Financial Services Committee Chairman Barney Frank and House Agriculture Committee Chairman Collin Peterson published a joint concept paper that calls for the SEC or the CFTC to oversee the regulation of over-the-counter derivative trading. The concept paper focuses on the possible impact of naked CDSs on cash stocks and proposes even tougher measures on naked CDSs than those in the Treasury’s bill. For example, as well as proposing closer supervision of certain types of speculative positions, it labels naked CDSs “speculative trading” and proposes to ban such trading altogether. However, we have some doubts about some of this criticism and of the proposed stricter regulation of over-the-counter derivatives in general and of CDSs in particular. We cannot deny that spreads were very volatile after the collapse of Lehman Brothers in September 2008 and that this volatility was a cause of 46. As of early July 2010, the bill is in House-Senate conference as part of Congress’s comprehensive financial reform legislation. 47. Commission of the European Communities (2009a, 2009b, 2009c).

03-0404-1 chap3.indd 71

7/12/10 6:01 PM

72

kei kodachi and tetsuya kamiyama

uncertainty on financial markets. However, since April 2009 the International Swaps and Derivatives Association, to ensure security and transparency, has required its members to select auction settlement as the method for calculating payments due following a credit event affecting a reference entity. The aim of auction settlement is to avoid a shortage of cash bonds (and resulting price discrepancies) in the case of physical settlement and a shortage of reference prices in the case of cash settlement. Such auctions have actually been held. There is an observation that participation has been high (95 percent of those eligible to participate), that settlement has been efficient (with aggregate payments halved as a result of netting), and that the prices achieved have been fair (similar to the latest cash bond prices).48 Inasmuch as the problem lay with the interconnectedness of CDSs, we think that the use of International Swaps and Derivatives Association auctions, the standardization of CDSs, the requirement to use clearinghouses, the encouragement of the use of exchanges, and the recording and publication of trading and position data by repositories should suffice. The question of how exactly to standardize CDSs referenced to single issues has still to be resolved. As far as the problem of the interconnectedness of CDSs, which is seen by some as one of the causes of the financial crisis, is concerned, we need to remember that the credit markets themselves are interconnected and that the default of a participant in the credit market will translate into a loss for its creditors, however that is transmitted. Some therefore take the view that the interconnectedness of these instruments is not a problem.49 With regard to the proposals to tighten regulation of naked CDSs, we would like to point out, first, that unlike naked short selling the risk of default in the physical settlement of naked CDS transactions is minor, as the underlying securities are not necessarily needed to settle the transaction. Also any restrictions on naked CDSs could affect future transactions. For example, their prohibition would limit selling opportunities for initial purchasers. If that happened, there would probably be fewer incentives to purchase them with a view to selling them (that is, initial demand would decline), and supply would be reduced. Financial institutions would have difficulty managing their risk using CDSs. Unlike credit ratings from banks and rating agencies, the credit information provided by CDSs reflects the market’s assessment of an entity’s creditworthiness and is an invaluable and irreplaceable point of access to the credit Â�market’s Â�pricing

48. Helwege and others (2009). 49. Wallison (2008).

03-0404-1 chap3.indd 72

7/12/10 6:01 PM

seven questions on financial reform

73

function.50 A reduction in the supply of CDSs would deprive both investors and regulators of a means of assessing the creditworthiness of the financial sector in general and financial institutions in particular as well as risk affecting the efficiency of the cash bond market. A further point is that the debate about tightening the regulation of overthe-counter derivatives has been conducted from the perspective of swap dealers and investors. In the United States, however, nonfinancial corporations use over-the-counter derivatives to hedge the risk of fluctuations in the price of raw materials. We therefore hope that any reforms to the rules governing the trading of these derivatives will allow nonfinancial corporations to trade them for hedging purposes (for example, by exempting certain entities or not imposing onerous reporting requirements).

Are the Rules Proposed for Short Selling Appropriate? Following the fall in share prices on world stock markets in 2008, regulators in Europe and the United States imposed a series of temporary restrictions on short selling (especially the naked short selling of shares in financial institutions), as it was seen as having contributed to the fall. The time limit was then extended in European countries and in some countries even extended indefinitely. In the United States moves are under way to replace the temporary restrictions on short selling, which expired in October 2008, with permanent restrictions. In April 2009 the SEC proposed two approaches to short selling:51 —As a marketwide, permanent approach it proposed either a short-selling price test based on the last sale price or tick (uptick rule) or a short-selling price test based on the national best bid (modified uptick rule). —As a security-specific, temporary approach it proposed a ban on short selling if there was a severe decline in the price of that security (circuit breaker rule). In August 2009 the commission reopened the public comment period to receive comments on an alternative uptick rule that would allow short selling only at an increment above the national best bid. (Although the alternative uptick rule was included in the April draft, public comment was not invited at that time.) In February 2010, the commission finalized the rule and applied the alternative uptick rule in combination with the circuit breaker rule. In July 2009 the commission adopted a rule on naked short selling (selling shares short without borrowing on them in advance) that excludes from naked short selling a market participant who fails to deliver by the first day after the 50. Wallison (2008). 51. U.S. Securities and Exchange Commission (2009a).

03-0404-1 chap3.indd 73

7/12/10 6:01 PM

74

kei kodachi and tetsuya kamiyama

failure, regardless of whether it has a long or short position, until that position is closed out. The debate about the costs and benefits of short selling has been going on for a long time. However, we use this opportunity to take a fresh look at the arguments for and against, starting with the basics. The two benefits of short selling are —It makes for a more effective price discovery function by reflecting negative views of either the market or individual securities. —It increases the number of market participants, and therefore liquidity, by enabling investors who have negative views of either the market or individual securities to participate. To argue that short selling should be subject to tighter regulation, even if that impairs such benefits, one would have to demonstrate that short selling was responsible for market failures and that the benefits of regulation outweighed the costs. In our view, however, while the stock market panic of 2008 may have been a case of market failure, it was not caused by short selling. Also we would argue that it makes more sense to tackle market manipulation head on by means of market abuse rules than by restricting short selling.52 We think this is borne out by the following observations on what happened in the United States between July and August 2008, when short selling of financials was prohibited: stocks about which investor opinion was divided became overvalued; and in the absence of short sellers, who are considered to be well informed, spreads widened and market quality suffered as other investors demanded a higher risk premium.53 Furthermore, the following observations have been made about the ban on naked short selling during the same period: the ban failed to stem the fall in share prices and actually reduced market liquidity and pricing efficiency; and not only was naked short selling not responsible for the fall in U.S. share prices in 2008 but prices actually increased after credit downgrades.54 This suggests to us that the newly adopted alternative uptick rule might impose unnecessary restrictions on the market’s price discovery function and liquidity. In fact, the SEC itself has admitted that, although the alternative uptick rule could be applied at less cost and more quickly because there is no need to check the latest price data, it is more restrictive than the modified uptick rule (because it limits short selling to prices above the national best bid) and impairs the benefits of legitimate short selling and could therefore affect market liquidity and the efficiency of its price discovery function. 52. Gruenewald, Wagner, and Weber (2009). 53. Autore, Billingsley, and Kovacs (2009). 54. Fotak, Raman, and Yadav (2009).

03-0404-1 chap3.indd 74

7/12/10 6:01 PM

seven questions on financial reform

75

We also think that it would be inappropriate and unjustifiable to apply a permanent tightening of the rules to dealing with a stock market decline that is presumably temporary (at least, not permanent). However, we do think it is justifiable to apply circuit breaker rules to short selling temporarily, to deal with temporary problems. The United Kingdom’s Financial Services Authority has adopted a similar approach. It takes the view that, although it is unjustifiable to apply permanent and direct restrictions on short selling, it would be justifiable to reimpose temporary restrictions on short selling in an emergency.55 However, in either case it is not clear who would decide and on what grounds. We also have to think about the practical consequences for broker-dealers of short selling (especially uptick) rules. Under the SEC’s new rule, broker-dealers would be required to establish, maintain, and enforce written policies and procedures reasonably designed to prevent the execution or display of any short sale order that did not conform to the rule. As this would require them, at the very least, to monitor national best bids in real time, they would incur the cost of installing and operating the necessary systems.56 This (in addition to the aforementioned impact on the market’s liquidity and the efficiency of its price discovery function) would have to be taken into account in any consideration of whether to tighten the rules on short selling. Some observers point out that, in the case of naked short selling, the security and reliability of settlement are not major problems and that there is no need to have different rules for naked short selling from those for normal short selling.57 These observers point out the following: —Naked short selling helps reduce pricing errors, the volatility of share price returns, bid-ask spreads, and the volatility of pricing errors. —Clearinghouses usually supply the necessary shares in the case of a failure to deliver. —While the failure to deliver continues, the buyer in the short sale receives interest from the seller. However, we accept that, unlike naked CDSs, the security and reliability of settlement are more likely to be endangered in the case of naked short selling than in the case of normal short selling because the securities in question have to be supplied. We therefore agree that some degree of restriction on naked short selling is reasonable. However, in view of the above observations on naked short selling, we wonder whether there might not have been more indirect alternatives 55. U.K. Financial Services Authority (2009a). 56. U.S. Securities and Exchange Commission (2009a). 57. Fotak, Raman, and Yadav (2009); Wallison (2008).

03-0404-1 chap3.indd 75

7/12/10 6:01 PM

76

kei kodachi and tetsuya kamiyama

to a de facto ban on short selling of the kind the SEC has imposed (like financial penalties for failure to deliver).

Do Hedge Funds Need to Be Regulated? Tighter regulation of hedge funds had been on the agenda for some time. However, as a result of the financial crisis and the increased volatility that went with it, the G-20 agreed at the London summit in April 2009 to introduce a registration system for hedge funds. As a result, in July 2009 the U.S. Treasury proposed the Private Fund Investment Advisers Registration Act of 2009.58 By amending the Investment Advisers Act of 1940, the 2009 act introduces the new concept of the private fund, which includes hedge funds and private equity funds. The act proposes that U.S. and nonU.S. private funds that meet certain criteria register as investment advisers. In order to avoid systemic risk, the act proposes that the SEC be given the powers to require private fund investment advisers to keep records and make reports. The SEC would give this information to the Federal Reserve, which is responsible for monitoring systemic risk, and a new body, the Financial Services Oversight Council. This is not the first time that the United States has tried to force hedge funds to register. In December 2004 the SEC promulgated a rule to require hedge funds (but not private equity funds) to register on the grounds that, under the Investment Advisers Act of 1940, “clients” means final investors. However, in June 2006 a federal court overturned this rule on the grounds that the SEC had exceeded its powers and arbitrarily interpreted the wording of a piece of legislation. Similarly, in July 2007 the SEC adopted an antifraud provision under the Investment Advisers Act of 1940. However, its main aim was to protect investors, not to require hedge funds to register or to tighten the regulation or supervision of hedge funds in order to prevent systemic risk. In contrast, the Private Fund Investment Advisers Registration Act of 2009 is a more far-reaching piece of legislation in that it requires not only hedge funds but also private equity funds to register and seeks to tighten regulation and supervision in order to prevent systemic risk. In Europe, especially continental Europe, countries such as France and Germany, which have always been hostile toward private equity firms such as hedge funds, are in the process of adopting even stricter rules than those of the United States. An EU directive of April 2009 proposes a new framework for managers of alternative investment funds.59 This directive adopted the same approach as 58. As of July 2010, the Private Fund Investment Advisors’ Act of 2009 is in the House-Senate conference as part of the Wall Street Reform and Consumer Protection Act. 59. Commission of the European Communities (2009a).

03-0404-1 chap3.indd 76

7/12/10 6:01 PM

seven questions on financial reform

77

the United States and covered not only hedge funds but also any funds, such as private equity funds, that are not undertakings for collective investments in transferable securities. The approach is also similar in that the directive requires fund managers to report to the authorities in order to control systemic risk. However, whereas private equity funds have to be registered under U.S. legislation, alternative investment fund managers have to be authorized under the EU directive. The directive proposes that, in order to be granted authorization, a fund manager has to give the authorities details of matters such as its principal shareholders or members, a program of activity, its funds, and how it will safeguard its customers’ assets.60 It also imposes capital adequacy requirements on alternative investment funds. A fund is required to have €125,000 in regulatory capital plus 0.02 percent of any funds under management in excess of €250 million. However, we cannot help having some doubts about such moves to tighten the regulation of hedge funds, especially under current market conditions. The present financial crisis is generally considered to have resulted from the mistakes of those involved in the securitization process (for example, lax credit checks by U.S. mortgage lenders), not from speculation by hedge funds. While hedge funds did suffer outflows because investors were forced to rebalance their portfolios or found themselves with no more cash to invest, they were not responsible for the crisis. We therefore find it odd that tighter regulation and supervision of hedge funds should be discussed as a possible response to the present financial crisis. In fact, with market conditions the way they are, we think now is not the time to discuss the regulation and supervision of hedge funds. Hedge funds are a major source of liquidity to the markets, reportedly accounting for about 20 percent of turnover in the New York Stock Exchange’s listed stocks and just over 30 percent of turnover in the London Stock Exchange’s listed stocks.61 Trading by hedge funds has also reportedly not only increased market efficiency, improved price determination, and increased investor options but also mitigated the impact of the financial crisis. In other words, while the banks are called on by the authorities to trim their assets in order to comply with their capital adequacy requirements in time of crisis, hedge funds, which are exempt from this requirement, are in a position to supply liquidity when it is most needed.62 Or to put it another way, imposing capital adequacy rules on hedge funds might reduce their trading assets, while imposing curbs on short selling might inhibit their investment activity. Similarly, greater disclosure would force hedge 60. The measure is scheduled for a July 2010 vote by the EU Parliament. 61. Financial Policy Forum (2006). 62. Danielsson, Taylor, and Zigrand (2005).

03-0404-1 chap3.indd 77

7/12/10 6:01 PM

78

kei kodachi and tetsuya kamiyama

funds to show their hand to potential rivals and inevitably impair the effectiveness of their idiosyncratic investment strategies. Therefore, far from ending the financial crisis, tighter regulation of hedge funds would, in our view, be more likely to restrict the activity of the very source of liquidity that might achieve this. Tighter regulation and supervision of hedge funds should therefore be considered in its own right and not as a solution to the financial crisis. Regulation and supervision of hedge funds need to be tightened in some way because there actually was a case, notably that of Long-Term Capital Management, in which a hedge fund brought the global financial system to the brink of disaster. However, in this case the problem was seen as one of excessive leverage, and the regulators should be able to deal with such cases indirectly by tightening their supervision of prime brokers and obtaining information from them about the hedge funds’ activities, including their use of leverage. This could have been done without having to amend any legislation and is the approach already adopted by the BCBS and the International Organization of Securities Commissions in their sound practices recommendations.63 We think it would be unwise to adopt the kind of regulation and supervision that would restrict the investment activity of hedge funds (for example, registration and disclosure requirements). Differences in the way Europe and the United States tighten regulation can also be a problem, because differences in the way small operations such as hedge funds are treated will inevitably lead to regulatory arbitrage. Rules therefore need to be aligned not only between Europe and the United States but also globally. The newly established Financial Stability Board should have a key role to play here.

Conclusion We think it would be unwise to apply indiscriminately and in an unmodified form tighter regulation and supervision of investment banking as a response to misregulation in the advanced capital markets and the mistakes of major financial institutions in emerging Asian countries and the rest of the world. In China the country’s bank regulator has been preparing for the adoption of Basel II. However, the banks to which Basel II would apply seem to have mixed feelings about the prospect. For example, Chen Siqing, vice president of Bank of China, argues that, although Basel II proved itself during the crisis, it is based on the situation in the West and is not the only model for how banks should be managed and supervised.64 If Basel II is introduced in developing countries it 63. Davies and Green (2008). 64. Chen (2009).

03-0404-1 chap3.indd 78

7/12/10 6:01 PM

seven questions on financial reform

79

should be done gradually and in a way geared to their situation. In introducing Basel II, Bank of China faces a number of constraints. These include information technology (such as the need for concentrated risk information processing and the need to have a complete array of operational procedures) and research and development and human resources (such as the need to collect enough data and to develop risk measurement models). China’s financial authorities are also thinking in terms of the realities of China’s banking industry. Luo Ping, director of the China Banking Regulatory Commission’s training department, writes that when China joined the BCBS on March 12, 2009—along with other emerging economies such as Brazil, India, Korea, Mexico, and Russia—its aim was to take an active part in drawing up the international rules on banking regulation and safeguarding the interests of Chinese banks.65 He also mentions the need to represent the interests of emerging economies in committee discussions. Since emerging economies generally have less developed capital markets and their financial systems depend on bank lending, they have to ensure that their banks remain sound while providing the capital their growing economies need. We therefore foresee occasions when emerging economies will object to attempts to apply Basel II uniformly on the basis of G-20 discussions. In this connection we would like to quote from Takafumi Sato, commissioner of Japan’s Financial Services Agency: Just asking for more capital will not cure the disease of capitalism. To avoid a recurrence of crises, we need a paradigm change. A global community adopting a uniform platform is vulnerable to a virus, as we have witnessed during the current financial pandemic. Capital adequacy regulations should be designed to foster diversity in business models, demanding the right level of capital for the business type of the bank in question.66 The business models of financial institutions vary from country to country and those among developing economies that rely heavily on indirect finance are particularly different from those in the West. We think that, while removing barriers to the free movement of people, goods, and capital across global boundaries is all very well, a diversity of business models and markets is necessary for the macroeconomic diversification effect that will help to create a global economy that can withstand crises. From that point of view, it would be better if financial regulation and supervision accorded with each country’s circumstances. 65. Luo (2009). 66. Takafumi Sato, “Tightening Capital Rules Could Increase Risk-Taking,” Financial Times, June 30, 2009.

03-0404-1 chap3.indd 79

7/12/10 6:01 PM

80

kei kodachi and tetsuya kamiyama

References Acharya, Viral V., and Phillip Schnabl. 2009. “How Banks Played the Leverage Game.” In Restoring Financial Stability: How to Repair a Failed System, edited by Viral V. Acharya and Matthew Richardson. Wiley Finance. American Securitization Forum. 2009. “ASF Model RMBS Representations and Warranties.” Autore, Don M., Randall S. Billingsley, and Tunde Kovacs. 2009. “Short Sale Constraints, Dispersion of Opinion, and Market Quality: Evidence from the Short Sale Ban on U.S. Financial Stocks.” Social Science Research Network. Basel Committee on Banking Supervision. 2005. “Trading Book Survey: A Summary of Response.” ———. 2009. “Revisions to the Basel II Market Risk Framework: Final Version.” Brunnermeier, Markus, and others. 2009. The Fundamental Principles of Financial Regulation. Geneva Reports on the World Economy 11. International Center for Monetary and Banking Studies. Chen Siqing. 2009. “A Contemplation on China Banking Sector’s Fulfilling Basel II under the Circumstance of International Financial Crisis.” China Finance, no. 13. Clementi, Gian Luca, and others. 2009. “Rethinking Compensation in Financial Firms.” In Restoring Financial Stability: How to Repair a Failed System, edited by Viral V. Acharya and Matthew Richardson. Wiley Finance. Commission of the European Communities. 2009a. “Proposal for a Directive of the European Parliament and of the Council on Alternative Investment Fund Managers and Amending Directives 2004/39/EC and 2009/ . . . /EC.” April 30. ———. 2009b. Communication from the Commission. “Ensuring Efficient, Safe, and Sound Derivatives Markets.” ———. 2009c. Commission Staff Working Paper. “Consultation Document: Possible Initiatives to Enhance the Resilience of OTC Derivatives Markets.” Danielsson, Jon, Ashley Taylor, and Jean-Pierre Zigrand. 2005. “Highwaymen or Heroes: Should Hedge Funds Be Regulated?” Journal of Financial Stability. October. Davies, Howard, and David Green, 2008. Global Financial Regulation: The Essential Guide. Cambridge, U.K.: Polity Press. European Commission. 2009. “Proposal for a Directive of the European Parliament and of the Council Amending Directives 2006/48/EC and 2006/49/EC as Regards Capital Requirements for the Trading Book and for Resecuritisations and the Supervisory Review of Remuneration Policies.” Financial Accounting Standards Board. 2009. Statement of Financial Accounting Standards 166 (Accounting for Transfers of Financial Assets); Statement of Financial Accounting Standards 167 (Amendments to FASB Interpretation FIN 46[R]). Financial Policy Forum. 2006. “Hedge Funds Again: A Response to Deregulation.” Â�Washington. Financial Stability Forum. 2009. “Report of the Financial Stability Forum on Addressing Procyclicality in the Financial System.” Basel. Fitch Ratings. 2009. “Off-Balance-Sheet Accounting Changes.” June 22. Fotak,Veljko, Vikas Raman, and Pradeep K. Yadav. 2009. ”Naked Short Selling: The Emperor’s New Clothes?” Social Science Research Network.

03-0404-1 chap3.indd 80

7/12/10 6:01 PM

seven questions on financial reform

81

Gorton, Gary, and Andrew Metrick. 2009. “Slapped in the Face by the Invisible Hand: Banking and the Panic of 2007.” Paper prepared for Financial Markets Conference, Federal Reserve Bank of Atlanta. May. Greenlaw, David, and others. 2008. “Leveraged Losses: Lessons from the Mortgage Meltdown.” Report 2. U.S. Monetary Policy Forum. Group of Twenty. 2008. “Declaration on Financial Markets and the World Economy. ———. 2009a. “Declaration on Strengthening the Financial System.” ———. 2009b. “Declaration on Further Steps to Strengthen the Financial System.” Gruenewald, Seraina, Alexander F. Wagner, and Rolf H. Weber. 2009. “Short Selling Regulation after the Financial Crisis: First Principles Revisited.” Social Science Research Network. Helwege, Jean, and others. 2009. “Credit Default Swap Auctions.” Federal Reserve Bank of New York. Hendricks, Darryll, John Kambhu, and Patricia Mosser. 2007. “Systemic Risk and the Financial System.” Economic Policy Review. November. International Monetary Fund. 2008. Global Financial Stability Report. October. Jaffee, Dwight, and others. 2009. “Mortgage Origination and Securitization in the Financial Crisis.” In Restoring Financial Stability: How to Repair a Failed System, edited by Viral V. Acharya and Matthew Richardson. Wiley Finance. Kashyap, Anil K., and others. 2008. “Rethinking Capital Regulation.” National Bureau of Economic Research. Luo Ping. 2009. “Understanding Basel Committee – On China Joining Basel Committee.” China Finance, no. 12. Singh, Manmohan, and James Aitken. 2009. “Deleveraging after Lehman—Evidence from Reduced Rehypothecation.” Working Paper 9/42. International Monetary Fund. Sirri, Erik. 2009. Testimony before the U.S. House Committee on Agriculture. October 15. Taylor, Ashley, and Charles Goodhart. 2007. “Procyclicality and Volatility in the Financial System: The Implementation of Basel II and IAS39.” In Procyclicality of Financial Systems in Asia, edited by Stefan Gerlach and Paul Gruenwald. Macmillan. Taylor, John B. 2009. Testimony before the U.S. House of Representatives, Subcommittee on Domestic Monetary Policy and Technology Committee on Financial Services. July 9. Tobias, Adrian, and Hyun Song Shin. 2009. “The Shadow Banking System: Implications for Financial Regulation.” Staff report. Federal Reserve Bank of New York. U.K. Financial Services Authority. 2009a. “Short Selling.” Discussion Paper 09/1. London. ———. 2009b. “The Turner Review: A Regulatory Response to the Global Banking Crisis.” London. U.S. Department of the Treasury. 2009a. Principles for Reforming the U.S. and International Regulatory Capital Framework for Banking Firms. September. ———. 2009b. “Financial Regulatory Reform, a New Foundation: Rebuilding Financial Supervision and Regulation.” June. U.S. Government Accountability Office. 2009. Testimony before the U.S. House of Representatives, Subcommittee on Capital Markets, Insurance, and Government Sponsored Enterprises. March 5. U.S. Securities and Exchange Commission. 2009a. “Amendments to Regulation SHO.” April 8. ———. 2009b. Litigation Release 2103. SEC v. Jon Paul Rorech, et al. May 5. Wallison, Peter J. 2008. “Everything You Wanted to Know about Credit Default Swaps but Were Never Told.” American Enterprise Institute.

03-0404-1 chap3.indd 81

7/12/10 6:01 PM

03-0404-1 chap3.indd 82

7/12/10 6:01 PM

christopher c. geczy

4

The Future of the Hedge Fund Industry

E

xaminations of the future of hedge funds are daunting tasks, not in the least because understanding hedge funds is still a subject of intense academic, industry practitioner, regulatory, and legislative examination. Even the very definition of a hedge fund defies easy characterization, an unusual irony given that, at least for the foreseeable future, hedge funds stand a substantial chance of remaining fixtures on the global landscape of wealth management. Perhaps the clearest definition of a hedge fund may be that it is a simple compensation contract and little else, systematically speaking. Other useful definitions may reference the fact that hedge funds find safe harbors from the 1940 Investment Company Act, the Securities Act, and other regulations. These safe harbors may facilitate certain investment strategies and trades, including those that use substantial leverage and great degrees of illiquidity. What is clear from both the academic and practitioner literature is that hedge funds do not necessarily hedge. By nature, any examination of the important elements of the hedge fund industry is subjective and incomplete. That said, the analysis offered in this chapter tries for objectivity by depending on research, the historical record, regulations, and a quantitative assessment of risk. I would like to thank Richard Herring, Robert Litan, the Financial Institutions Center at Wharton, and the Brookings Institution for financial support. Kyle Binder, Michelle Zhang, Kevin Burke, Helen Nguyen, Jamie Doran, and Amanda Wagner provided excellent research assistance.

83

04-0404-1 chap4.indd 83

7/12/10 6:05 PM

84

christopher c. geczy

The Wharton Hedge Fund Survey of Pension Consultants A number of well-known surveys examine forecasted industry trends from both the institutional perspective and the net worth perspective. The Casey, Quirk, and Associates/Bank of New York survey, for example, queries institutional investors in hedge funds.1 But these institutional surveys do not necessarily query institutional consultants or focus on them to any great extent. With the purpose of understanding what these providers of investment advice see as important trends and to understand what they perceive as their clients’ views, in the fourth quarter of 2009 and the first quarter of 2010 we conducted a survey of the largest fifty investment consultants for pension funds. We measured assets under advisement, numbers of clients, and their ranking by Thomson Financial and Nelson.2 Of the consultants who responded (30 percent), 20 percent have thirty to a hundred employees and another 20 percent have more than a hundred. Respondents have thousands of clients who invest more than $7 trillion in assets, with most respondents (70 percent) having more than $7 billion under advisement. Type of client ranges from large state, corporate, and Taft-Hartley plans to endowments and foundations. Twenty-two percent of consultants indicate that their clients allocate 10–20 percent of their funds to hedge funds and funds of funds; 33 percent say that their clients allocate 5–10 percent to these funds; and 44 percent record a 5 percent allocation.

Trends Uncovered by the Survey Answers to our survey provide insights into trends among both hedge fund consultants and their clients. The following examines several of these areas. When queried about the most important near-term trends in the hedge fund industry, every respondent mentions increased transparency in hedge fund manager strategies and positions. This notion relates to the more general theme of a power shift to investors from managers, as most respondents also mention decreasing fees, better liquidity terms, and more rigorous operational due diligence as being demanded by their clients and increasingly supplied by them, with more than 50 percent mentioning each of these items. Other trends mentioned include declining leverage, consolidation, and institutionalization, broadly defined.

1. Bank of New York (2009). 2. Thomson Financial (2003).

04-0404-1 chap4.indd 84

7/12/10 6:05 PM

the future of the hedge fund industry

85

More than 90 percent of consultant respondents cite increased regulation as one of the most important issues facing stakeholders in the industry at large. With the possibility of increased regulation imposing greater costs and limitations on investment activities as well as decreased risk of fraud, consultants indicate that whether and how hedge funds fall under increased regulatory oversight or enforcement is a critical element in how they might advise clients in the future. In addition, likely reflecting their and their clients’ experience during the recent financial crisis, many respondents (more than 50 percent) note issues related to performance, such as net asset values that have fallen significantly lower than high-water marks; too much market beta in hedge fund strategies; and alpha/beta separation. Finally, respondents emphasize continued concern over transparency, fees, and operational risk. Given the heterogeneity of trading strategies and potential systematic risk exposures of hedge funds and funds of funds, consultants report that integrating them appropriately into their clients’ portfolios is an important issue, stressing strategy selection as critical to client success. One issue specifically noted by more than 90 percent of respondents is the choice among direct hedge fund investment, fund of funds investment, and hedge fund replicators. Hedge fund replicators generally seek to mimic hedge fund performance by investing in fairly liquid underlying instruments (I discuss replicators more fully below). Consultants indicate that replicators offer greater liquidity and transparency. Further, fees for fund of funds are perceived as prohibitively high. However, respondents also indicate that about 50 percent of clients who invest in hedge funds do so via funds of funds. Also noted, although secondarily (fewer than 55 percent of respondents), is a trend toward strategy-focused funds of funds rather than multistrategy funds of funds. Expanded opportunities are available to clients, as many elite funds once again open for investment, and clients are in a better position to negotiate terms with more transparency. As part of the survey, we asked for arm’s-length views about how investors see areas of concern in the industry generally. More than 50 percent of respondents indicate that their clients identify financial controls, risk management, monitoring, and operational due diligence as critically important, being largely motivated by concern about fraud. In addition, 33 percent of clients, although concerned about exposure to credit risk, are also interested in opportunities to take risks. Pension clients vary in their knowledge and comfort level with hedge funds. Consultants describe most of their pension fund clients (more than 80 percent) as having only limited knowledge, with a small minority (the remaining 20 percent)

04-0404-1 chap4.indd 85

7/12/10 6:05 PM

86

christopher c. geczy

being quite knowledgeable. The percentage of pension clients invested in hedge funds and funds of funds reflects this proportion: most respondents report that less than half of their pension clients are invested in either of these funds. Of pension funds that do invest in such funds, their average allocation is 2–10 percent of their portfolios. Pension plans with assets under management of more than $50 million are more likely than smaller plans to invest in hedge funds and funds of funds. Most consultants expect that their clients will increase this allocation, but not substantially. This is consistent with consultant recommendations: no consultant surveyed recommends or expects a decrease in allocation (88 percent are satisfied with their experience with hedge funds, but the financial crisis has caused pension clients to review and reevaluate their current managers).

Issues Uncovered by the Survey One of the benefits of the survey is the clarity it offers around certain issues: liquidity, risk, fees, fair valuation, regulation, replication, and strategy. The following discussion explores each of these issues. Regarding liquidity: a reasonable initial lockup period is six to twelve months, according to the majority of consultants, and this is in line with client expectations. Most consultants report that their pension clients are not invested in funds with gates, either because they were never raised or because they were lowered. It is clear that consultants prefer to not recommend funds with gate provisions; however, about half of the consultants who responded are willing to recommend them if the client’s liquidity horizon permits the investment and if the manager has a strong track record. The typical redemption period that applies to current clients is sixty days. Regarding risk, consultants use such risk metrics as volatility, market beta, and correlations, including variable rates, drawdown, skewness, and kurtosis. The risks of greatest concern to consultants and pension clients are headline risk, process risk, beta expansion risk, event risk, and credit risk. About fees: although the recent return experience of hedge funds might cause some to speculate on fee compression in the hedge fund industry, survey results do not support this view. The typical management fee charged by hedge funds to pension clients is 1.5–2.0 percent. The typical performance fee is 20 percent, but approximately a third of respondents report a performance fee of 10 percent for hedge funds that are not funds of funds. For funds of funds, median reported fees are a 1 percent management fee and a 10 percent incentive fee. About half of funds of funds are subject to hurdle rates, while all pension clients are invested in funds with high-water marks. Surprisingly, no clients have clawback terms.

04-0404-1 chap4.indd 86

7/12/10 6:05 PM

the future of the hedge fund industry

87

The fair valuation principles outlined in the various iterations of the Statement of Financial Accounting Standards 157 (including the final version, SFAS 157-e) may be harbingers of a new direction for hedge funds. However, most consultants indicate that the principles of SFAS 157 have not altered the reporting of hedge funds and funds of funds. Moreover, the general consensus is that SFAS 157 will have little impact on allocations to private funds, another surprising result given that the volatility of less liquid classes of investments—like distressed debt, measured market exposures, and potentially correlations across standard and alternative benchmarks—is likely to increase. The dominant regulatory issues are possible registration and operational requirements by the Securities and Exchange Commission. Other issues are the trading of over-the-counter derivative contracts and possible restrictions on short selling and speculation in futures markets. In sum, clients desire regulation to make hedge funds more open and transparent, while their consultants fear that regulation may entail increased costs. Although these costs could be passed on to investors in the form of greater fees, they might have the effect of eliminating consultants’ ability to sell certain investments short or to use leverage. Hedge fund replication, which is typically either factor-related replication or trade-related replication, typically invests in liquidly traded underlying assets.3 According to consultants the advantages of replication are low fees, transparency, and liquidity. The most noted disadvantage is (factor) model risk; replicators are described as being backward looking, having high tracking error, and being based on factor models that may themselves be in error. Lack of active management is also mentioned as a disadvantage. Moreover, consultants describe their clients as either unaware of, or neutral toward, replication. This lack of a strong view of replication on clients’ part may reflect one of several characteristics: the recent advent and untested history of replication, the belief that hedge funds provide value added above commonly available investments, or that certain benchmarks (from active management or other sources) are absent in. Client strategies revealed by the survey include primary allocation to credit opportunities, which is not surprising given the performance of credit strategies in 2009, including those concentrated in high-yield debt. Long-short equity, a predominant subclass, is second, above event-driven strategies focusing on distressed opportunities. Neither of these concentrations is surprising, given recent 3. Hedge fund replication traces its roots to academics like Fung and Hsieh (2003); Lo (2005); Kat (2007); Asness, Krail, and Liew (2001), who point out that hedge funds have evinced significant exposures to market indexes like the S&P 500 and other familiar benchmarks as well as other correlates that may not be as common.

04-0404-1 chap4.indd 87

7/12/10 6:05 PM

88

christopher c. geczy

performance. Multistrategy, event driven multistrategy, and recovery-focused strategy align in the same manner. While credit opportunities, long-short equity, and event driven strategies will be recommended in the next twelve months by consultants, the large increase in allocations will be elsewhere. Specifically, fixed-income arbitrage (which at the time the survey was conducted received no “votes” from consultants in that no respondents indicated that their clients allocated at all) was identified as a target strategy: 43 percent of consultants indicate that they would newly recommend it in the next twelve months. In addition, 38 percent of consultants target emerging-market strategies. Likewise, 50 percent indicate that they will recommend so-called exotic strategies.

The Value of Hedge Funds, Now and in the Future One way to understand the various reasons that hedge funds may be beneficial to the overall portfolio is to examine the following four factors. —Vanilla risk premiums: hedge funds provide exposure to systematic risks. —Exotic beta: hedge funds take risks that may be viewed as systematic and capture risk premiums otherwise unavailable to traditional or more constrained investors. —Alpha: hedge fund managers identify mispriced securities or invest with less constraint than managers of nonprivate funds. Transactions costs and fees stay low enough to retain the resulting benefit. —Market, benchmark, timing: these factors define the ability ex ante to alter risk or to benchmark exposures (U.S. equity market beta, say) as the investment opportunity sets changes and premiums rise or fall.

Vanilla Beta Put simply, popular aggregate hedge fund index returns, substrategy returns, fund of funds aggregate returns, and individual hedge fund returns appear to have historically been substantially and statistically significantly correlated with typical benchmarks. These indexes, like the S&P 500 and the Barclays (Lehman) aggregate bond index can be invested in simply and inexpensively.4 Consider for instance the five-year rolling monthly correlations between the S&P 500 and various hedge fund indexes from Credit Suisse.5 The estimated 4. Moreover, aggregate hedge fund indexes appear to have been correlated with popular factors or benchmark portfolios like those of Fama and French (1993); Carhart (1997); Chan and others (2005). 5. These monthly correlations are patterned after Hedge Fund Research (2009).

04-0404-1 chap4.indd 88

7/12/10 6:05 PM

the future of the hedge fund industry

89

correlation between the fund (capital) weighted composite and the S&P 500 index varies over time but is never below 50 percent and averages approximately 65 percent over the sample. Moreover, measured over the five years previous to December 2008, the point estimate of the correlation is 71 percent. Of course, this correlation may arise due to the underlying weightings in equity-oriented funds that do not fully hedge equity risk, and the close tracking of the equity hedge index S&P 500 correlation with the overall fund-weighted composite supports this notion. In any case, though, the clear fact is that the “hedge” in hedge fund is largely illusory. In addition, an aggregate hedge fund index may be thought of as a stratified representation of funds in the hedge fund universe or as a feeless fund of funds in which the underlying funds are ostensibly open for investment at a particular point in time, according to the construction methodology of the index. It is a common benchmark used to track aggregate hedge fund returns. A number of characteristics of a diversified portfolio of hedge funds may be appealing to some investors. First, the overall volatility of the cumulative return shown by the aggregate hedge fund index is clearly less than the volatility of the cumulative return of the S&P 500, although not as low as that of the aggregate bond index. However, the aggregate hedge fund index is best thought of as reflective of an agglomeration of risks, some of which can be easily misestimated when appropriate estimation technologies are not used. The result of such a mischaracterization may be an underassessment of true volatility. In addition, the aggregate hedge fund, as typified by the Credit Suisse index, weathered the equity market downturn at the turn of the last century. However, it is also clear from the figure that the aggregate hedge fund index displays some measure of correlation with equities, a correlation that appears to grow over time. Moreover, the equity market downturn apparent in 2007 is eventually mirrored by the hedge fund index. Nonetheless, a number of points are worth mentioning. First, equities started their drawdown in July 2007 (some hedge fund strategies suffered in 2007 as well, including those trading structured subprime debt), whereas the aggregate hedge fund did not begin its corresponding drawdown until the fall of 2008, about the time that Lehman Brothers filed for bankruptcy and that other financial institutions experienced tremendous distress. In addition, in October 2008 regulatory authorities banned short selling by a large number of financial institutions (about 800 in the United States), trades that anecdotal evidence suggests a significant number of hedge funds would have otherwise made. While subsequent research suggests that the ban did not achieve its intended effect—of bringing order to markets—and in fact appears to have denuded price discovery, the event appears

04-0404-1 chap4.indd 89

7/12/10 6:05 PM

90

christopher c. geczy

Figure 4-1. Asset Class and Hedge Fund Returns, 2000–04 Annualized return (percent) Classical asset classes Hedge funds

20

Macro Convert Emerging market Distressed Equity L/S U.S. mid-cap Event driven 10 Futures Fixed income arb U.S. small-cap LT gov U.S. corp 5 U.S. value stocks U.S. mortgage Non-U.S. bonds Treasury bills MSCI Japan EAFE Ex Japan

15

Equity mkt neutral

0

5

10

–5 –10

15

20 U.S. growth stocks Short sellers

25

Annualized volatility (percent)

to correlate with the steepest portion of the aggregate hedge fund drawdown in the fall of 2008. Finally, the nadir of market return during the period was not echoed in hedge funds, and they have participated in the subsequent rally after the first quarter of 2009. Another useful perspective on the role of hedge funds in investor portfolios highlighting performance lessons of the recent past can be found in figures 4-1, 4-2, and 4-3. Figure 4-1 plots in mean-volatility space a number of long-only asset classes that might be available both to institutional and to retail investors, along with a number of typical hedge fund subindexes from Credit Suisse, with coordinates calculated using monthly data from July 1999 through June 2004. The volatilities in the figures do not make the volatility corrections discussed here either to hedge fund or to standard benchmarks, some of which certainly exhibit meaningful degrees of autocorrelation. Apparent in figure 4-1 are the often-cited high Sharpe ratio characteristics of hedge funds, most subclasses evincing high average returns per unit of measured volatility. It is important to point out that, although some hedge fund subclasses have substantially lower volatility than typical long-only equity indexes on the face of it, this volatility is often mismeasured. Adjustments used here typically result in larger estimates of volatility.6 Nonetheless, during a bull market, a rising tide raises nearly all boats (figure 4-2). That is, when the worldwide equity bull market saw the S&P 500 compounding annually at just under 25 percent, with 6. Geczy, O’Connor, and Proskine (2010).

04-0404-1 chap4.indd 90

7/12/10 6:05 PM

the future of the hedge fund industry

91

Figure 4-2. Asset Class and Hedge Fund Returns, Bull Market, 1998–2000 Annualized return (percent) 25

Classical asset classes Hedge funds

U.S. mid-cap

20 Convert 15 Equity mkt

Emerging market

Macro Distressed Equity L/S Event driven Futures 10 Fixed income arb LT gov U.S. mortgage 5 Non-U.S. bonds Treasury bills U.S. corp neutral

0

MSCI Japan

S&P500

5

10

EAFE Ex Japan U.S. value stocks

15

U.S. small-cap

20

25

–5 10

Annualized volatility (percent)

Figure 4-3. Asset Class and Hedge Fund Returns, Bear Market, 2000–02 Annualized return (percent) U.S. corp LT gov U.S. mortgage Distressed Equity mkt neutral Treasury bills 5 Non-U.S. bonds Event driven Emerging Convert 10 5 –5 market

Short sellers Futures Macro

15

15

U.S. value stocks

–15

20

25 U.S. mid-cap U.S. small-cap

S&P500

–25 –35

Classical asset classes Hedge funds

MSCI Japan

EAFE Ex Japan U.S. growth stocks

Annualized volatility (percent)

a standard deviation estimated to be below 15 percent, hedge funds shared in the increase. While they did not generally have as high a return as equity classes, they did appear to have roughly as great Sharpe ratios. However, the term structures of credit risk and duration embodied in the performance of the classical fixedincome categories were not volatile. The effects of the equity drawdown during the bear market of August 2000 through September 2002, were strong. The various equity classes representing

04-0404-1 chap4.indd 91

7/12/10 6:05 PM

92

christopher c. geczy

exposures across dimensions often associated with informative spreads in equities (for example, value and growth) in fact appear highly clustered in mean-volatility space. That is, if investors sought diversification within equity subclasses simply by naïvely spreading allocations across them, they suffered when a common equity effect drew down. However, during the same bear market in equities, the fixed-income categories, which were so lacking in volatility during the bull market, performed relatively well. Hedge funds remained in between the two, some subclasses surely having suffered due to common equity exposure, while others having little such risk fared well. The subsequent bull and bear markets of March 2003 through May 2007 and the bear market of November 2007 through February 2009 repeat this pattern, especially for the classical long-only asset classes. My analysis suggests that during the substantial equity rally all equity classes did well. The EAFE Ex Japan index annualized return was just under 30 percent, with a volatility of just under 12 percent. U.S. value stocks had annualized returns of approximately 20 percent, with an estimated standard deviation of just under 9 percent. Once again, the term structures of credit risk and duration were flat in the dimension of annualized return across annualized volatility. The hedge fund classes again follow the familiar pattern in that they have estimated Sharpe ratios roughly equivalent to those of the classical equity classes, although both had lower average returns and standard deviations. Moreover, during the substantial bear market of November 2007 though February 2009 equity diversification was dominated by a common and severe equity drawdown, with the clustering of risk and reward apparent for value and growth; for small, mid, and large cap; and for foreign and domestic equity indexes.7 In addition, fixed income indexes outperformed equities, although indexes for nonU.S. bonds, long-term government bonds, and U.S. corporate bonds saw negative or just barely positive annualized returns during the period. Finally, while hedge funds suffered, a fact that is not surprising (given the substantial equity market correlation for the average hedge fund and especially for equity long and short) is that they did not suffer as badly as equities; some categories actually did well. The average hedge fund in the Credit Suisse blue chip index was down approximately 20 percent in 2008.8 In addition, futures-oriented hedge funds, global macro hedge funds, and short sellers reported returns that were significantly better than either other hedge fund categories or equities. Once again, 7. The average annualized return across categories was about –45 percent, with relatively little variability round that average. 8. See www.hedgeindex.com.

04-0404-1 chap4.indd 92

7/12/10 6:05 PM

the future of the hedge fund industry

93

diversification across a dozen classical asset classes really amounts to diversification across two: stocks and bonds. These market-cycle style effects are surely not mysterious or unknown. Hedge funds, for example, simply do not fully hedge equity risk. We would therefore expect them to suffer when equity risk does not pay, ex post. Furthermore, that equity subclasses such as value stocks and growth stocks appear to cluster more tightly in mean standard deviation space during market crises and periods of extreme downside volatility is a fairly well-known characteristic of equities.9 If market or systematic risk varies through time and across market cycles, then unless idiosyncratic risk rises during periods of crisis in a way that offsets increases in market or systematic volatility, we would expect correlations to rise at those times as well. The ex post messages of figures 4-2 and 4-3 are, one, that this fundamental connection to common equity effects may have been hedged or diversified away to some extent via allocation to hedge funds but, two, that it would only have been a partial hedge. That said, even the potential for this kind of “tail event” diversification might seem promising for investors in general and at times of crisis, when diversification is perhaps most needed and scarcest. Factor Exposure It is now well understood that measuring risks of investments that are not highly liquidly traded or marked to market is a challenge. The essential issue is that returns measured for an asset that is not liquidly traded or not well marked might not reflect all relevant contemporaneous information. That is, its returns might be smoothed, or smeared, across time, resulting in volatility, benchmark, or factor exposures, and other risk and performance metrics might be measured too low relative to what truer marks would provide, often inducing autocorrelation in returns and model residuals.10 The result typically is an increase in exposures and a decrease in relative performance measures, like multiple factor alphas.11 Factor Models Regression tables demonstrate that common benchmarks, most of which are available to investors in the form of index funds, exchange-traded funds, futures â•⁄ 9. In fact, as Fama and French (1993), among others, point out, it is the spread between value and growth (net of a common market effect in equities) that may be the most interesting representation of the two, rather than long-only exposures to either one individually. 10. The work of Dimson (1979) and Scholes and Williams (1977) has been applied by Asness, Krail, and Liew (2001); Lo (2005); Woodward (2004); Geczy, O’Connor, and Proskine (2010) to hedge fund and other asset class returns. 11. I apply simplified Scholes and Williams (1977) adjustments below.

04-0404-1 chap4.indd 93

7/12/10 6:05 PM

94

christopher c. geczy

contracts, swaps, and other structures, are correlated with the returns of hedge funds.12 For instance, the aggregate hedge fund portfolio multiple-factor contemporaneous market beta is estimated to be 0.23, with a t statistic of 5.81. Moreover, the aggregate hedge fund portfolio has a Lehman Bond aggregate loading of 0.59, with a t statistic of 3.77. Also, the coefficients on the large minus small (LMS) and of the value minus growth (VMG) portfolios are significant at standard levels at –0.19 and –0.08, respectively. Certain strategies have correlations to other such typical benchmarks or factors. For example, the convertible arbitrage index has a –0.07 loading on the LMS (t statistic of –2.98), with exposures to the credit spread and the term spread of 0.20 (t statistic of 2.26) and –0.20 (t statistic of –1.99), respectively. The event-driven substrategy has a lagged equity market beta of 0.06, with a t statistic of 2.39 and contemporaneous and statistically significant loadings on the bank, LMS, VMG, and term spreads, all of which have appeared in the literature and are reasonably easily traded by practitioners. Both the present and future of the hedge fund industry will have to confront how easily these effects may be packaged in the form of relatively inexpensive, liquid, and transparent investments.

Exotic Beta Recent advances in the characterization of hedge fund strategies via examination of their payoffs and distributional qualities have led academics and practitioners alike to propose replicating benchmarks that appear more exotic than what one might traditionally find in a style or asset allocation analysis.13 These more exotic benchmarks or strategy effects appear to be correlated to hedge fund returns and, in the absence of alpha, may nonetheless represent an important source of value delivered by hedge funds from the perspective of allocating investors. Merger arbitrage historically has generated a payoff that looks like a long call with a positive premium; in addition, hedge funds may take higher-order exposures to volatility and tail risk in commodities, currencies, and other futures contracts and model the exposure via lookback options on the underlying benchmark.14 In addition, hedge fund variation can be explained by buy-write types of strategies, typified by the BXM index, among others.15 12. For the regressions I use, see Chan and others (2005). For other regressions, see Fung and Hsieh (2003). Also see the work of Dimson (1979); Scholes and Williams (1977); Asness, Krail, and Liew (2001); Lo (2002). 13. Fung and Hsieh (2003); Chan and others (2005); Mitchell and Pulvino (2001); and Geczy (2006), among others, discuss such exotic exposures. 14. See Mitchell and Stafford (2000) and Fung and Hsieh (2003), respectively. 15. Jaeger (2005).

04-0404-1 chap4.indd 94

7/12/10 6:05 PM

the future of the hedge fund industry

95

I consider a number of such factors as a way to understand how hedge funds may systematically be exposed to the financial industry via either trading or counterparty activities. In addition, I include the level and the percentage change in the volatility index of the Chicago Board Options Exchange (CBOE VIX) so as to capture implied volatility on the S&P 500 futures contract and, ostensibly, exposure to tradable risk. Moreover, I include a tradable version of the PastorStambaugh liquidity benchmark.16 The results of the work of N. Chan and colleagues are strongly repeated here.17 For example, it appears that the VIX is an important correlate of the returns of a number of strategies, including convertible arbitrage, fixed-income arbitrage, and long-short equity, as is the banking industry. In addition, a number of hedge funds are exposed to the liquidity effects embodied in the tradable liquidity factor. The number of factors (selected by loading p values of less than 5 percent) range from four for managed futures (the adjusted R 2 for which is only 21.4 percent) to thirteen for the event-driven category. The aggregate hedge fund index selects ten factors overall (with an adjusted R 2 of 54.5 percent). If the results of the work of N. Chan and colleagues are extended by nearly five years, a number of important differences in estimated exposures appear. Dropped from the regression due to their exceedingly large p values were the S&P 500 and its square, bank returns at the first lag, a U.S.-dollar trade-weighted currency basket, gold returns, and the aggregate bond index. In general, the number of factors declined: only two factors for managed futures, nine for convertible arbitrage, five for the aggregate hedge fund index, six for the event-driven category, and four for the distressed factor selection count. Fixed-income arbitrage dropped selected regressors as well, including all first- and second-order measures of aggregate U.S. equity exposure. Certain categories experienced very large increases in higher-order exposures to the S&P 500, largely due to the market shocks of 2007 and 2008 as well as changes in mark-to-market rules. These higher-order benchmarks, the square and cube of the S&P 500, are meant to capture tail effects and skewness, which ostensibly changed dramatically during the crisis. Further, a number of correlates appear to have declined in importance. For example, the CBOE VIX was not selected for inclusion in any models. These changes in exposures are likely not to surprise those who understand hedge funds as a set of heterogeneous trades and various exposures to risks, irrespective of any apparent or fleeting alpha, that respond to changes in the investment opportunity set. In that light, the fact that hedge funds are unlikely to be 16. Pastor and Stambaugh (2003). 17. Chan and others (2005).

04-0404-1 chap4.indd 95

7/12/10 6:05 PM

96

christopher c. geczy

compared easily to benchmarks with loadings that remain relatively near constant or range-bound may be seen as part of the advantage offered to investors. And the changes between Chan’s results and mine may arise for the same reason. However, given the tremendous market changes between 2007 and the end of 2008, it is difficult to identify the source uniquely. That said, though, I identify below the importance of an important regulatory change that arose during the period, namely Statement of Financial Accounting Standards 157.

Replication The out-of-sample success of a new breed of investment options known as hedge fund replicators, some of which have even become available in the form of mutual funds and exchange-traded funds, underscores the practical implication of the regression exercises described above, although to be sure what I discuss above also points out certain limitations of replicators. It is my view that replication may open the door to better understanding of the systematic drivers of alternative investment strategies, their roles in investor portfolios, and the implications for asset allocation. Generally, hedge fund tracker proponents count among their advantages that they generally take positions in more liquid securities than the average hedge fund. Indeed a number of tracker products are registered under the 1940 Investment Company Act as mutual funds or other vehicles that have similar liquidity. In addition, some trackers provide high degrees of transparency of their holdings, some going so far as to publish their direct holdings (either to meet regulatory disclosure rules or simply to enhance investor awareness). Another advertised benefit is the elimination of hedge fund manager risk and operational risk, which have been long identified as sources of concern in the hedge fund arena (following the famous Capco study of 2003).18 The Madoff Ponzi scheme, discovered in 2008 amid the flight to liquidity that the markets experienced and the great number of Ponzi schemes reportedly discovered by the SEC subsequently, underscores this potential tracker advantage and no doubt has enhanced the perception of tracker value in the marketplace.19 Also, the ability of trackers to accept smaller minimum investments, and the potential ability for counterparties to more easily understand and track the inherent risk, allow for structured products to be created around them. Finally, most trackers are substantially less expensive than the typical fund of hedge funds, which may ultimately deliver returns net of the aggregate expenses of underlying managers on the order of 5–8 percent, a tremendous hurdle for many investors to 18. Feffer and Christopher (2003). 19. As of September 12, 2009, the SEC (www.sec.gov) reports the discovery and prosecution of about sixty Ponzi or similar schemes.

04-0404-1 chap4.indd 96

7/12/10 6:05 PM

the future of the hedge fund industry

97

surmount in deciding to invest with hedge funds. Proponents of various replication strategies often tout these advantages as evidence of how alternative strategies and exposures may be “democratized” for investors who fall outside of the traditional classes eligible for investment (under safe harbors like Sections 3(c)(1) and 3(c)(7) of the Investment Company Act of 1940). Of course, hedge fund trackers are themselves subject to a number of criticisms, which vary depending on the replication approach taken, but one aspect that is nearly universal is the choice of target to be replicated (say, an aggregate hedge fund index subindex of choice or a fund of funds index). At the heart of the criticism is that, by tracking an aggregate average, one is minimizing tracking error using an uninspiring performer rather than tracking funds substantially above the median. In addition, a tracker that targets a fund-of-funds index may essentially be tracking performance already net of substantial layers of fees. Even if a tracker does track an aggregate hedge fund index, the notion of charging substantially less for performance that is already net of large hedge fund fees is in some sense a contradiction. Perhaps a more interesting target would be the gross performance of an index. That said, proponents may counter that the mere fact that the average hedge fund historically may have added value to otherwise less diversified portfolios— even in the presence of the enormous aggregate fees embedded in fund of funds returns—is by itself indicative of the value of increasing the span of asset and strategies in investor portfolios. In addition, since some trackers have outperformed funds of funds historically—albeit over a fairly short time period—the high fees complaint is in some sense moot, at least for the time being. Replication technologies fall into three categories: factor or style-based replication; distributional replication; and mechanical trading-based replication, including holdings-based replication.20 Factor and style-based technologies attempt to use tradable versions of some of the regressors. For example, the Merrill Lynch factor model hedge fund tracker in July 2009 allocated to six underlying components the following:21 –10.2 percent to the S&P 500 –2.4 percent to the Russell 2000 index 14.4 percent to the MSCI EAFE index

20. Geczy (2007). 21. See http://gmi.ml.com/factormodel. Perhaps this tracker built on the work of Fung and Hsieh (2003) and Hasanhodzic and Lo (2007), who find that hedge fund replication may be quite feasible by taking positions in U.S. dollars (US dollar index), bonds (Lehman Corp. AA index), credit spreads (such as that between the Lehman BAA index and treasuries), equity (the S&P 500), and commodities (such as the Goldman Sachs commodity index).

04-0404-1 chap4.indd 97

7/12/10 6:05 PM

98

christopher c. geczy

24.2 percent to the MSCI emerging markets index 16.7 percent to a USD/EUR spot index 74.0 percent at one-month LIBOR. In 2009 a series of hedge fund tracking exchange-traded funds appeared on the market. For example, IndexIQ launched an exchange-traded fund (ETF) that aims to track a proprietary aggregate hedge fund index by investing directly in other ETFs. In other words, it is itself a kind of fund of funds in that it is an ETF of ETFs. As of the end of July 2009 the top ten holdings in the QAI portfolio were reported to be long positions in emerging markets (24.8 percent in EEM and 8.1 percent in VWO), medium-term Treasuries (16.8 percent in SHY), diversified and short-term bonds (6.6 percent in AGG and 4.5 percent in BSV), a basket of currencies (6 percent in DBV), and high-yield corporate bonds (3.98 percent in HYG). In addition, QAI had short exposures via long allocations to short-selling ETFs, namely, short U.S. small-capitalization companies (via 4.7 percent in TWM, a levered short ETF), and a short position in Europe, Australia, and the Far East (EFU, a levered short ETF tracking the inverse of the MSCI EAFE index). As with the Merrill Lynch factor model, apparently QAI has maintained a substantial positive exposure to emerging markets and medium-term duration and diversified and lower-quality bonds while carrying short positions in U.S. short-term Treasuries and generally U.S. equity. Another example is the ART mutual fund managed by Goldman Sachs. Although the correlations with hedge fund indexes are shockingly high, the weaknesses of this technique are many. Again, the explicit pursuit of tracking the average instead of tracking the outstanding may simply not be appealing, although customized indexes of historically outperforming hedge funds may ameliorate this concern (it is of course important to weigh the impact of survivor and selection biases in this alternative). In addition, the factor or benchmarkbased approaches are in substantial measure based on analysis of historical data. As such, they may not capture real-time changes in exposure important for the tracking of underlying hedge funds. A hybrid approach, discussed below, that mixes tracking with following certain hedge fund manager holdings through their Form 13 filing promises to address some of these concerns. In addition, again, the out-of-sample performance of the replicators is compelling. Other hedge fund replication techniques include the distributional approach.22 In this approach, using technologies from the derivatives portfolio replication literature, the so-called distributional tracker does not directly attempt to mimic

22. This approach is advocated by Kat and Palaro (2005), among others.

04-0404-1 chap4.indd 98

7/12/10 6:05 PM

the future of the hedge fund industry

99

the period-by-period returns of hedge fund aggregate indexes. Rather, it attempts to replicate the (unconditional or conditional) statistical distribution of the hedge fund portfolio of interest. This notion makes strong sense if one views an overall portfolio distribution via the multivariate distribution of the underlying components, in which case the distribution tracker is optimally parameterized to attempt to make overall portfolio outcomes (lower variance, skewness, kurtosis, and so on) distributionally more attractive than without the tracker. This approach essentially tries to carve out interesting, important, or useful features of the distributions of hedge fund returns, as desired by an investor, by trading other instruments, which are often fairly liquid. This can be a very powerful idea from the perspective of an investor who can wait for those characteristics to appear in performance realizations. For example, H. Kat proposes trading futures contracts to replicate desired elements of hedge fund distributions.23 While the actual performance of products pursuing this technique is not easily made publicly available, one criticism that has likely hampered its adoption is easy to identify. Most investors, institutional or otherwise, track performance over time and understand it via the historical record of market and economic events. The distributional replication approach promises the important goal of replication that an investor should ultimately care about (payoff distributions) and in fact allows the emphasis on certain characteristics or constraints (like low betas). Another approach often referred to as passive replication attempts to follow hedge fund strategies mechanically, thereby taking on systematic exposure. For example, one approach to merger arbitrage is simply to take long positions in the targets of all announced takeovers and short positions in acquirers, without the in-depth, fundamental, deal-by-deal analysis that is typical. A variant of this approach considers the actual holdings of hedge fund managers that are large enough to be required to file Form 13F (those with $100 million or more) or Form 13G (those that take greater than 5 percent positions in a given public firm’s equity float). One ostensible advantage of this approach over pure factorbased replication is that it may pick up changes in strategy faster than treatments based only on historical returns. A strong analogy exists between traditional returns-based style analysis and holdings-based style analysis. A number of studies have investigated this notion generally, examining the potential for replicating some portfolio manager performance based on required filings).24 Of course, the same criticisms of holdings-based research apply here 23. Kat (2007). 24. See for example, Frank and others (2004).

04-0404-1 chap4.indd 99

7/12/10 6:05 PM

100

christopher c. geczy

as well: namely that managers can and do obfuscate, or window dress, their reported holdings at the reporting juncture.25 That having been said, a number of such holdings replicators have emerged. For example, AlphaClone examines the holdings of hedge fund managers, including the famous Tiger Cubs, funds whose managers were staked and trained by famed hedge fund investor Julian Robertson.26 AlphaClone produces various lists of stocks held by Tiger Cubs in concentrated positions or in commonly held positions that are rebalanced quarterly. The 2009 NACUBO-Commonfund Study of Endowments, which is closely watched by the industry, tabulates the aggregate allocations and performance over various time frames of just under 850 reporting endowments, ranging in size from less than $25 million to multiple billions of dollars (typically endowments belonging to such universities as Harvard, Yale, Princeton, and the University of Pennsylvania). The baseline asset allocation of small endowments holds 38 percent in domestic (public) equities, 27 percent in public fixed income, 13 percent in international equities, and 13 percent in alternative assets, which include private equity, venture capital, hedge funds, and energy and natural resources (including timberland), and distressed debt (table 4-1). To examine whether, on the face of it, hedge fund trackers might add value hypothetically, I examine the historical record of the performance of this reported “policy” allocation based on the assumption of annual rebalancing (without transactions costs) to these and other asset class weights, using standard indexes to represent their performance. Obviously I would prefer to judge the actual historical performance of small endowments, but no usable record of the historical performance of these endowments is available. I view this record as conservative in the sense that, over the longest historical period of examination reported in the study (ten years), the smallest endowments dramatically underperform the largest (by hundreds of basis points per year, on average). Those large endowments that appear to have outperformed may have done so due to their allocations to hedge funds and other alternatives, although some also suggest that their recent underperformance is directly linked to their most illiquid allocations.27 To gauge at a high level whether a hedge fund tracker might add value to the simple aggregate policy allocations, I sequentially allocate to a replicator, again under the presumption that the allocations are rebalanced back to the indicated 25. See Lakonishok and others (1991). 26. See www.alphaclone.com. 27. See Lerner, Schoar, and Wang (2008) on outperformance; and Geczy, O’Connor, and Proskine (2010) on underperformance.

04-0404-1 chap4.indd 100

7/12/10 6:05 PM

the future of the hedge fund industry

101

Table 4-1.╇ Small Endowment Allocation Weights to Hedge Fund Replicator

Asset class

Index

Baseline small endowment asset allocationa

Small endowment with 15% Small reallocation Small Small endowment to hedge fund endowment endowment with 15% tracker, and with 25% with 40% reallocation substantial reallocation reallocation to hedge reduction in to hedge to hedge fund tracker equities fund tracker fund tracker

Domestic equities

S&P 500

38.0

32.3

26.0

28.5

22.8

Fixed income

Barclay’s Capital U.S. Aggreate

27.0

23.0

27.0

20.3

16.2

International equities

MSCI World

13.0

11.1

13.0

9.8

7.8

Alternative strategiesb

Credit Suisse Hedge Fund

13.0

11.1

13.0

9.8

7.8

Short-term securities/ cash/other

Citigroup 3-month T-bill

9.0

7.7

6.0

6.8

5.4

0.0

15.0

15.0

25.0

40.0

HF Replicator Actual Hedge Fund Replicator/ Hypothetical Hedge Fund Replicator

a. The dollar-weighted asset allocation for small endowments, according to the 2009 NACUBO– Commonfund Study of Endowments. A small endowment is defined as one of less than $25 million. b. Alternative strategies are categorized in the NCSE as follows: private equity (LBOs, mezzanine, M&A funds, and international private equity); marketable alternative strategies (hedge funds, absolute return, market neutral, long/short, 130/30, and event-driven and derivatives); venture capital; private equity real estate (noncampus); energy and natural resources (oil, gas, timber, commodities, and managed futures); and distressed debt.

policy weights. For instance, I allocate 15 percent to a given hedge fund replicator pro rata, with resulting weights shown in table 4-1. I use two replicators, one constructed hypothetically by minimizing a function of tracking error with the Credit Suisse hedge fund index by allocating to a set of exchange-traded funds historically available in January 1999. The second replicator is one that is commercially available. Its actual track record ranges from January 2003.

04-0404-1 chap4.indd 101

7/12/10 6:05 PM

102

christopher c. geczy

Table 4-2.╇ Small Endowment Allocation, Hypothetical Performance Using Hedge Fund Replicator, January 1999–February 2010 Item Hypothetical Hedge Fund replicatora 2009 small endowmentb Small endowment with 15% reallocation to hedge fund tracker Small endowment with 15% reallocation to hedge fund tracker, and substantial reduction in equities Small endowment with 25% reallocation to hedge fund tracker Small endowment with 40% reallocation to hedge fund tracker S&P 500

Return (%)

Standard Downside deviation risk (%) (%)

Beta vs. market

Sharpe ratio

Worst month

Worst 12 months

10.70

8.60

6.98

0.35

0.90

–13.78

–25.73

4.03 5.05

8.33 8.03

6.35 6.19

0.51 0.49

0.13 0.26

–9.47 –10.13

–24.87 –24.52

5.38

7.50

5.80

0.45

0.32

–9.65

–22.98

5.72

7.89

6.13

0.47

0.35

–10.55

–24.25

6.73

7.81

6.13

0.44

0.48

–11.20

–23.91

0.80

15.87

11.99

1.00

–0.14

–16.79

–43.32

a. We use SPY, IVV, BWX, CFT, AGG, DBV, DBC, SHY, TLT, EEM, EFA, and VWO, specifically, to mimic hedge funds. b. The dollar-weighted asset allocation for small endowments, according to the 2009 NACUBO– Commonfund Study of Endowments. A small endowment is defined as one of less than $25 million.

Despite the highly hypothetical nature of the exercise, the results are clear and powerful. Historically, adding either tracker improved risk-adjusted performance in the form of the overall portfolio Sharpe ratio, which increases with increased allocations to the tracker, with one exception. The exception is that the specific, arbitrary, deallocation of equities and a 15 percent replicator weight results in a marginally greater Sharpe ratio than a pro rata reallocation for the actual replicator calculation (table 4-3). Market performance during the January 2003 to February 2010 period driving this result is not surprisingly important for this result. In all cases, though, the allocation results in decreased volatility and increased return relative to the base case. Furthermore, statistical tests of the difference in Sharpe ratios among the cases are presented in tables 4-2 and 4-3 (the simple small endowment policy allocation versus a pro rata 15 percent reallocation to a given hedge fund tracker versus a reduction in equities rather than a pro rata reallocation and so on). The only exception is that the difference

04-0404-1 chap4.indd 102

7/12/10 6:05 PM

the future of the hedge fund industry

103

Table 4-3.╇ Small Endowment Allocation Using Merrill Lynch Factor Model, January 2003–February 2010 Item€

Return (%)

Standard Downside deviation risk Beta vs. (%) (%) market

Sharpe ratio

Worst month

Worst 12 months

Merrill Lynch factor modela

5.98

6.12

4.76

0.35

0.60

–6.39

–16.09

2009 small endowmentb Small endowment with 15% reallocation to hedge fund tracker Small endowment with 15% reallocation to hedge fund tracker, and substantial reduction in equities Small endowment with 25% reallocation to hedge fund tracker Small endowment with 40% reallocation to hedge fund tracker S&P 500

6.15 6.14

8.12 7.72

6.41 6.11

0.55 0.52

0.47 0.49

–9.47 –9.00

–24.87 –23.41

6.27

7.27

5.76

0.48

0.54

–8.53

–21.86

6.13

7.46

5.91

0.50

0.51

–8.68

–22.41

6.11

7.11

5.64

0.47

0.53

–8.21

–20.91

5.30

14.55

11.46

1.00

0.20

–16.79

–43.32

a. The Merrill Lynch Factor Model is commercially available. Its actual track record ranges from January 2003. b. The dollar-weighted asset allocation for small endowments, according to the 2009 NACUBO– Commonfund Study of Endowments. A small endowment is defined as one of less than $25 million.

is not rejected (and the Sharpe ratio point estimate is not larger). The p values for the successive increases in Sharpe ratios are the same in each case except that noted below 5 percent. Replication Success Insofar as simple tracking models are useful in investor portfolios via their abilities ex ante to improve asset allocation as well as outcomes, it is as least as much a statement about the incompleteness of the allocation strategy as it is about the trackers themselves. Simple vanilla beta models, albeit perhaps with fewer short sales constraints and time-varying exposures than many strategic allocations have, can be thought of in the traditional academic context of portfolio spanning. That is, if the inclusion of a hedge fund tracker improves portfolio Sharpe ratios, then in a real sense it must have been the case that the allocation was suboptimal or incomplete to begin with.

04-0404-1 chap4.indd 103

7/12/10 6:05 PM

104

christopher c. geczy

The famous study published in 1986 and the numerous subsequent studies demonstrating essentially the same results have been read by the industry to suggest that strategic asset allocation at the highest level drives most variation in portfolio returns.28 If in the long run hedge fund trackers and the beta components of returns add value, they are likely to do so either because they simply complete the span of asset exposures in ways that investors strategically fail to do or because they tactically take advantage of an investment opportunity set that expands or contracts over time. The notion that they “add alpha” is simply hard to accept, again implying the only possible remaining conclusion: asset allocation is just not what it should be as applied today. The Future of Replication Other perhaps as interesting paths for the future may involve further attempts at replicating illiquid or substantially difficult to access strategy performance (like private, large-scale infrastructure exposure), the inclusion of more exotic, carrytrade-like components, and continued separation of systematic sources of covariation in common hedge fund trades. Even if hedge fund replication does not find its way into institutional or retail portfolios, it represents challenges to the historical approach, laden as it is with fees, illiquidity, and transparency. Given how accessible vanilla replication appears to be, replication technology may likely be useful for hedge funds and especially funds of funds to equitize their asset flows. That is, rather than parking inflows in cash in advance of investment, they may capture benchmark-like returns while finding investments in which to deploy assets. Moreover, simply as benchmarks, hedge fund replicators represent tremendous challenges to the average fund of funds, which historically may not be benchmarked at more than the return on Treasury bills or LIBOR plus some hurdle like 3 percent. Comparison to replicated returns certainly raises the bar.

Alpha Traditionally, hedge funds may have been seen as potential sources of alpha, which for now I will generally describe as excess expected return having adjusted for exposure to systematic risks. However, the estimation of alpha critically depends on getting the risk model used for measuring performance right. While the perhaps surprising out-of-sample success of factor-based hedge fund replicators may indicate that it is potentially possible and in fact useful to capture much time-series variation and perhaps otherwise apparent risk-adjusted outperformance of hedge funds, given the heterogeneity of the strategies followed by hedge 28. Brinson, Hood, and Beebower (1986).

04-0404-1 chap4.indd 104

7/12/10 6:05 PM

the future of the hedge fund industry

105

fund managers, the notion that we may be able to capture all exposures using tradable factors in a parsimonious model seems a stretch.29 In addition, the effects of survivor or selection biases are surely apparent, at least in part, in the examination of individual hedge funds or hedge fund indexes like those described by others.30 Moreover, the fact that some lower-cost hedge fund factor-based replicators appear to have outperformed aggregate hedge fund target indexes suggests, at a minimum, that fees are meaningful deterrents to the production of benchmark-beating performance. Furthermore, W. Fung and colleagues demonstrate that, for a sample of about 1,600 funds over the period 1995–2005, the alphas were explained by more factors than during previous periods.31 These alphas were significant only during a middle subperiod (October 1998 to March 2000), after the liquidity crisis that arose in August 1998 with the Russian debt default and the failure of Long-Term Capital Management to just before the worldwide drawdown in equities near the turn of the century. Alphas ranged as high as 1.90 percent a month for dedicated short sellers and 1.14 percent a month for risk arbitrage funds. Seven of the fifteen hedge fund indexes studied evince statistically significant historical alpha. However, the inclusion of an additional five years sees alphas decline across the board, only two being significant at the usual level. While this decline in aggregate alphas, read in concert with estimated sensitivities, may not portend a trend, it does suggest that future allocations to hedge funds should carefully take into account the true source of value: alpha or risk exposures and corresponding premiums, whether exotic or not.

Market or Factor Timing The evidence on the market timing ability of hedge funds is quite sparse and generally negative, although one study reports some marketing timing ability on the part of “market timing” hedge funds.32 However it must be noted that these funds are not necessarily attempting to time markets in the sense typically understood. Rather than trying to tilt opportunistically toward or away from, say, equity at the expense of fixed-income exposure, these funds often try to time 29. Of course I do not necessarily need to require that a benchmark be traded or connected to a traded instrument. Even in the real world, it may be possible to engineer via financial structuring payoffs that mimic the exotic performance of exotic instruments. 30. Malkiel and Saha (2005), and others, suggest that positive alphas may in fact overstate the historical alpha. 31. Fung and others (2006). 32. For a negative assessment, see Fung, Xu, and Yau (2002); for a positive one, see Chen and Liang (2007).

04-0404-1 chap4.indd 105

7/12/10 6:05 PM

106

christopher c. geczy

the points at which mutual funds strike their net asset value and take advantage of misalignments in valuations. This activity was the subject of much scrutiny a number of years ago when illegal market-timing activities on behalf of market participants were discovered and prosecuted. At least at monthly horizons, hedge funds appear not to have been able in aggregate to time the market. Nonetheless, it is certainly possible that some funds may have. Nothing about the hedge fund experience in the recent crisis—or about the positive results noted above—suggests that one should expect hedge funds to be able to time markets in the traditional sense.

Hedge Funds, Private Equity, and Marking to Market The year 2008 marked a watershed period for many investment strategies, especially those that either had inherent liquidity risk exposures and those that were exposed either to high yield or equity risk. On the regulatory front, Statement of Financial Accounting Standards (SFAS) 157 (“Fair Value Measurements”), issued in 2006, became effective in November 2007 but was not adopted fully until November 2008. SFAS 157 governs the principles of fair valuation and rules about when marking to market of illiquid assets is appropriate. It provides a definition of what fair value means and direction on how fair valuation must be done, and it requires expanded disclosure about fair valuation measurements. For example, it requires certain U.S. entities to classify their assets based on the certainty with which fair values can be calculated. It creates a hierarchy of three asset levels. Level 1 assets are those whose values are observable and reflect quoted prices for identical assets or liabilities in active markets to which the reporting entity has access on a given measurement date. Level 2 assets are valued in ways other than quoted market prices in active markets for identical assets (or liabilities). These can include inputs from inactive markets or from inputs that are then transformed (such as, “yield curves observable at commonly quoted intervals”). Level 3 assets are those that may not be legitimately valued as level 1 or level 2 assets and have values based on the reporting entity’s “own assumptions regarding what market participants would use in valuing the asset or liability.” These assumptions include those made about risk. A number of events of the market crisis of 2007–08 affected the adoption of SFAS 157, including drops in home prices and sales, Bear Stearns filing for bankruptcy, and the failure of Lehman Brothers in September 2008. At the end of that month, in fact, the SEC and the Financial Accounting Standards Board issued a joint statement clarifying the implementation of fair value accounting in

04-0404-1 chap4.indd 106

7/12/10 6:05 PM

the future of the hedge fund industry

107

cases in which markets are inactive or disorderly, explaining that “forced liquidations” are not reflective of fair value or an orderly transaction. Moreover, in April 2009 the FASB released an official update to SFAS 157 easing mark-to-market rules in cases in which markets are unsteady or inactive. Although consultants in the Wharton survey did not identify the mark-tomarket changes of SFAS 157 to be of major concern to industry stakeholders, it is clear that the very act of marking illiquid assets to market may affect hedge funds’ reported returns, especially for strategies that are less liquidly traded. For instance, distressed debt may be classifiable as level 2 assets, in which case adjusted prices from exchanges or inactive dealer markets may be required to be used in place of entity assumptions. And even if such an obligation is classified as level 3, the rule still asserts that “the fair valuation measurement objective remains the same, that is, an exit price from the perspective of a market participant that holds the assets.” The reporting entity cannot ignore reasonably available information. This approach differs from previous allowable treatments, which included the possibility of maintaining an “entry” price from the purchase of the asset and good faith updates on the part of the entity. That is, keeping the asset at book value (and therefore potentially underreporting volatility and asset value changes) was possible. Given the potential importance of SFAS 157 and the importance of illiquidity in the estimation of common risk measures like standard deviation, market, and benchmark betas, I examine the effects of marking to market in the data.33 To do so, I specify the following regression, using dummy variables that indicate the required adoption of SFAS 157, largely in the third quarter of 2008. The regression is essentially a simple extended market model in which the standard excess market return is the regressor (here, the S&P 500). RHF is a given Credit Suisse hedge fund index for alternative hedge fund styles or, in the single case of equity market-neutral index, the Hedge Fund Research Index, HFRX, here included as a third variable measuring equity market neutral.34 RTB is the return on the one-month Treasury bill. DSFAS157 is an indicator variable for the adoption of SFAS 157, specified as the third quarter of 2008. RSP500 is the S&P 500 total return index. And RSP500, t–1 is the S&P 500 lagged one month. Inclusion of the lagged market excess return is intended to account for 33. The factor exposures and correlations are identified by Lo (2005); and more recently, Geczy, O’Connor, and Proskine (2010), who study the implications of illiquidity for broadly construed asset classes that are typically a part of the so-called endowment model. 34. See www.hedgeindex.com.

04-0404-1 chap4.indd 107

7/12/10 6:05 PM

108

christopher c. geczy

the well-known effects of illiquidity on the measurement of market exposures. In addition, the dummy variables in the regression pick up changes in alpha, in market beta, and in lagged market beta pertaining to the time period assumed to reflect full adoption of SFAS 157. I caution that this factor model does not fully characterize hedge funds. Rather, in this regression, I isolate the effects of SFAS 157 relative to U.S. exposure. RHF – RTB = aHF + aHF, 2 DSFAS157 + bHF, 1 (RSP500 – RTB ) + bHF, 2 [DSFAS157(RSP500 – RTB )] + bHF, 3 (RSP500, t–1 – RTB ) + bHF, 4 [DSFAS157(RSP500, t–1 – RTB )] + eHF . The results indicate strong effects for the Boolean dummy variables in a number of strategies and results. First, the alpha shift parameters, aHF, 2 , estimated for nine out of sixteen hedge fund categories, are statistically significant at standard levels, where fewer than one would normally be expected under the null. All shifts in alpha that are statistically significant are negative. In addition, six of sixteen slope shift parameters, bHF, 2 , are significantly positive at standard levels and better, with some indicating an upward shift in the estimated value of market exposure as many as ten standard errors away from zero. For example, for convertible arbitrage, market beta, which was estimated to be essentially zero, was measured after the adoption of SFAS 157 as 0.38 (t statistic 4.20). Fixed-income arbitrage, which also has an estimated contemporaneous beta that is insignificantly different from zero, shifts upward to 0.41, which is over six standard errors from zero. The specific styles for which the effects in question are prevalent are convertible arbitrage, short sellers, equity market neutral, fixed-income arbitrage, global macro, and multistrategy. Each of these categories is expected except for short sellers, who generally, although not always, trade liquid equities. Moreover, distressed bond hedge funds evince little measurable changes in market exposure during the period of the adoption of SFAS 157. Nonetheless, with the exception of equity market neutral, underlying trades in the categories indicated are notoriously prone to the problems of illiquidity and mark to market. However, SFAS 157 enforces valuations closer to market marks, and my results are consistent with SFAS 157 making a difference in the way returns are less smooth than before its adoption. In addition, categories of strategies like managed futures and long/short equity, which generally trade liquid assets, have no such effect. It should be noted that the R 2s for the Credit Suisse market-neutral indexes are surprisingly strong and are entirely a result of the feeder funds (such as Broadmarket Prime) of the fraudulent investment managed and promulgated by Bernard

04-0404-1 chap4.indd 108

7/12/10 6:05 PM

the future of the hedge fund industry

109

Madoff, who was convicted of perpetrating the world’s largest Ponzi scheme.35� In November 2008 the Credit Suisse market neutral was down approximately 40 percent, a return nearly entirely caused by the –100 percent return of the Madoff feeder funds. The market-neutral HFRX is not so affected. It is highly likely that this extreme outlier (approximately eight standard deviations from the average monthly return) drives the high R 2s and beta shifts indicated by the dummy variables. Consistent with this conclusion is that I see no such shift for the HFRX. Finally, there is no identifiable effect for a shift in the lagged effect of the excess market return at t – 1. Again, this result is consistent with the notion that SFAS 157 made a palpable difference in the way otherwise very smoothed out returns would have been reported.

Bernard Madoff, Regulation, and Enforcement One of the most interesting areas highlighted by the Wharton survey is regulation and enforcement. Understanding the interest of regulators, legislators, and investors is not difficult amid the fallout of the crisis and the discovery of the Madoff Ponzi scheme (along with other alleged frauds prosecuted by the SEC). While the details of the Madoff saga have been well covered in the industry press and academic literature, it is useful to point out that so-called operational risk and its assessment was and will continue to be an important element in the hedge fund industry and one that is certain to continue to receive the attention of academic and other stakeholders.36 Moreover, between the approximate time of the Madoff revelation in the fall of 2008 and just under a year hence, the SEC pursued approximately fifty cases of alleged Ponzi schemes or similar fraud. With increased scrutiny of regulators in the wake of the Madoff discovery, and increased enforcement, it is likely that still more frauds and scams will be identified. Regulation of hedge funds beyond more stringent enforcement of current laws and regulations will evolve in one or both of two ways. First, both legislators and regulators will continue to directly address hedge fund regulations. For instance, in the fall of 2009 members of committees in both the U.S. House of Representatives and the U.S. Senate proposed bills that would require hedge fund advisers and some other similar pools to register with the SEC and be subject to increased 35. For Broadwater Prime, see www.hedgeindex.com. 36. For example, Brown, Fraser, and Liang (2009) and Brown and others (2009) suggest that operational risk not only is important but can be quantitatively assessed and used in portfolio construction and risk management.

04-0404-1 chap4.indd 109

7/12/10 6:05 PM

110

christopher c. geczy

reporting and disclosure requirements as well as to record-keeping requirements in line with those that apply to investment advisers that register with the SEC. In early 2009 the European Union proposed a directive on alternative investment fund managers that would require such fund managers to become authorized to market to investors in any member state. Information proposed to be disclosed by managers under the directive includes certain characteristics of the underlying strategies and investments, the identities of certain large investors, and the maximum leverage used by the manager. Current debate on the directive centers on such questions as to whether foreign fund managers would be required to treat the European Union as a single market or whether they would be required to treat it country by country. Second, legislators and regulators would regulate the activities or instruments that hedge funds trade. For instance, in August 2009 the Obama administration proposed the Over-the-Counter Derivatives Markets Act of 2009. This proposed regulation, in some sense a first step toward even full use of central clearinghouses and exchanges, stipulates that certain over-the-counter derivatives need to be cleared by a clearing organization falling under the aegis of the SEC or the Commodity Futures Trading Commission (CFTC). In addition these derivatives contracts (like standardized credit default swaps) would be required to be traded via execution facilities regulated by the SEC or the CFTC. Moreover, nonstandard derivatives would require greater capital and be subject to more restrictions on the definition of an investor eligible to invest in such derivatives. In addition, in the first quarter of 2010 the SEC reenacted a version of the so-called uptick rule restricting short sales in securities that have fallen 10 percent from the previous close. Short sales are banned under the rule unless the current price is above the national best bid and will remain in place until the close of the following business day. Whether and how further regulation and enforcement will play out, the basic point is clear: that there exists a great deal of uncertainty and that’s what was identified in our survey. Consultants responding to the survey identify this further regulation both as an investor protection and as a cost. This point of view is not inconsistent with other regulatory experiments, like the SEC’s ban on short sales of nearly a thousand financial stocks implemented in October 2008. Emergent research suggests that the ban not only was ineffective in preventing market prices from falling but also hindered price discovery and reduced market quality in affected stocks.37 While further regulation is likely, it is not clear that its benefits will outweigh its costs. 37. Boehmer, Jones, and Zhang (2009).

04-0404-1 chap4.indd 110

7/12/10 6:05 PM

the future of the hedge fund industry

111

Summary and Conclusion Some hedge fund trends are more difficult to characterize than those discussed above. One of these is a continued overlap between private equity-oriented strategies like mezzanine and distressed financing, with many hedge funds tilting their strategies toward higher yield fixed-income exposures. This trend is likely to continue into the future, based on our survey and analysis. In addition, since much of the stress in the hedge fund industry—apparent as part of the recent economic crisis—arguably arose from mismatching assets and liabilities, the business model of both hedge funds and funds of funds is likely to match better the liquidities of underlying investments with liquidity at the fund level, a clear area of concern among the responses from the survey of consultants. Certainly institutional investors and their consultants will be incorporating even more due diligence in this important area. Also apparent from our survey is an expression of interest in separately managed accounts as a method for improving transparency, monitoring, and control of managers. It seems likely that this demand will be met with supply, although, given the secrecy under which a large portion of the industry operates, it is not clear that it will adopt separately managed accounts, in which hedge fund managers simply advise pools owned by investors and held in custody outside the aegis of the manager. It is tempting to comment on fee compression for hedge funds. However, anecdotal evidence contrasts with the idea that fees are likely to decline at the underlying hedge fund level. The advent of replication and other more liquid, transparent, and cheaper ways of accessing the type of risk exposures or trades made by some hedge funds may in fact provide the type of benchmarking that results in decreased fund of funds fees. But managers with the best track records by any number of measures will still be likely to demand the stratospheric fees that make the industry the object of desire of many business school students. The industry may in fact target alpha as a metric upon which fees are calculated. But at best a bimodal distribution of fees is likely, because the so-called best funds will continue to convince investors that they provide something unique, even in the face of low-cost, transparent, liquid alternatives.

References Asness, C., R. Krail, and J. Liew. 2001. “Do Hedge Funds Hedge?” Journal of Portfolio Management 28. Bank of New York. 2009. “Mellon and Casey Quirk Analysis.” Boehmer, E., C. Jones, and X. Zhang. 2009. “Shackling Short Sellers: The 2008 Shorting Ban.” Research Paper 34-09. Johnson School.

04-0404-1 chap4.indd 111

7/12/10 6:05 PM

112

christopher c. geczy

Brinson, G., R. Hood, and G. Beebower. 1986. “Determinants of Portfolio Performance.” Financial Analysts Journal 42. Brown, S., T. Fraser, and B. Liang. 2009. “Hedge Fund Due Diligence: A Source of Alpha in a Hedge Fund Portfolio Strategy.” CFA Digest. Brown, S., and others. 2009. “Estimating Operational Risk for Hedge Funds: The w-Score.” Financial Analysts Journal 65. Carhart, M. 1997. “On Persistence in Mutual Fund Performance.” Journal of Finance 52. Chan, N., and others. 2005. “Systemic Risk and Hedge Funds.” NBER Book on Risks of Financial Institutions, Topic: Systemic Risk. Chen, Y., and B. Liang. 2007. “Do Market Timing Hedge Funds Time the Market?” Journal of Financial and Quantitative Analysis 42. Dimson, E. 1979. “Risk Measurement When Shares Are Subject to Infrequent Trading.” Journal of Financial Economics 7. Fama, E., and K. French. 1993. “Common Risk Factors in the Returns on Stocks and Bonds.” Journal of Financial Economics 33. Feffer, S., and K. Christopher. 2003. “Understanding and Mitigating Operational Risk in Hedge Fund Investments: A Capco White Paper.” Frank, M., and others. 2004. “Copycat Funds: Information Disclosure Regulation and the Returns to Active Management in the Mutual Fund Industry.” Journal of Law and Economics 47. Fung, H. G., X. E. Xu, and J. Yau. 2002. “Global Hedge Funds: Risk, Return, and Market Timing.” Financial Analysts Journal 58. Fung, W., and D. Hsieh. 2003. “Asset-Based Style Factors for Hedge Funds.” Financial Analysts Journal 58. Fung, W., and others. 2006. “Hedge Funds: Performance, Risk, and Capital Formation.” Paper prepared for the AFA meetings, 2007, Chicago. Geczy, C. 2006. “Alternative Assets and Exposures: Toward a Definition.” IMCA Monitor. ———. 2007. “Hedge Fund Replication.” Managed Funds Association Reporter. Geczy, C., J. O’Connor, and L. Proskine. 2010. “The Endowment Model of Portfolio Management Revisited: Reassessing the True Risk, Return, and Liquidity Characteristics.” Working Paper. Wharton School. Hasanhodzic, J., and A. Lo. 2007. “Can Hedge Fund Returns Be Replicated? The Linear Case.” Journal of Investment Management 5. Hedge Fund Research. 2009. Annual Summary. Jaeger, L. 2005. “Through the Alpha Smoke Screens: A Guide to Hedge Fund Return Sources.” Euromoney Institutional Investors. Kat, H. 2007. “Alternative Routes to Hedge Fund Return Replication.” Journal of Wealth Management 10. Kat, H., and H. Palaro. 2005. “Who Needs Hedge Funds? A Copula-Based Approach to Hedge Fund Return Replication.” Working Paper 27. London: Alternative Investment Research Centre, Cass Business School. Lakonishok, J., and others. “Window Dressing by Pension Fund Managers.” American Economic Review Papers and Proceedings 81. Lerner, J., A. Schoar, and J. Wang. 2008. “Secrets of the Academy: The Drivers of University Endowment Success.” Journal of Economic Perspectives 22. Lo, A. 2002. “The Statistics of Sharpe Ratios.” Financial Analysts Journal 58.

04-0404-1 chap4.indd 112

7/12/10 6:05 PM

the future of the hedge fund industry

113

———. 2005. “The Dynamics of the Hedge Fund Industry.” Charlotte, N.C.: CFA Institute. Malkiel, B., and A. Saha. 2005. “Hedge Funds: Risk and Return.” Financial Analysts Journal 61. Mitchell, M., and T. Pulvino. 2001. “Characteristics of Risk and Return in Risk Arbitrage.” Journal of Finance 56. Mitchell, M., and E. Stafford. 2000. “Managerial Decisions and Long-Term Stock Price Performance.” Journal of Business 73. Pastor, L., and R. Stambaugh. 2003. “Liquidity Risk and Expected Stock Returns.” Journal of Political Economy 111. Scholes, M., and J. Williams. 1977. “Estimating Betas from Nonsynchronous Data.” Journal of Financial Economics 5. Thomson Financial. 2003. “Pension Fund Consultant Survey.” Woodward, S. 2004. “Measuring Risk and Performance for Private Equity.” Sand Hill Â�Econometrics.

04-0404-1 chap4.indd 113

7/12/10 6:05 PM

04-0404-1 chap4.indd 114

7/12/10 6:05 PM

john e. core wayne r. guay

5

Is There a Case for Regulating Executive Pay in the Financial Services Industry?

S

ince at least as early as the 1950s, the press, the general public, politicians, and academic researchers have remarked on the high levels of pay in the United States for the chief executive officers of corporations. These observers question whether these pay levels are fair and whether incentive structures are appropriate.1 As an example of this view, shortly after his election in 1992, President Bill Clinton signed Internal Revenue Code section 162(m), which limits the tax deductibility of an executive’s compensation to $1 million, with an exception for performance-based compensation. This law promised to rein in “excessive” executive compensation. (Ironically, many argue that this bill may have had the unintended consequence of increasing equity compensation and total compensation.) Similarly, a widely cited book argues that flawed compensation arrangements have been widespread, persistent, and systemic.2 Moreover, it claims that these deficiencies are symptomatic of defects in underlying governance structures, which allow executives to influence their boards of directors. We would like to thank Richard Herring and participants at the 2009 research conference, “After the Crash: The Future of Finance,” sponsored by the Tokyo Club Foundation for Global Studies, Brookings Institution, and Wharton Financial Institutions Center. We thank Rahul Vashishtha for excellent research assistance. We thank Wharton Financial Institutions Center for financial support. 1. See, for example, Murphy (1999). 2. Bebchuk and Fried (2004).

115

05-0404-1 chap5.indd 115

7/12/10 6:06 PM

116

john e. core and wayne r. guay

In 2007 Senator Barack Obama introduced “say on pay” legislation intended to allow shareholders greater influence over executive pay.3

Debate over U.S. Executive Pay This popular resentment of executive compensation appears to stem at least partly from a perception of growing income inequality. Two researchers use the ratio of CEO pay to worker pay as a measure of income inequality, noting, “A comparison of executive pay to the earnings of a typical worker provides insight into the evolution of earnings inequality at the top of the income distribution.”4 They show that CEO pay relative to that of average worker pay increased sharply between 1970 and 2000, rising from about 30:1 to approximately 120:1. Another study provides data on the fraction of taxable income earned by the highest 10 percent of taxpayers.5 These data show an increase in the share of income earned by the top 10 percent of taxpayers, from about 33 percent in the mid-1970s to almost 50 percent in 2006. (Interestingly, this percentage in 2006 was roughly the same as in the late 1920s and early 1930s.) Finally, and more related to our focus on financial industry executives, a third study, which examines the proportion of 2004 taxable income earned by various groups of high-income individuals (specifically individuals in the top 0.1 percent tax bracket), shows that CEOs and other highly compensated executives of nonfinancial companies make up 3.9 percent of this high-income bracket.6 In contrast, investment bankers and fund managers (hedge fund, venture capital, and private equity) constitute 5.2 percent and 4.8 percent of this bracket, respectively. Thus within the United States the financial services industry has a substantial share of very highly compensated individuals. In addition to arguing that executive pay levels are too high, critics of current executive compensation practices also argue that pay for performance is largely absent in executive compensation plans. These arguments appear to rest largely on the following two points: an observation that executives typically receive substantial pay even in years when earnings and stock returns are poor, and a belief that executives are free to unwind their equity incentives at will (that is, by exercising stock options and by selling stock). 3. S. Brush and M. O’Brien, “President Obama’s Say on Pay,” The Hill, June 10, 2009. 4. Frydman and Saks (2007, p. 7). 5. Piketty and Saez (2003). 6. Kaplan and Rauh (2009).

05-0404-1 chap5.indd 116

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 117 The debate over executive compensation, however, is not one sided. For example, two studies examine several of the frequent complaints about executive compensation and argue that many of these concerns are either incorrect or overstated.7 These authors point out that market forces heavily influence executive compensation, that U.S. executives have substantial pay for performance as a result of their holdings of stock and options, and that pay for performance for U.S. executives is typically much larger than for executives in all other countries. Finally, they point out that contracts with executives in many cases anticipate, and try to minimize, costs stemming from managerial power and other agency conflicts. Undoubtedly, executive compensation and incentives will continue to be a hotly debated issue for years to come, and we do not try to settle these disputes in this chapter. Rather, we begin by offering a basic descriptive analysis of CEO pay levels and incentives and a comparative analysis of CEO pay and incentives in the financial services industry. We then describe recent proposals to regulate executive pay in the financial services industry (and more generally) and discuss the merits of such regulation. In summary, although we agree broadly with regulators’ views on the principles that should guide executive compensation practices, we believe that many of these principles are already embodied in the typical executive compensation plan. We also have reservations about whether several of the regulatory proposals would achieve their stated objectives.

Benchmarking U.S. Executive Pay In absolute terms, U.S. CEOs are highly paid. However, many other individuals are also highly paid: hedge fund managers, doctors, lawyers, athletes, musicians, and actors, to name a few. Thus one needs to provide more than just dollar amounts of pay to make a convincing argument against existing pay practices. We observe two general approaches to explore the issue of whether U.S. CEO pay is “too high.” One approach argues that the pay-setting process for CEOs is flawed due to various problems with corporate governance. And if the pay-setting process is flawed, one can readily extrapolate that CEO pay levels and incentives would also be flawed. Another approach is to abstract away from the details of the pay-setting process and to compare CEO pay levels with various benchmarks. We briefly discuss this latter approach in the next section and discuss the former approach later in the chapter, when we address recent regulatory proposals to “fix” executive pay. 7. Core, Guay, and Thomas (2005); Kaplan (2008).

05-0404-1 chap5.indd 117

7/12/10 6:06 PM

118

john e. core and wayne r. guay

To begin, we note that much of the angst about CEO pay is targeted at CEOs of large corporations (CEOs at small corporations are not the ones attracting attention for multimillion-dollar bonuses, large severance packages, and pension plan payouts). While CEO pay at large firms has grown considerably over time, so has the size of the large firms that these CEOs manage. In a study comparing the growth in compensation of S&P 500 CEOs with the growth in the S&P 500 firm index over time, it can be seen that the sharp growth in executive compensation after the 1970s is accompanied by a similar growth in the stock market value of firms.8 Economic theory predicts that larger firms are more difficult to manage and demand more talented CEOs.9 Thus growth in executive compensation may, at least in part, stem from the fact that, not surprisingly, large U.S. corporations are much larger (and arguably more complex) today than they were thirty years ago. The empirical evidence is consistent with this hypothesis—when the average firm is larger, executive pay is higher.10 Another benchmark for the compensation of top U.S. executives might be fees earned by managers who run businesses other than publicly traded corporations, such as managers of hedge funds or private equity firms. In 2005 aggregate total pay received by the top five executives across all exchange-traded firms in the United States was approximately $50 billion. These firms had a total market capitalization of roughly $20 trillion. Therefore, total pay as a proportion of market value is about 0.25 percent, which suggests an “asset management fee” of 0.25 percent for the top five executives. By way of comparison, this percentage is significantly smaller than the 2 percent of assets managed and 20 percent of the profits commonly charged by hedge funds and private equity firms (it is also smaller than fees charged by almost all actively managed mutual funds). To be fair, this comparison is clearly not “apples to apples,” in the sense that hedge fund, private equity, and mutual fund fees may include compensation for more than just five executives, as well as administrative costs. At the same time, one might argue that a multinational corporation is a considerably more complex organization to manage than many of these other types of firms. Regardless, asset management fees provide an interesting comparison to U.S. executive pay. Still another benchmark is CEO pay in countries other than the United States. One study uses the United Kingdom as a benchmark against which to examine whether CEO pay in the United States appears unusually high (the study also compares U.S. CEO pay with CEO pay in other, non-U.K., European â•⁄ 8. Frydman and Saks (2007, figure 5). â•⁄ 9. Gabaix and Landier (2008). 10. Gabaix and Landier (2008).

05-0404-1 chap5.indd 118

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 119 Figure 5-1. U.S. and U.K. CEO Annual Pay and Performance Incentives, 2003 $ thousand United States United Kingdom

2,500 2,000 1,500 1,000 500 0

Annual pay

Performance incentives

Source: Conyon, Core, and Guay (2009).

countries). The authors note that the economies of these two countries “share important governance features, but the United Kingdom is generally considered to be less afflicted by problems of excessive executive compensation.”11 They find that median U.S. CEO pay in 2003 (defined as the sum of salary, bonus, grantdate value of restricted stock and options, and benefits and other compensation) was about 40 percent greater than that for U.K. CEOs (figure 5-1). The authors also note, however, that executives are expected to demand greater pay when their compensation is subjected to greater risk through greater equity performance incentives. They go on to show that U.S. CEOs bear substantially greater equity risk than U.K. CEOs (or equivalently hold much more equity performance incentives), where equity risk is measured as the sensitivity of the value of the CEO’s stock and option portfolio to a 10 percent change in stock price. Specifically, U.S. CEOs bear more than five times the equity incentive risk of U.K. CEOs. Further, the authors show that, after making a risk adjustment to total pay for differences in equity incentives (and for a range of assumptions about CEO wealth and risk aversion), there is no evidence that U.S. CEOs’ riskadjusted pay is significantly greater than that of U.K. CEOs. In summary, we do not expect the debate over excessive executive pay in the United States to end soon. But it is important to recognize the difference between 11. Conyon, Core, and Guay (2009, p. 2).

05-0404-1 chap5.indd 119

7/12/10 6:06 PM

120

john e. core and wayne r. guay

policy influenced by a two-sided social and economic debate and policy based on assertions stemming from only one side of the debate. It appears to us that current regulations are based on arguments from the side of the debate that asserts that U.S. CEO pay is too high and that U.S. CEOs do not have enough performance incentives. However, as we have begun to discuss above and will more clearly illustrate below, these assertions do not appear to be supported by evidence.

U.S. Executive Pay in the Financial Industry Although one might explore many aspects of financial executives’ pay, we begin by asking whether there are differences in the compensation structure for financial industry CEOs vis-à-vis CEOs of nonfinancial firms (as we discuss briefly below, one might also consider whether there should be differences in CEO pay across these industries). We focus our analysis on 2006, the year before the credit crisis began. For this analysis we rely on CEO compensation as reported by Standard and Poor’s Execucomp database. We first identify “banks” as financial firms with SIC codes between 6000 and 6300. We then exclude firms that are mostly concerned with investment advice, pure brokerage business, or wire transferring and that do not match well with the definition of lending institutions.12 The resulting sample of ninety-two firms in 2006 includes both banks and investment banks. We then define nonfinancial firms as those firms not having a one-digit SIC code that begins with six. Each year, we match each financial firm to a similar nonfinancial firm, using a propensity score procedure that controls for beginning-of-year market value, stock return performance, stock return volatility, and CEO tenure. These characteristics have been shown in prior research to explain CEO pay and incentives. By matching firms on determinants of CEO pay and incentives, any observed differences in pay and incentives between financial and nonfinancial firms are expected to be attributable to differing compensation practices across these industries. Figure 5-2 shows median CEO total annual pay over the period 1992–2006 for the banks and the matched sample of nonfinancial firms. Total annual pay is defined as the sum of salary, bonus, grant-date value of restricted stock and options, and benefits and other compensation. The trends for banks and nonfinancial firms are fairly highly correlated, with no persistent difference in pay levels across the two groups (bank CEO pay was somewhat higher in the 1992– 2002 period but somewhat lower in the 2003–06 period).

12. Following Fahlenbrach and Stulz (2009).

05-0404-1 chap5.indd 120

7/12/10 6:06 PM

regulating executive pay in the financial services industry?

121

Figure 5-2. CEO Annual Pay, Banks and Nonfinancial Firms, 1992–2006 $ thousand Nonfinancial 4,000 Banks

3,000 2,000 1,000 0 1992

1994

1996

1998

2000

2002

2004

2006

Source: Authors’ calculations.

Figure 5-3 shows the total annual pay and the equity incentive structure of CEOs of banks and nonfinancial firms in 2006, the year before the credit crisis began. CEO performance incentives are measured as the change in the value of the CEO’s stock and option portfolio for a 10 percent change in the stock price. Risk-taking incentives are measured as the change in the dollar value of the CEO’s option holdings for a 10 percent change in the volatility of the stock price.13 Bank CEOs received about 35 percent less annual pay in 2006 than CEOs of matched nonfinancial firms. Further, the dollar magnitudes of performance and risk-taking incentives of bank CEOs are slightly smaller than those of CEOs in the matched nonfinancial firms. Figure 5-4 replicates this analysis after restricting the sample to the twenty-four largest banks and their corresponding matched nonfinancial firms. In this subsample, the results flip. The CEOs of these large banks received 24 percent greater pay in 2006 but also have greater performance and risk-taking incentives. Thus it may be important to distinguish between exec-

13. CEO stockholdings are not included in the computation of risk-taking incentives. Empirical evidence in Guay (1999) and Parrino and Weisbach (1999) illustrates that risk-taking incentives stemming from stockholdings are typically small except in firms with a high probability of financial distress. Although most of the sample firms examined here were healthy in 2006, clearly many of the banks are now suffering financial difficulty. Given this fact, we return to this issue of risk-taking incentives stemming from CEO stockholdings below.

05-0404-1 chap5.indd 121

7/12/10 6:06 PM

122

john e. core and wayne r. guay

Figure 5-3. CEO Annual Pay and Incentives, Banks and Nonfinancial Firms, 2006 a $ thousand Nonfinancial Banks 4,000 3,000 2,000 1,000 0

Annual pay

Performance incentives

Risk-taking incentives

Source: Authors’ calculations. a. Annual pay is the sum of salary, bonus, grant-date value of restricted stock and options, plus benefits and other compensation. CEO performance incentives are measured as the change in the value of the CEO’s stock and option portfolio for a 10 percent change in the stock price. Risk-taking incentives are measured as the change in the dollar value of the CEO’s option holdings for a 10 percent change in the volatility of the stock price.

utives of smaller banks and executives of large banks when one draws inferences or makes recommendations regarding executive pay. A key feature of the incentive measures in figures 5-3 and 5-4 is their focus on incentives stemming from holdings of stock and options, as opposed to incentives stemming from year-to-year changes in annual pay. It is frequently (but incorrectly) claimed that many top executives have weak performance incentives because their annual pay is not highly sensitive to recent firm performance (that is, that annual pay for many executives is substantial even in years when earnings or stock price performance is poor). It is well documented, however, that the majority of the typical CEO’s performance incentives comes from significant holdings of stock and options, as opposed to changes in annual pay.14 For example, consider the case of large banks (figure 5-4). For the median large bank CEO, the value of his or her stock and option portfolio will change by about $13.4 million with a 10 percent change in the stock price. If the stock price drops 14. Jensen and Murphy (1990); Hall and Liebman (1998); Core, Guay, and Verrecchia (2003); Core, Guay, and Thomas (2005).

05-0404-1 chap5.indd 122

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 123 Figure 5-4. CEO Annual Pay and Incentives, Twenty-Four Largest Banks and Nonfinancial Firms, 2006 a $ thousand Nonfinancial Banks 15,000

10,000

5,000

0

Annual pay

Performance incentives

Risk-taking incentives

Source: Authors’ calculations. a. Annual pay is the sum of salary, bonus, grant-date value of restricted stock and options, plus benefits and other compensation. CEO performance incentives are measured as the change in the value of the CEO’s stock and option portfolio for a 10 percent change in the stock price. Risk-taking incentives are measured as the change in the dollar value of the CEO’s option holdings for a 10 percent change in the volatility of the stock price.

by 20 percent, the value of the CEO’s equity portfolio will fall by about $26.8 million.15 Regardless of whether annual pay declines as well, the CEO has been strongly punished for a declining stock price. We also compare the structure of annual pay for bank CEOs and for nonfinancial firm CEOs. Figures 5-5 and 5-6 show the breakdown of CEO pay in 2006 into cash pay (salary and bonus), equity-based pay (restricted stock and option grants), and other pay for the full bank sample and for the large bank samples, respectively. The figures also present the pay breakdown for the matched nonfinancial firm CEOs. For the full sample of banks, figure 5-5 shows that bank CEOs receive a greater proportion of their pay in the form of cash and that CEOs of nonfinancial firms receive a greater percentage of pay in the form of equity. For the sample of large banks, figure 5-6 shows just the opposite, with 15. The incentives reported in figures 4-3 and 4-4 reflect the change in stock and option portfolio values for a marginal change in stock price and volatility. For discreet changes, such as a 10 or 20 percent change in stock price, the marginal sensitivities are only rough approximations, because the payoffs from stock options are a nonlinear function of the stock price.

05-0404-1 chap5.indd 123

7/12/10 6:06 PM

124

john e. core and wayne r. guay

Figure 5-5. CEO Pay Components, Banks and Nonfinancial Firms, 2006 a $ thousand Bank CEO pay structure Other pay $537

Equity pay $852

Salary and bonus $846

Nonfinancial CEO pay structure Other pay $713

Equity pay $1,674

Salary and bonus $950

Source: Authors’ calculations. a. CEO pay is the sum of salary, bonus, grant-date value of restricted stock and options, plus benefits and other compensation.

large bank CEOs receiving a greater proportion of pay in the form of equity and a somewhat smaller proportion of pay in cash. Equity pay is shown to be the single largest component of CEO compensation for both banks and nonfinancial firms, with the proportion of equity being greatest among large banks. Overall, the analysis suggests that the compensation and incentive structures of bank CEOs are similar to those of CEOs of nonfinancial firms. As a final point, we note that critics of U.S. CEO compensation and incentives frequently argue that equity pay has a limited effect on building CEO incentives because of too short vesting requirements as well as the relative ease with which CEOs can “unwind” their equity incentives through option exercises and stock sales.16 Such arguments, however, are not consistent with the empirical evidence. First, as noted above, U.S. CEOs hold substantially more equity than CEOs in other countries. Second, U.S. CEOs hold substantial amounts of equity well after vesting requirements have lapsed. One study, for example, reveals that in the period 1993–2003 vested stock and options accounted for more than half of all equity holdings by S&P 500 U.S. CEOs.17 Finally, an examination of the possibility that bank CEOs cashed out their equity positions in advance of the recent financial crisis finds no evidence that CEOs attempted to liquidate their equity positions in the period leading up to 16. Bebchuk and Fried (2009). 17. Core, Guay, and Thomas (2005, table 3).

05-0404-1 chap5.indd 124

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 125 Figure 5-6. CEO Pay Components, Twenty-Four Largest Banks and Nonfinancial Firms, 2006 a $ thousand Bank CEO pay structure (24 largest banks) Other pay $3,171

Equity pay $9,271

Salary and bonus $1,019

Nonfinancial CEO pay structure (matched to 24 largest banks) Other pay $2,225

Equity pay $5,984

Salary and bonus $1,442

Source: Authors’ calculations. a. CEO pay is defined as the sum of salary, bonus, grant-date value of restricted stock and options, plus benefits and other compensation.

the credit crisis.18 Instead, on average, CEOs in this sample lost $30 million in stock and option value, with the median CEO loss being more than $5 million.

Regulating U.S. Executive Pay The credit crisis triggered a flood of proposals from various sources to regulate executive compensation, particularly in the financial services industry. In his statement on June 10, 2009, Treasury Secretary Timothy Geithner proposed aligning compensation practices with the interests of shareholders and promoting the stability of firms and the financial system by applying the following five principles: —Compensation plans should measure and reward performance. —Compensation should account for the time horizon of risks. —Compensation practices should be aligned with sound risk management. —Golden parachutes and supplemental retirement packages should be reexamined to ensure alignment of the interests of executives and shareholders. — Transparency and accountability are encouraged in the process of setting compensation. 18. Fahlenbrach and Stulz (2009).

05-0404-1 chap5.indd 125

7/12/10 6:06 PM

126

john e. core and wayne r. guay

Secretary Geithner’s proposed principles are quite straightforward and noncontroversial, so much so in fact that it is plausible that compensation practices already largely conform to them. The principles appear to be consistent with compensation practices that maximize shareholder value, which is the objective stated by nearly every board of every publicly traded corporation in the United States. Further, over the last two decades great strides have been taken by firms and regulators to enhance executive compensation disclosure and transparency as well as the independence of directors sitting on compensation committees. Most of the rhetoric favoring additional regulation of executive compensation does not appear to rely on evidence or careful analysis as to why existing compensation practices are flawed (since there is little evidence that these practices are systemically flawed). Rather it seems to rely on arguing backward from the observation that an unfortunate state of affairs is plaguing the banking industry (that is, when a crisis is observed, one then navigates backward to find a culprit). Then, based on the long-standing, preferred political idea that certain individuals are paid too much or improperly, one uses the argument that flawed compensation practices must have been partly responsible for the financial crisis. This logic is then harnessed to a preferred political change. One proposal to regulate bankers’ pay provides an example of this argument: The best available evidence suggests that the more questioned forms of incentive compensation did not affect financial institutions’ performance during the financial crisis and therefore it is improbable that they were key contributing factors to the global credit crisis. That being said, executive compensation is a perennial media flash point in democratic politics that lends itself easily to political grandstanding, and the current financial crisis is no exception, as it is self-evident that there were egregious instances where financial institutions’ executives and traders did extremely well for themselves while taxpayers have or will be picking up the check. Given an environment in which there is widespread political unease over executive compensation, we advance in this article what we consider to be a superior regulatory approach to that adopted by Congress, and to what existed prior to the barrage of crisis-related compensation legislative and regulatory initiatives.19 In essence, although the authors quoted agree that there is no relation between pay and the financial crisis, because pay bothers a lot of people and politicians, they propose regulating it. We, however, suggest that, if one believes that executive compensation practices are flawed, a natural first step would be to direct 19. Bhagat and Romano (2009, p. 2).

05-0404-1 chap5.indd 126

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 127 attention toward those responsible for the pay-setting process, that is, the board of directors and its compensation committee, rather than to indirectly attack the problem by regulating the outcomes of the board’s decisionmaking process. Further to this point, if one were to determine that the decisionmakers in the compensation-setting process were competent and aligned with shareholders, this would tend to call into question whether, in fact, executive compensation practices are indeed flawed (unless one takes issue with the premise that boards should be aligned with shareholders, which may be reasonable in settings in which taxpayers’ or the general public’s interests conflict with those of shareholders). On June 10, 2009, the U.S. Treasury Department issued an interim final rule on the Troubled Asset Relief Program (TARP), establishing standards for regulating compensation practices at firms receiving government assistance (in which the number of employees subject to the regulations depends on the amount of TARP assistance received). Broadly speaking, the new rules require the following: —Salary payments are not restricted, but salary payments above $500,000 are encouraged to be in the form of vested stock with sales restrictions. —Bonuses are limited to a third of total compensation and must be paid in restricted stock. —Clawback provisions are imposed on any bonus subsequently determined to have been earned based on materially inaccurate financial statements or performance metrics. —Severance and change-in-control payments are prohibited for the most highly compensated executives. —Disclosure on perquisite consumption is increased, and firms are required to adopt a luxury expenditure policy. —Tax gross-ups are prohibited on all forms of compensation. —Firms must permit an annual nonbinding “say on pay” shareholder vote. —Compensation committees must be composed of independent directors. —Risk-taking and earnings management incentives are periodically assessed. —A special master oversees compensation for the top hundred executives at firms receiving “exceptional financial assistance” from the TARP. In June 2009 Kenneth R. Feinberg was appointed as the Treasury Department’s special master for overseeing compensation of the five most senior executives and the next ninety-five highest-paid employees at the seven firms receiving “exceptional financial assistance” under the TARP (AIG, Bank of America, Citigroup, General Motors Co., GMAC Inc., Chrysler Group LLC, and Chrysler Financial).20 20. In December 2009, Bank of America arranged to repurchase all of the $45 billion in preferred shares issued under the TARP program and was thereby released from Feinberg’s proposed compensation arrangements.

05-0404-1 chap5.indd 127

7/12/10 6:06 PM

128

john e. core and wayne r. guay

In October Special Master Feinberg released the proposed Â�compensation arrangements for the five most senior executives and the next twenty highest paid employees of these firms. The arrangements closely followed the Treasury Department’s interim final rules (Feinberg recently continued his charge by proposing compensation structures for the next seventy-five highest-paid employees at these firms). Upon releasing these proposals, Feinberg said, “I’m hoping that the methodology we developed to determine compensation for these individuals might be voluntarily adopted elsewhere,” raising the question of whether and how his proposals might be applied to a broader array of corporations. Our assessment and critique of the Treasury Department’s and Feinberg’s proposals follows.

Compensation Committee Independence The Treasury Department’s interim final rules and Feinberg’s proposed compensation arrangements require the compensation committee that sets executive pay to be composed solely of independent directors. The firms are also required to disclose details related to advice received by compensation consultants. These requirements certainly seem reasonable, but they are in fact already required by stock exchange listing standards and are in place at most corporations. The relatively minor differences between existing requirements and the proposed requirements seem unlikely to have any meaningful significance for most firms.

Restrictions on Salary, Bonus, and Option Payments The TARP restrictions do not appear to place strict limits on overall executive pay (which is an important degree of freedom to maintain if these institutions are to compete in the market for executive talent).21 Rather, the proposal encourages the majority of compensation to be conveyed through salary paid in the form of restricted stock—in other words, stock that is vested at grant date but that cannot be sold for several years. The movement toward stock salary (cash salary is limited to $500,000 or less for most executives) is largely created by the additional restriction on bonus/incentive payments to no more than a third of total pay (or 21. However, TARP participants as well as market analysts appear to view these proposed compensation restrictions as onerous. For example, on December 2, 2009, Bank of America announced it had reached an agreement with the Treasury Department to repurchase all of the $45 billion in preferred shares issued under the TARP program. In response to this announcement, many press reports and analysts indicated that the early repayment of TARP funds was likely, in part, due to a desire to stave off Feinberg’s proposed pay restrictions. For example, a Morgan Stanley Research report states: “We believe TARP repayment is a significant positive for BAC as it eliminates the competitive disadvantage relative to peers who had already repaid TARP in terms of new business generation, talent retention/attraction, and compensation. External CEO candidates may be more likely to talk now that BAC is no longer restricted by the pay czar” (Graseck and Pate, 2009). Emphasis added.

05-0404-1 chap5.indd 128

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 129 50 percent of salary), which also must be paid in restricted stock. Further, stock options are noticeably absent from the list of acceptable forms of compensation. The main thrust of these restrictions appears to be an explicit requirement that all aspects of top executive pay be closely tied to future stock price performance, to promote greater equity performance incentives (or at the least to require payments in stock instead of cash). On October 22, 2009, Feinberg released his proposed compensation arrangements for the seven exceptional assistance TARP firms. The Treasury Department’s press release for those proposals was titled “Reform Pay Practices for Top Executives to Align Compensation with Long-Term Value Creation and Financial Stability,” thus appearing to label the push toward equity compensation as reform. At that time, Feinberg stated, “There is entirely too much reliance on cash and there’s got to be a better way to tie corporate performance to long-term growth.” Further, in his October 22, 2009, letters to each of the exceptional TARP-assisted firms proposing compensation payments, Feinberg repeatedly emphasizes compensation explicitly tied to future corporate performance and discourages (or prohibits) compensation not explicitly performance based. This push toward greater equity incentives, however, is quite puzzling given the evidence presented above that top U.S. executives hold more equity performance incentives than top executives in all other countries. Most top U.S. executives already receive the majority of their annual pay in the form of stock, options, or other performance-based pay that carry substantial multiyear vesting restrictions. Further, most of these executives hold large amounts of restricted stock and options well after those equity grants have become vested. Thus it seems implausible that most U.S. executive compensation arrangements suffer from insufficient equity performance incentives. Or said a different way, it is difficult to see how regulators can so confidently infer that U.S. top executives do not already have sufficient equity performance incentives.22 Further to this point, we illustrate in figure 5-7 how Feinberg’s proposed compensation changes are expected to have a rather small effect in augmenting equity performance incentives for most CEOs in the banking industry. We use 2008 data so we can see the effect of the proposals using the most current data available (although a drawback is that we have 2008 compensation data for only sixtyeight banks). We compare banking CEOs’ equity incentives at the beginning of 22. Extensive theory and empirical evidence emphasizes that executives (or any employee) must be compensated for bearing equity incentive risk. So if more equity ownership is imposed on executives, they must receive higher compensation. If executives are required to hold large amounts of risky equity without compensation, their tendency is to find ways to reduce the risk of their equity—for example, to reject risky projects even if they are good for shareholders.

05-0404-1 chap5.indd 129

7/12/10 6:06 PM

130

john e. core and wayne r. guay

Figure 5-7. Ratio of Existing CEO Equity Incentives to Feinberg’s Proposed Incremental CEO Incentives, 2008 a All banks 80 60 40 20 0

P25

P50

P75

Source: Authors’ calcuations. a. The figure compares banking CEOs’ equity incentives at the beginning of 2008 with the incremental equity incentives that these CEOs would receive during that year if they had followed Feinberg’s recommendations during 2008. For this analysis, we assume that total CEO compensation would be held constant under the Feinberg proposals but that cash pay would be limited to $500,000 a year, with all remaining pay being in the form of stock. To estimate the incremental equity incentives, we compute the difference between the projected equity pay under Feinberg’s proposal and the equity pay that the banking CEOs actually received during 2008. We then compare the CEOs’ existing equity incentives due to their beginning-of-year holdings of stock and options to the incremental incentives stemming from the Feinberg proposal. Beginning incentives are used to compute the percentiles shown.

2008 with the incremental equity incentives that these CEOs would receive during that year if they had followed Feinberg’s recommendations during 2008. For this analysis, we assume that total CEO compensation would be held constant under Feinberg’s proposals but that cash pay would be limited to $500,000, with all remaining pay being in the form of stock. To estimate the incremental equity incentives, we compute the difference between the projected equity pay under Feinberg’s proposal and the equity pay that the banking CEOs actually received during 2008. Finally, we compute the CEOs’ existing equity incentives due to their beginning-of-year holdings of stock and options as a multiple of the incremental incentives stemming from Feinberg’s proposal. Figure 5-7 shows existing incentives as a multiple of incremental incentives at the 25th, 50th, and 75th percentiles for banking CEOs. The median CEO already has twenty-five times the incentives that Feinberg’s proposal would have added in 2008 had it been implemented at these banks. At the 75th percentile, this ratio is

05-0404-1 chap5.indd 130

7/12/10 6:06 PM

regulating executive pay in the financial services industry?

131

greater than 80:1, and even at the 25th percentile, the ratio is 12:1. Further, these ratios are similar across the 25th, 50th, and 75th percentiles if the analysis is alternatively conducted on the sample of large banks examined in figures 5-4 and 5-6. Overall, figure 5-7 suggests that Feinberg’s proposals may have little economically significant effect on overall CEO equity performance incentives. Further, because Feinberg’s proposals do not appear to address restrictions on executives’ ability to rebalance these equity incentives, it is difficult to know how much, if any, incremental equity performance incentives would be imposed by the Feinberg plan (even if one were to consider the effects of the plan after it has been in place for multiple years). For example, consider an executive who holds a large portfolio of vested stock and options. If such an executive’s compensation is tilted toward greater equity pay, it is not hard to imagine that the executive might just sell off a portion of his or her existing equity portfolio to maintain the same equity risk as before the Feinberg plan (which may well be an acceptable response by executives if equity incentives are already at appropriate levels). For an executive who holds no vested stock or options (or holds out-of-the-money options), the additional equity risk imposed by the Feinberg plan might be more difficult for the executive to avoid. The analysis shown in figure 5-7, however, shows that this scenario applies to relatively few CEOs. As a final point, and related to the foregoing, note that in this chapter we focus on CEOs and CEO incentives. We do this under the assumption that, if the CEO has the right incentives, he or she will ensure that his or her subordinates also have the right incentives. This is in fact the way most public corporations operate: the board sets CEO compensation but delegates to the CEO the operating decision of how to compensate his or her subordinates (with approval by the board). In contrast, the Feinberg approach micromanages compensation. It not only sets CEO pay but also sets pay for ninety-nine of the CEO’s subordinates. This approach likely imposes costs on the subordinates and on the firm. While being forced to hold more stock may have little effect on the CEO (as the above analysis suggests), it may have large effects on lower-level employees who do not own much stock and are now forced to hold it. It may not be in the interests of the firm for these employees to own much stock (if it were, one could argue that they would already hold it). It is optimal for the CEO to have a large ownership stake because he or she is typically the employee with the greatest control and impact on firm performance. In contrast, a lower-level executive (a division president, for example) has substantial control over division performance but may have very little impact on or control over firmwide performance. Increasing this manager’s incentives tied to overall firm performance, and likely lowering the relative weight on division performance, quite conceivably will lead the manager

05-0404-1 chap5.indd 131

7/12/10 6:06 PM

132

john e. core and wayne r. guay

to make worse decisions about the division and, as a consequence, lower division and firm profit. An odd feature of Feinberg’s proposals is that he presumes that, even if the CEO’s compensation and incentives are properly structured, further regulatory action is required to ensure that lower-level employees have appropriate compensation and incentives.

Clawback Provisions The Treasury Department’s interim final rules and Feinberg’s proposed compensation arrangements require firms to adopt clawback provisions, which recoup bonuses paid to executives based on financial statements or performance metrics that turn out to be materially inaccurate (these clawback provisions are somewhat more expansive than the provisions under the Sarbanes-Oxley Act of 2002). It seems difficult to oppose the concept of clawback provisions on principle. At the same time, one might question how significant would be the incremental incentives provided by clawback provisions. Public disclosure of materially inaccurate financial statements is a “big deal” for most corporations and attracts negative publicity, regulatory scrutiny, and legal liability. These egregious infractions typically result in CEOs being substantially penalized both monetarily and by the labor market (high probability of being fired, loss of reputation, and large wealth declines through stock price effects on stock and option portfolio). As a result of these existing penalties, it is difficult to envision that clawback provisions will provide enough incremental incentives to top executives to prevent their fraudulent financial reporting (however, clawbacks may be important with respect to lower-level employees, whose manipulation of performance measures is less likely to attract public scrutiny).23

Severance and Change-in-Control Payments The Treasury Department’s interim final rules prohibit severance and change-incontrol payments to the top five executives of all TARP firms during the TARP period. Feinberg’s proposals go further, prohibiting additional accruals into these plans for many lower-level executives. As stated in the Treasury Department’s October 22, 2009, press release, the concern over these payments is that they 23. From a practical perspective, clawbacks may also sometimes be difficult to operationalize. Many performance measures used in executive compensation plans are based on financial accounting numbers and are often tied to multiyear accounting measures. Mechanically, misstated financial accounting performance measures in one period influence financial accounting measures in future periods (overstated earnings in one period, for example, must eventually reverse in a future period). Thus it can be nontrivial to determine the effect of the misstated performance measures on bonus payouts. Further, the compensation setting process is dynamic and is likely to be a complex function of firm performance over time.

05-0404-1 chap5.indd 132

7/12/10 6:06 PM

regulating executive pay in the financial services industry?

133

“often serve to enrich executives rather than provide reasonable compensation during unemployment, and often do not enhance the long-term value of a company.” This criticism, although likely applicable to some extreme cases, such as Disney’s $130 million severance payment to Michael S. Ovitz, seems unlikely to apply more broadly and is also not supported by theoretical or empirical evidence. Severance agreements can provide the insurance needed to attract talented executives who are leaving a successful career at one firm to embark on a new, and potentially risky, career at another firm (although in such cases, a phaseout of the severance agreement over time might be appropriate). Severance agreements can also induce executives to be forthcoming in disclosing bad news and can provide useful risk-taking incentives.24 Further, change-in-control agreements can provide incentives for executives to forfeit their own job in the event that selling the company is beneficial to shareholders. Although the risk-taking incentive benefits of severance agreements may not be appealing to regulators of TARP firms, the benefits related to attracting talented executives and encouraging disclosure of bad news seem particularly important for TARP firms. Given the financial hardship of many TARP firms (as well as the restrictions on compensation plans), attracting high-quality executives to TARP firms is likely to be quite difficult. Timely and transparent disclosure of bad news is also critical for TARP firms to allow regulators and investors the greatest amount of time to take action in the event of a firm’s collapse. Finally, regarding restrictions on change-in-control arrangements, it is hard to see why the Treasury Department would wish to discourage the sale of distressed firms to other, more healthy corporations (to the contrary, in the fall of 2008 the Treasury Department and the IRS promoted the sale of distressed firms by allowing acquirers to more easily utilize the net operating losses of distressed target firms in the banking industry).25

Perquisites and Tax Gross-Ups The Treasury Department’s interim final rules and Feinberg’s proposed compensation arrangements require firms to adopt an “excessive or luxury expenditures policy” as well as to prohibit tax gross-ups on any forms of compensation for employees covered by Feinberg’s proposals. Further, perquisites exceeding $25,000 to any covered employee require extensive disclosure and justification. 24. On severance agreements, see Eisfeldt and Rampini (2004); Inderst and Mueller (2005); Laux (2005); on risk-taking incentives, see Ju, Leland, and Senbet (2002). 25. For example, see J. Drucker, “PNC Stands to Gain from Tax Ruling—Acquisition of National City Will Bring Billions in Deductions, Experts Say,” Wall Street Journal, October 30, 2008.

05-0404-1 chap5.indd 133

7/12/10 6:06 PM

134

john e. core and wayne r. guay

Perquisites (and to some extent, tax gross-ups) are often a lightning rod for negative press coverage and public outrage. High-profile cases of expensive office furnishings, complementary company jet travel, and luxurious conferences have attracted significant negative publicity and public outrage. Although such payments rarely have a significant impact on the company’s bottom-line profitability (and company payment of certain personal expenditures can sometimes be advantageous from a tax perspective), it seems reasonable for regulators to ensure that firms receiving taxpayer assistance do not attract unnecessary public outrage. Further, to the extent that such perquisites and tax payments are an important compensation component in the competitive market for executive labor, paying the executive additional amounts of compensation and then allowing the executive to purchase these perquisites privately would seem to be much more discreet and effective.

Nonbinding Say-on-Pay Shareholder Vote A common proposal put forward by critics of U.S. executive compensation is the requirement that shareholders be allowed to vote annually on executive compensation plans. And although the vote would be nonbinding, the belief (or hope) is that directors would adhere to shareholders’ preferences.26 In 2007 Senator Obama sponsored say-on-pay legislation, but the proposal did not pass. The Treasury Department’s interim final rules and Feinberg’s proposed compensation arrangements require firms to permit a shareholder vote to approve compensation of executives. Although shareholder say-on-pay voting might on the surface appear to be a valuable corporate governance mechanism, it seems difficult to envision a shareholder vote on pay leading to more efficient executive compensation packages (except in particularly egregious cases). There are very good reasons that shareholders allocate most decision rights to a board of directors.27 Making efficient decisions about most complex corporate activities requires considerable expertise, time, and company-specific information. As researchers who have devoted many hours to understanding corporate executive compensation practices, we can personally attest to the time it would take an outsider using public disclosures to 26. Since 2002 shareholders of corporations in the United Kingdom have voted annually on executive compensation packages. The outcome of this voting mechanism has frequently been negative, and although votes are not binding, companies do often adhere to them (recent examples in which a majority of shareholders have voted against management remuneration plans include Royal Dutch Shell PLC, Royal Bank of Scotland Group, Bellway PLC, and Provident Financial PLC). At the same time, there is no evidence that say-on-pay proposals change the level or the growth of CEO pay (Ferri and Maber, 2008). 27. Jensen and Meckling (1976).

05-0404-1 chap5.indd 134

7/12/10 6:06 PM

regulating executive pay in the financial services industry?

135

assess whether or not a specific compensation plan was appropriate for a specific firm at a specific point in time. Corporate directors, on the other hand, not only have expertise that most shareholders do not have but also have a wealth of company-specific information that is used in decisionmaking. In light of these issues, it seems unlikely that most individual, or even institutional, shareholders would take the time to become sufficiently informed to identify deviations between a firm’s existing compensation plan and the optimal compensation plan.

Periodic Assessment of Risk-Taking Incentives The Treasury Department’s interim final rules and Feinberg’s proposed compensation arrangements require firms to review employee compensation plans with “senior risk officers” to ensure that such plans are aligned with sound risk management practices and limit features that might lead senior executive officers to take “unnecessary and excessive risks.” This emphasis on risk-taking incentives seems reasonable given the weak financial health of the exceptional assistance TARP firms, the significant financial leverage of these firms, and the taxpayers’ creditorlike interest in these institutions. It is well known that shareholders can have incentives to take risks at the expense of creditors, particularly when the corporation is financially distressed.28 These risk-taking incentives derive from the limited liability feature of shareholders’ interests, effectively making common stock an option on the value of the firm (that is, limited downside risk and unlimited upside potential). Further, significant leverage, which is common in nearly all financial institutions, can exacerbate these risk-taking incentives. Part or all of TARP assistance came in the form of creditlike capital (either debt or preferred stock), thereby giving taxpayers a vested interest in mitigating the risk-taking incentives of stockholders in these firms. Taxpayers are also the indirect providers of FDIC insurance on certain bank deposits and retirement accounts, and insurance providers obviously have a vested interest in mitigating the risk-taking behavior of the insured. It is important to remember, however, that taking on the right amount of investment and operating risk is essential to successfully compete within any industry and that even creditors want firms to take on some risk. Remember also that senior executives are generally undiversified with respect to firm-Â� specific wealth, thereby requiring boards to carefully consider the appropriate level of both performance incentives and risk-taking incentives. Equity performance incentives expose executives to personal risk and can cause executives to forgo positive net present value investments if those projects are risky; 28. Jensen and Meckling (1976); Myers (1977); Parrino and Weisbach (1999).

05-0404-1 chap5.indd 135

7/12/10 6:06 PM

136

john e. core and wayne r. guay

risk-taking incentives, in the form of stock options or other compensation with convex payoffs, can serve to offset some of this reluctance by executives to take on risk.29 The main point here is simply that too much or too little risk-taking incentives can cause incentive problems, as can too much or too little equity performance incentives. To determine what constitutes “excessive” risk-taking incentives requires one to first determine what constitutes “appropriate” risk-taking incentives. Risktaking incentives generally stem from nonlinearities in compensation payoffs, whereby the sensitivity of payoffs on the down side is lower than the sensitivity of payoffs on the up side. The Treasury Department’s interim final rules and Feinberg’s proposed compensation arrangements offer little in the way of specific details on risk-taking incentives, although one can presumably infer from statements by regulators that their underlying assumption is that executives currently have too much risk-taking incentives and too little performance incentives (we are unaware, however, of any empirical evidence supporting this assumption). The main thrust of Feinberg’s proposals is to eliminate most cash (and other nonperformance-related) pay in favor of stock-based pay. As noted above, depending on the firm’s financial situation, stock-based pay can either increase or decrease risk-taking incentives. In general, greater stock-based pay can potentially either mitigate or exacerbate any existing incentive alignment problems, depending on whether the executive had the right amount, too much, or too little equity incentives to begin with. Given that U.S. senior executives already hold substantial amounts of equity and receive most of their pay in equity, and given that it is difficult to argue that U.S. executives do not have enough performance incentives, it seems unlikely that requiring greater equity holdings will improve incentives and shareholder outcomes. If requiring greater equity holdings will have any substantial incremental incentive effects (which is questionable given the evidence in figure 5-7), the most likely outcome is that executives will be encouraged to take on less risk (potentially to the detriment of equity investors). We also note that Feinberg’s proposals do not include stock options in any of the compensation plans. This is presumably due to a concern that the nonlinear payoffs of stock options would promote “excessive” risk-taking incentives. Such a concern may or may not have merit, however, depending on how the option grants and other elements of the compensation plan are structured.30 Some of the other elements of the Treasury Department’s and Feinberg’s proposals, such as 29. Coles, Daniel, and Naveen (2006). 30. See Lambert, Larcker, and Verrecchia (1991) for arguments that option holdings can frequently result in lower risk-taking incentives.

05-0404-1 chap5.indd 136

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 137 clawbacks, say-on-pay proposals, tax gross-ups, and perquisites, appear to have little to do with risk-taking incentives. Rather, these elements appear to be aimed at limiting or preventing negative or embarrassing publicity about executive compensation at TARP firms as well as possibly encouraging proactive changes to corporate governance (or at least creating such a perception).

Better Disclosure of Executive Incentives One difficulty in determining the merits of the proposals by both the Treasury Department and Special Master Feinberg is the lack of transparent proxy statement disclosure regarding the performance and risk-taking incentives of senior executives. In recent years, the Securities and Exchange Commission expended considerable effort in augmenting required disclosure of compensation arrangements. This disclosure, however, largely emphasizes the level of executive pay rather than the incentives embedded in executive pay and equity holdings. For example, existing disclosures do not allow one to easily determine the overall sensitivity of executive equity holdings to firm performance. So although firms are required to disclose the general form of bonus and equity incentive plans, rough details about performance measures included in those plans, and data on stock and option holdings, it would likely be informative to many shareholders if the firm also provided quantitative details (possibly in tabular form) of the overall sensitivity of the executives’ wealth to various performance measures (such as stock price and earnings). These details could also include quantitative disclosure about the extent of nonlinearities in executive compensation plans and equity holdings so that the investing public could better understand executives’ risk-taking incentives. (As a template for such disclosures, note that many firms currently provide similar tabular disclosures for the risks underlying financial instruments.)

Conclusion In this chapter we review and critique recent regulation of TARP recipients proposed by Treasury Secretary Geithner and implemented by Special Master Feinberg. Before doing this we review the long-standing debate over executive pay and incentives and the concern among some that U.S. CEO pay is excessive and that U.S. CEOs do not have enough performance incentives. We explain how U.S. CEOs do, in fact, have substantial equity performance incentives (and much more than CEOs in other countries, such as the United Kingdom). We also explain why greater performance incentives are expected to give rise to greater pay. Further, as one study shows, after making a risk adjustment to total pay for

05-0404-1 chap5.indd 137

7/12/10 6:06 PM

138

john e. core and wayne r. guay

equity incentives, there is no clear evidence that U.S. CEOs receive significantly more risk-adjusted pay than U.K. CEOs.31 It appears to us that most of the recent regulations and proposals regarding executive compensation proposed by Treasury Secretary Geithner and implemented by Special Master Feinberg are based on arguments from only the side of the debate that asserts that U.S. pay is too high and that U.S. performance incentives are too low. As we note above, these arguments have little empirical support. Geithner’s proposed principles to guide executive compensation seem noncontroversial, in the sense that it is difficult to argue with the principles as stated and also that many or most firms already appear to design compensation practices to conform to such principles. Of more concern, however, are the rules for implementing the principles, as proposed by Feinberg. Some of the requirements are reasonable, in that most firms already follow them or in that it is sensible—or not costly—to adopt them. These requirements include compensation committee independence, clawbacks, reduced perquisites, and tax gross-ups. More concerning are restrictions on severance and change-in-control payments and on the level and composition of executive compensation and incentives. Evidence suggests that the typical firm uses severance and change-in-control payments efficiently; removing them from the risky and financially unhealthy TARP firms seems only to make it more difficult for these firms to attract and retain good talent. The same is true for restrictions on executive pay and incentives. U.S. executives hold more equity performance incentives than executives in any other country, so it is difficult to argue that they do not have enough incentives. Further, and particularly in light of these incentives, there is no evidence that U.S. executives are systematically overpaid. Finally, although it is reasonable for regulators to be concerned about risk-taking incentives for executives, there is often confusion over what constitutes risk-taking incentives as well as a failure to recognize that some risk taking is necessary to compete effectively within any industry. Although the language of regulators nearly always states that there is a dual goal (better aligning compensation practices with the interests of shareholders and promoting the financial stability of firms; see the statements by Geithner above), how regulators prefer to implement these goals can differ from shareholders’ preferences. And it may be that the final implementation emphasizes the stability and low risk taking favored by government claimholders over the value creation favored by stockholders. 31. Conyon, Core, and Guay (2009).

05-0404-1 chap5.indd 138

7/12/10 6:06 PM

regulating executive pay in the financial services industry? 139

References Bebchuk, L., and J. Fried. 2004. Pay without Performance: The Unfulfilled Promise of Executive Compensation. Harvard University Press. ———. 2009. “Paying for Long-Term Performance.” Working Paper. Harvard Law School. Bhagat, S., and R. Romano. 2009. “Reforming Executive Compensation: Simplicity, Transparency, and Committing to the Long-Term.” Working Paper. Yale University. Coles, J., N. Daniel, and L. Naveen. 2006. “Managerial Incentives and Risk-Taking.” Journal of Financial Economics 79. Conyon, M., J. Core, and W. Guay. 2009. “Are US CEOs Paid More than UK CEOs? Inferences from Risk-Adjusted Pay.” Working Paper. University of Pennsylvania. Core, J., W. Guay, and R. Thomas. 2005. “Is CEO Compensation Inefficient Pay without Performance?” University of Michigan Law Review 103. Core, J., W. Guay, and R. Verrecchia. 2003. “Price versus Nonprice Performance Measures in Optimal CEO Compensation Contracts.” Accounting Review 78. Drucker, J. 2008. “PNC Stands to Gain From Tax Ruling—Acquisition of National City Will Bring Billions in Deductions, Experts Say.” Wall Street Journal, October 30. Eisfeldt, A., and A. Rampini. 2004. “Letting Go: Managerial Incentives and the Reallocation of Capital.” Working Paper. Northwestern University. Fahlenbrach, R., and R. Stulz. 2009. “Bank CEO Incentives and the Credit Crisis.” Working Paper. OSU Fisher College of Business. Ferri, F. and D. Maber. 2008. “Say on Pay Vote and CEO Compensation: Evidence from the UK.” Working Paper. Harvard University. Frydman, C., and R. Saks. 2007. “Executive Compensation: A New View from a Long-Term Perspective, 1936–2005.” Technical Report 2007-35. Federal Reserve Board. Gabaix, X., and A. Landier. 2008. “Why Has CEO Pay Increased So Much?” Quarterly Journal of Economics 123. Graseck, B., and C. Pate. 2009. “Bank of America Quick Comment: Catalyst #1 Comes through; TARP Repayment a Big Positive.” Morgan Stanley Research North America. Guay, W., 1999. “The Sensitivity of CEO Wealth-to-Equity Risk: An Analysis of the Magnitude and Determinants.” Journal of Financial Economics 53. Hall, B., and L. Liebman. 1998. “Are CEOs Really Paid Like Bureaucrats?” Quarterly Journal of Economics 113. Inderst, R., and H. Mueller. 2005. “Keeping the Board in the Dark: CEO Compensation and Entrenchment.” Working Paper. New York University. Jensen, M., and W. Meckling. 1976. “Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure.” Journal of Financial Economics 3. Jensen, M., and K. Murphy. 1990. “Performance Pay and Top-Management Incentives.” Journal of Political Economy 98. Ju, N., H. Leland, and L. Senbet. 2002. “Options, Option Repricing, and Severance Packages in Managerial Compensation: Their Effects on Corporate Risk.” Working Paper. University of Maryland/University of California at Berkeley. Kaplan, S. 2008. “Are U.S. CEOs Overpaid?” Academy of Management Perspectives 22. Kaplan, S., and J. Rauh. 2009. “Wall Street and Main Street: What Contributes to the Rise in

05-0404-1 chap5.indd 139

7/12/10 6:06 PM

140

john e. core and wayne r. guay

the Highest Incomes?” Review of Financial Studies Advance Access (http://rfs.oxfordjournals. org/cgi/content/abstract/hhp006). Lambert, R., D. Larcker, and R. Verrecchia, 1991. “Portfolio Considerations in Valuing Executive Compensation.” Journal of Accounting Research 29. Laux, V. 2005. “Board Independence and CEO Turnover.” Working Paper. Frankfurt: Goethe University. Myers, S. 1977. “Determinants of Corporate Borrowing.” Journal of Financial Economics 5. Murphy, K. 1999. “Executive Compensation.” In Handbook of Labor Economics, edited by O. Ashenfelter and D. Card. Vol. 3. Amsterdam: North-Holland. Parrino, R., and M. Weisbach. 1999. “Measuring Investment Distortions Arising from Stockholder-Bondholder Conflicts.” Journal of Financial Economics 53. Piketty, T., and E. Saez. 2003. “Income Inequality in the United States, 1913–1998.” Quarterly Journal of Economics 118.

05-0404-1 chap5.indd 140

7/12/10 6:06 PM

Contributors

John E. Core Wharton School, University of Pennsylvania

Tetsuya Kamiyama Nomura Institute of Capital Markets Research

Yasuyuki Fuchita Nomura Institute of Capital Markets Research

Kei Kodachi Nomura Institute of Capital Markets Research

Christopher C. Geczy Wharton School, University of Pennsylvania

Robert E. Litan Kauffman Foundation and Brookings Institution

Wayne R. Guay Wharton School, University of Pennsylvania

Alan McIntyre Oliver Wyman Group

Richard J. Herring Wharton School, University of Pennsylvania

Michael Zeltkevic Oliver Wyman Group

141

06-0404-1 bm.indd 141

7/12/10 6:09 PM

06-0404-1 bm.indd 142

7/12/10 6:09 PM

Index

ABCP. See Asset-backed commercial paper Accounting: accounting standards, 68; fair value accounting, 59, 106–07; international reporting standards, 65–66; off-balance-sheet entities, 68 After the Credit Crash: The Future of Finance (conference; 2009), vii, 1–2 AIG, 2, 70, 127 AlphaClone, 100 Arbitrage: convertible arbitrage, 94, 95; fixed-income arbitrage, 88, 95; merger arbitrage, 94, 99; regulatory arbitrage, 4, 64, 65–66, 78 Asia, 67 Asset-backed commercial paper (ABCP), 56 Assets: asset levels, 106; bank trading-book holdings, 64–65; fair-market accounting and, 59; fair value of, 106; long-term assets, 16, 25, 56, 60, 66; marking to market of, 106; price effects and margin spirals, 58–59; risk-weighted assets, 65; strategic asset allocation, 104; value of, 57. See also Banks and banking; Hedge funds; Statement of Financial Accounting Standards Auction settlement, 72

Bank of America, 2, 4, 40, 42, 127 Bank of China, 78 Banks and banking: assets of, 55–56, 57, 58, 59, 64, 65, 66; bailouts and, 12; bank branches, 21, 26, 47, 48; bank failures, 49; bank holding companies, 4, 27; banking crisis of 1930s, 14; borrowing by, 57, 66; business models of, 14–15, 27, 28, 47–48, 49, 79; capital levels and, 30, 32, 43, 44, 47, 54–61; cash, checks, and debit cards, 21, 22, 23, 31; CEO compensation in, 120–25; changing environment for U.S. commercial banking, 29–51; competition, 32–33; cost efficiency and cost control, 47–49; definition of, 120; deposits, 18, 20–21, 26–27, 33, 44–47, 66–67; deterioration of performance of, 2–3, 11; in emerging market economies, 67; equity levels, 15; financial crisis of 2004–09, 24–27, 28, 54, 56, 61–62; funding of bank positions, 56, 66; future scenarios of, 3, 11, 12–13, 28–51; “gap” investments and earnings, 18, 24, 42; golden era of (1993–2003), 2, 11, 12, 16–24, 30, 42, 45, 48; growth rates of, 26–27; historical

143

06-0404-1 bm.indd 143

7/12/10 6:09 PM

144 performance of, 13–27; human capital in, 49–50; institutional positioning of, 39–42; investment banks and banking, 3–4; management execution by, 42–51; mergers and acquisitions, 21, 22, 49; purchase of Treasuries, 30; recession of 1981 and, 15; regulation of, 12, 13, 15, 18, 21, 28, 30–32, 33, 35, 37, 40; regulatory arbitrage by, 64; revenues of, 13, 15, 19, 40, 41; risks of, 42–44; S&Ls/ thrifts (savings and loan banks), 16, 17, 27, 28, 49; shadow banking sector, 12, 27, 30, 32; targeted asset growth, 43–44; value at risk, 62–64. See also Economic issues; Loans and lending; Nonbanks; Securities and securitization; individual banks by name Banks and banking—fees: commercial banking revenues and, 13, 25–26, 40; deposit fees, 23–24, 30–31, 33, 42; financial crisis and, 11–12; free checking and, 21; during the golden era, 18, 19, 23; penalty fees, 23–24; role in banking industry business model, 14–15, 16; transaction account fees, 45 Banks and banking—profitability: analysis of scenarios, 33–37, 38–39; during the financial crisis, 2004–09, 27; future profitability, 28; during the golden era 1993–2003, 16–18; during the period 1980–92, 15–16; range of industry profitability, 13; rate-sensitive funding and, 42; regulation and, 12, 30, 32; service charges/fees on deposit accounts and, 23–24 Banks and banking—recommendations for: compensation and human capital, 50; cost controls, 49; country-specific leverage, capital requirements, 4; general description of, 4, 78; higher capital charges, 4; higher capital standards, 4, 60–61; innovation, 50–51; leverage quality, 66–67; limitation of kinds of assets, 4; management of deposits business, 46–47; mergers and acquisitions, 49; risk management, 4, 42,

06-0404-1 bm.indd 144

index 62–64; targeted asset growth, 44; tradingbook assets, 64 Barclays (Lehman) aggregate bond index, 88 Basel I, 61 Basel II: capital framework, 4; in China, 78–79; framework for market risk, 61–62; regulatory arbitrage and, 65; Basel III funding rules, 32, 46 Basel Accord of 1988, 61 Basel Committee on Banking Supervision (BCBS), 61, 62, 64, 78, 79 Bear markets, 91–92 Bear Sterns, 3–4, 24, 106 Bond indexes, 95 Bond markets, 54–55, 92. See also Treasury, Department of the Brazil, 79 Broadmarket Prime, 108–09. See also Madoff, Bernard Brokers, 58 Brookings Institution, vii–viii, 2 Bull markets, 90–91, 92 BXM index, 94 Capital: asset prices and, 58; capital adequacy rules, 65–66, 77; capital markets, 67; capital requirements for trading books, 61–65; cyclicality of regulatory capital framework, 62; financial crisis of 2007–09 and, 57; firesale externality and regulatory capital, 59–61; Long-Term Capital Management crisis of 1998, 55; value-risk-based capital adequacy ratios, 65. See also Banks and banking Casey, Quirk, and Associates/Bank of New York survey, 84 CBOE. See Chicago Board Options Exchange CDSs. See Credit default swaps CEO compensation. See Compensation standards and structures CFTC. See Commodity Futures Trading Commission Chan, N., 95 Chen Siqing, 78

7/12/10 6:09 PM

index Chicago Board Options Exchange (CBOE), 95 China, 78 China Banking Regulatory Commission, 79 Chrysler Financial, 127 Chrysler Group LLC, 127 Clearinghouses, 72, 75, 110 Clinton, William J. (“Bill”), 115 Citigroup, 127 Commerce (de novo bank), 26 Commodity Futures Trading Commission (CFTC; U.S.), 71, 110 Compensation standards and structures: as asset-management fee, 118; better disclosure of executive incentives, 137; CEO compensation, 7–8, 49, 115, 121–25; compensation committees, 127, 128, 138; components and definitions of, 119, 120, 122–25; debate over, 116–17; financial crisis and, 7, 49, 126; firm size and, 118; flaws in the pay-setting process, 117, 126–27; of financial institutions, 7; for hedge and equity funds, 7, 8; nonbinding say-onpay shareholder votes, 134–35; periodic assessment of risk-taking incentives, 135–36; political factors of, 126; regulation of, 7, 8–9, 115–16, 117, 120, 125–37; recommendations for, 126–27, 137; subordinates’ compensation, 131; Treasury proposals for, 127–28 Compensation standards and structures— specific: bonuses, 127, 128–29; clawback provisions, 127, 132, 138; earnings management incentives, 127; equity options, 116; equity performance incentives, 119; measurement of performance incentives, 121; performance incentives, 122, 124, 125, 126, 129–32, 135, 137–38; perquisite compensation, 127, 133–34, 138; restrictions on salary, bonus, and option payments, 128–32; risk-taking incentives, 121, 127, 133, 135–37, 138; severance and change-incontrol payments, 127, 132–33, 138; stocks and stock options, 116, 117, 121,

06-0404-1 bm.indd 145

145 122–25, 127, 128–31; tax gross-ups, 127, 133–34, 138 Compensation standards and structures— U.S.: benchmarking of U.S. executive pay, 117–20; executive pay in the U.S. financial sector, 120–25; regulation of executive pay, 125–37 Competition. See Economic issues Consumers, 26, 31, 43 Core, John, 7, 115–40 Credit cards. See Loans and lending Credit default swaps (CDSs), 69, 70–73. See also Derivatives Credit markets, 72–73 Credit ratings, 67, 72 Credit Suisse, 88–87, 90 Credit Suisse indexes, 89, 92 Debit cards. See Banks and banking Debt: collateralized debt obligations, 62, 66, 69, 71; distressed debt as level 2 assets, 107; structured subprime debt, 89 De novo banks, 26 Derivatives: derivatives portfolio replication, 98–99; risks and regulation of, 5, 70–73, 87; trading of, 110; use of, 73. See also Credit default swaps; Structured investment products Developing countries, 78. See also Emerging market economies Dot.com bubble, 24 EAFE Ex Japan index, 92 Economic issues: balance sheet factors, 13–14; changing macroeconomic environment for U.S. commercial banking, 29–30; competition, 12, 13, 20, 26, 31, 32–33, 43, 45, 128; costs of regulation, 31; deposits, 20–21; economic cycles, 62; economic stress, 62–64; GDP (gross domestic product), 14; household debt, 18–19; income inequality, 116; inflation, 12, 15, 29–30; interest rates, 14–16, 18, 19–20, 24–26, 29, 42, 45, 55; leverage and leveraging, 17, 32; marketoriented systemic risk, 54–55; maturity

7/12/10 6:09 PM

146 transformation, 56; recessions of 1981, 1991–92, 2008–09, 15, 16, 29. See also Banks and banking; Compensation standards and structures; Financial crisis of 2004–09 Emerging market economies, 67, 79. See also Developing countries Endowments, 100, 101t, 102–03 Equity indexes, 90 Equity markets, 89, 91–92, 93, 95 EU. See European Union Eurex Clearing, 71 Europe, 58, 68, 76. See also individual countries European Commission, 63, 67, 71 European Union (EU), 76–77, 110. See also individual countries Exchange-traded funds (ETFs), 96, 98 Execucomp database (S&P), 120 FASB. See Financial Accounting Standards Board FDIC. See Federal Deposit Insurance Corporation Federal Deposit Insurance Corporation (FDIC), 14, 27, 32, 45, 135 Federal Reserve, 2, 24, 32, 76 Federal Reserve Bank, 15, 30, 33 Fees, 7, 84, 85, 86. See also Banks and banking—fees Feinberg, Ken (Special Master for TARP Executive Compensation), 8, 127–28, 129–31, 136, 138 Financial Accounting Standards Board (FASB), 106–07 Financial crisis of 2004–09: bank losses and, 61, 64; casualties of, 2–5, 28, 89; compensation and, 7, 8; deposits and, 26–27; effects of, 29–37, 59; financial crisis after 2007, 27; G-20 reforms and, 53–79; lending and, 25–26; recovery from, 12, 29, 35, 90; Treasury effects, 24–25 Financial crisis of 2004–09—background and causes of: analysis by the G-20, 54; banking problems, 24–25, 65, 66; crisis

06-0404-1 bm.indd 146

index after 2007, 27; deposits, 26–27; Federal Reserve, 24–25; hedge funds, 5, 76, 77, 89–90, 91–93, 111; lending, 25–26; over-the-counter derivatives, 70, 71–72; prediction of, 1; short selling, 73, 74; subprime mortgage problem, 67; turmoil in liquidity markets, 56–57, 59 Financial holding companies, 31 Financial services, 8, 12, 32, 116 Financial Services Authority (U.K.), 75 Financial Services Oversight Council (U.S.), 76 Financial Stability Board, 78 Financial Stability Forum, 64 Financial systems, 55 Fire-sale externality, 59–60 Foreign banks, 26 Foreign Corrupt Practices Act (1977), 31, 32 France, 76 Frank, Barney (D-Mass.), 71 Fuchita, Yasuyuki, vii, 1–9 Funds of funds, 85, 86, 87, 89, 96–97, 98, 111 Fung, W., 105 Futures markets, 87, 95, 99 GDP (Gross domestic product). See Economic issues Geczy, Christopher, 5, 6–7, 83–113 Geithner, Timothy (secretary, U.S. Treasury), 8, 125–26, 138 General Motors Co., 70, 127 Germany, 76 Glass-Steagall Act (1933), 14 GMAC Inc., 26, 127 Gold, 95 Goldman Sachs, 4, 45, 98 Great Recession of 2008–09. See Financial crisis of 2008–09 Gross domestic product (GDP). See Economic issues Group of Twenty (G-20), 4, 53–54, 76 Guay, Wayne, 7, 115–40 “Haircuts,” 56, 57, 59, 66 Harvard, 100

7/12/10 6:09 PM

index Hedge funds: aggregate hedge fund portfolios, 94, 95; asset prices and, 58; benchmarks of, 93–94, 95–96, 111; business model of, 111; definition of hedge funds, 83; in diversified portfolios, 89–93, 97; effects of hedge fund trading, 77; fees and costs for, 7, 84, 85, 86, 87, 96–97, 105, 111, 118; financial crisis and, 5, 77, 89–90, 91–93, 111; future of, 5–6, 7, 94, 111; hedge (in hedge funds), the, 89, 93; hedge fund indexes, 88–89; investors in, 85–86, 87, 91–92; issues and trends revealed by hedge-fund survey, 84–88; liquidity in, 77–78, 84, 86, 87, 95; lock-ups, 6, 86; managers and management of, 99–100, 104–05, 110, 111, 118; marking-to-market of, 106–09; operation and performance of, 6, 84, 93–94, 97, 102, 111; provisions and characteristics of, 6, 90; recommendations for, 88, 111; registration of, 76, 109–10; regulation of, 76–78, 85, 87, 96, 109–10; reporting, 87, 109–10; risk factors of, 6, 85, 86, 87, 89, 96, 104, 109; selection and survivor biases, 105; spreads of, 94; trackers, 96–99, 100–04; trading strategies and, 85, 87–88, 94, 95, 96–105, 111; transparency in, 84, 85, 87, 96, 111; typology of, 6; volatility of, 89–93, 102 Hedge funds—value of: aggregate hedge fund index returns, 88; aggregate hedge fund portfolio multiple-factor contemporaneous market beta, 94; alpha, 104–05, 111; exotic beta, 94–96, 104; factor exposure, 93; factor models, 93–94; general factors, 88; market or factor timing, 105–06; replication, 87, 96–105; vanilla beta, 88–93, 103 Herfindahl-Hirschman Index, 22 Herring, Richard, vii, 1–9 Housing issues, 1, 2–3 ICE Clear Europe, 71 IMF. See International Monetary Fund IndexIQ, 98 India, 79

06-0404-1 bm.indd 147

147 IndyMac, 27 ING Direct, 26 Internal Revenue Code, 115 International Accounting Standard 27, 68 International Monetary Fund (IMF), 65 International Organization of Securities Commissions, 78 International Swaps and Derivatives Association, 72 Internet, 26 Investment Advisers Act of 1940, 76 Investment banks and banking. See Banks and banking Investment Company Act of 1940, 83, 96, 97 Investor Protection Act of 2009, 68 Investors, 5–6, 85. See also Hedge funds; Pension funds Kamiyama, Tetsuya, 4–5, 53–81 Kat, H., 99 Kodachi, Kei, 4–5, 53–81 Korea, 79 Lehman Brothers, 2, 3–4, 58, 72–72, 89, 106 Leverage, 58, 65–67 Liabilities: bank assets and, 56, 60; deposit liabilities, 45, 46f; hedge fund industry stress and, 111; level 1 assets and, 106; long-term assets and short-term liabilities, 15–16, 25, 27, 56, 66; market liabilities, 45 Liquidity: bank trading books and, 64–65; hedge funds and, 77–78, 84; marketability and, 66; maturity transformation and market liquidity, 54–56; turmoil in liquidity markets, 56–58, 59–60 Litan, Robert, vii, 1–9 Loans and lending: bank lending, 43–44; credit card lending, 19–20, 30, 44; demand for loans, 19–20, 43; financial crisis of 2004–09 and, 24, 25–26, 54; home equity lending, 19; real estate lending, 19–20; regulation of, 30; risk

7/12/10 6:09 PM

148 issues, 19, 42; U.S. credit liabilities, 26. See also Banks and banking; Securities and securitization Loans and lending—mortgages: financial crisis of 2004–09 and, 67; leading lenders, 40, 42; origination fees from, 40; subprime and alt-A mortgage lending, 2–3, 25–26, 42, 54, 67 Lobbies and lobbying, 31 London Stock Exchange, 77. See also Stock markets Long-Term Capital Management, 78 Luo Ping, 79 Madoff, Bernard, 96, 108–09 Margins, 57, 58–59 Market timing, 104–05 Marking to market, 95, 106, 107–09 McIntyre, Alan, 2, 3, 11–51 Merrill Lynch, 2, 4, 62 Merrill Lynch factor model, 97, 98 Metropolitan statistical areas (MSAs), 39–40, 41 Mexico, 79 Money market funds, 2 Morgan Stanley, 4 Mortgages. See Loans and lending— mortgages MSAs. See Metropolitan statistical areas Mutual funds, 56, 96, 98, 118 NACUBO-Commonfund Study of Endowments, 100 Naked short selling. See Short selling New York Stock Exchange (NYSE), 77. See also Stock markets Nomura Institute of Capital Markets Research, vii–viii, 2 Nomura Securities, 4 Nonbanks and nonfinancial firms, 4, 26, 33, 121, 123. See also Investment banks and banking NYSE. See New York Stock Exchange Obama, Barack H. (D-Ill.; president), 116, 134

06-0404-1 bm.indd 148

index Over-the-Counter Derivatives Markets Act of 2009, 71, 110. See also Derivatives Ovitz, Michael S., 133 Pastor-Stambaugh liquidity benchmark, 95 Pension funds, 6, 85–86. See also Investors Peterson, Collin (Democratic-Farmer-Labor Party-Minn.), 71 Ponzi schemes, 96, 108–09 Princeton, 100 Private equity firms, 118 Private equity funds, 76, 77 Private funds, 76, 87 Private Fund Investment Advisers Registration Act of 2009, 76 QAI portfolio, 98 Rajan, Raghuram, 60 Recommendations. See Banks and banking—recommendations; Compensation standards and structures; Hedge funds; Rules and regulations— recommendations “Reform Pay Practices for Top Executives to Align Compensation with Long-Term Value Creation and Financial Stability” (press release; Treasury), 129 Regulations. See Rules and regulations Rehypothecation, 58, 60, 61 Replication, 87, Repurchase (repo) market, 56–58, 59, 66, 68 Risk and risk management: commercial banking and, 42; concentration risk, 70; credit default swaps and, 72; credit risk, 42, 61, 65; default risk, 62; equity risk, 119, 131; fair-market accounting and, 59; in hedge funds, 6, 85, 86, 87, 96; incremental risk charge, 61–62; idiosyncratic risk, 93; investor views of, 85; liquidity risk, 42, 56; manager risk, 96; market and market-oriented risk, 54–55, 58, 60, 61, 93; measuring risk of investments, 93; operational risk, 61, 96, 109; risk-adjusted CEO compensation,

7/12/10 6:09 PM

index 119; risk-taking CEO compensation incentives, 121, 127, 133, 135–37, 138; stressed value-at-risk measure, 61–62; subprime mortgages and, 56; systemic risk, 76, 77, 93; value at risk, 62–64 Robertson, Julian, 100 Rules and regulations: accounting standards, 68; circuit breaker rules, 73, 75; compensation rules, 126–37; costs of, 31–32; disclosure requirements, 68; fire-sale externality and regulatory capital, 59–61; 5 percent risk-retention rule, 68; of hedge funds, 76–78, 85, 87, 96, 109–10; Investor Protection Act of 2009, 68; leverage ratios, 66–67; market risk rules, 62–64; off-balance-sheet vehicles, 68; over-the-counter derivatives, 70–73; regulation E (Federal Reserve), 31; regulatory arbitrage, 4, 64, 65–66, 78; risk retention rules, 68; short selling, 73–76; structured-product markets and, 67–70; subprime mortgage problem and, 67; uptick rules, 73–74, 75, 110 Rules and regulations—recommendations: general financial regulation, 79; G-20 recommendations, 4; hedge funds, 77–78; over-the-counter derivatives, 71–73; risk retention rules, 68–70; short-selling, 74–76 Russia, 79, 105 Safe harbors, 83, 97 S&P (Standard & Poor’s) 500 index, 88–89, 90–91, 95, 118 Sarbanes-Oxley Act of 2002, 132 Sato, Takafumi, 79 SEC. See Securities and Exchange Commission Securities Act (1933), 83 Securities and Exchange Commission (SEC): credit default swaps and, 70, 71; disclosure of compensation arrangements, 137; discovery of fraud by, 96, 109; enforcement by, 109, 110; fair value accounting and, 106–07; Over-theCounter Derivatives Markets Act and, 71;

06-0404-1 bm.indd 149

149 registration and operational requirements and, 76, 87, 109–10; short selling and, 73, 75–76, 110; uptick rule and, 74, 110 Securities and securitization: asset-back securities, 57, 62, 66; bank borrowing and, 57, 66; as bank collateral, 59; bank profitability and, 18, 19; charges on securitization exposure, 61; credit default swaps and, 72, 73; financial crisis of 2004–09 and, 25–26; funding of bank positions and, 56; goals and growth of, 19, 69; regulation of, 5, 69–70; resecuritization exposure, 62; risk diversification effect of, 68; short selling and, 75; as a source of credit, 69; subprime mortgage lending and, 2–3. See also Structured investment products Securities Exchange Act of 1934, 58 Securities Investor Protection Act (1970), 58 SFAS 157. See Statement of Financial Accounting Standards 157 Shareholders: in the banking industry, 11, 15, 27, 49; CEO compensation and, 125, 126, 127, 137, 138; change-in-control agreements and, 133; equity holdings and, 136; of private equity funds, 77; sayon-pay votes of, 8–9, 116, 134–35 Sharpe ratios, 90, 91, 92, 102 Short selling, 4, 5, 73–76, 77, 87, 89, 110 Standard & Poor’s. See S&P Statement of Financial Accounting Standards 157 (SFAS 157): adoption of, 106–07; fair valuation principles of, 87, 106; hedge funds and, 6–7; importance of, 96 Stock markets, 54, 70, 73, 74, 75, 77, 118 Structured investment products, 56, 67–70. See also Derivatives; Securities and securitization Supervisory Capital Assessment Program (2008), 27, 32 TARP. See Troubled Asset Relief Program Tax deductions for CEO compensation, 115 Thomson Financial and Nelson, 84 Tiger Cubs (funds), 100

7/12/10 6:09 PM

150 TIPS (Treasury inflation-protected securities). See Treasury, Department of the Trackers. See Hedge funds Trading books, 64–65 Treasury, Department of the: bank capital levels and, 60; bond yields and rates of, 15, 18; inflation-protected securities (TIPS), 29; limitation of executive compensation by, 8, 136; sale of distressed firms, 133; value at risk and, 64. See also Geithner, Timothy; Finberg, Ken; Troubled Asset Relief Program Treasury inflation-protected securities (TIPS). See Treasury, Department of the Troubled Asset Relief Program (TARP), 8, 27, 49, 127, 132, 135, 137 “Turner Review” (UK), 69

index United States (U.S.), 29, 58, 67, 117–37. See also Securities and Exchange Commission; Treasury, Department of€the University of Pennsylvania, 100 U.S. See United States Volatility index (VIX), 95 Volcker, Paul, 15 Washington Mutual, 21, 27 Wells Fargo, 40, 42 Wharton Financial Institutions Center, vii–viii, 2 Wharton Hedge Fund Survey of Pension Consultants, 84–88, 110 Yale, 100

United Kingdom (U.K.), 58, 118–19, 137–38

06-0404-1 bm.indd 150

Zeltkevic, Michael, 2, 3, 11–51

7/12/10 6:09 PM

Brookings Institution The Brookings Institution is a private nonprofit organization devoted to research, education, and publication on important issues of domestic and foreign policy. Its principal purpose is to bring the highest quality independent research and analysis to bear on current and emerging policy problems. The Institution was founded on December 8, 1927, to merge the activities of the Institute for Government Research, founded in 1916, the Institute of Economics, founded in 1922, and the Robert Brookings Graduate School of Economics and Government, founded in 1924. Interpretations or conclusions in Brookings publications should be understood to be solely those of the authors.

Nomura Foundation Nomura Foundation is a nonprofit public interest incorporated foundation which aims to address social and economic issues involving Japan and the rest of the world by devoting private sector resources to promote international exchanges and the interchange between social science theory and practice. The foundation also provides grants and scholarships to support the social sciences, arts and culture, and up-and-coming international artistic talent. In the area of World Economy Research Activities it sponsors research, symposiums, and publications on current trends in capital markets of advanced and emerging economies as well as on topical issues in global macroeconomic stability and growth. It relies on a network of institutions from Europe, the United States, and Asia to assist in organizing specific research programs and identifying appropriate expertise.

Nomura Institute of Capital Markets Research Established in April 2004 as a subsidiary of Nomura Holdings, Nomura Institute of Capital Markets Research (NICMR) offers original, neutral studies of Japanese and Western financial markets and policy proposals aimed at establishing a market-structured financial system in Japan and contributing to the healthy development of capital markets in China and other emerging markets. NICMR disseminates its research among Nomura Group companies and to a wider audience through regular publications in English and Japanese.

Wharton Financial Institutions Center, University of Pennsylvania The Wharton Financial Institutions Center is one of twenty-five research centers at the Wharton School of the University of Pennsylvania. The Center sponsors and directs primary research on financial institutions and their interface with financial markets. The Center was established in 1992 with funds provided by the Sloan Foundation and was designated as the Sloan Industry Center for Financial Institutions, the first such center designated for a service-sector industry. It is now supported by private research partners, corporate sponsors, and various foundations and nonprofit organizations. The Center has hundreds of affiliated scholars at leading institutions worldwide, and it continues to define the research frontier, hosting an influential working paper series and a variety of academic, industry, and “crossover” conferences.

06-0404-1 bm.indd 151

7/12/10 6:09 PM

Contributors: John E. Core, Wharton School, University of Pennsylvania ◆ Christopher C. Geczy, Wharton School, University of Pennsylvania ◆ Wayne R. Guay, Wharton School, University of Pennsylvania ◆ Tetsuya Kamiyama, Nomura Institute of Capital Markets Research ◆ Kei Kodachi, Nomura Institute of Capital Markets Research ◆ Alan McIntyre, Oliver Wyman Group ◆ Michael Zeltkevic, Oliver Wyman Group YASUYUKI FUCHITA is a senior managing director at the Nomura Institute of Capital Markets Research in Tokyo. He coedited Prudent Lending Restored (Brookings, 2009) with Richard J. Herring and Robert E. Litan and Pooling Money (Brookings, 2008) with Litan. RICHARD J. HERRING is the Jacob Safra Professor of International Banking and professor of finance at the Wharton School, University of Pennsylvania, where he is also codirector of the Wharton Financial Institutions Center. ROBERT E. LITAN is a senior fellow in Economic Studies at the Brookings Institution and vice president for research and policy at the Kauffman Foundation. His many books include Good Capitalism, Bad Capitalism, and the Economics of Growth Prosperity (Yale University Press, 2007), written with William J. Baumol and Carl J. Schramm.

NOMURA INSTITUTE OF CAPITAL MARKETS RESEARCH Tokyo www.nicmr.com/nicmr/english

Cover design and illustration montage by Claude Goodwin

BROOKINGS / NICMR

BROOKINGS INSTITUTION PRESS Washington, D.C. www.brookings.edu

After the Crash

In After the Crash, noted economists Yasuyuki Fuchita, Richard Herring, and Robert Litan bring together a distinguished group of experts from academia and the private sector to take a hard look at how the financial industry and some of its practices are likely to change in the years ahead. Whether or not you agree with their conclusions, the authors of this volume—the most recent collaboration between Brookings, the Wharton School, and the Nomura Institute of Capital Markets Research—provide well-grounded insights that will be helpful to financial practitioners, analysts, and policymakers.

Fuchita / Herring / Litan

A

s the global economy continues to weather the effects of the recession brought on by the financial crisis of 2007–08, perhaps no sector has been more affected and more under pressure to change than the industry that was the locus of that crisis: financial services. But as policymakers, financial experts, lobbyists, and others seek to rebuild this industry, certain questions loom large. For example, should the pay of financial institution executives be regulated to control risk taking? That possibility certainly has been raised in official circles, with spirited reactions from all corners. How will stepped-up regulation affect key parts of the financial services industry? And what lies ahead for some of the key actors in both the United States and Japan?

After the

Crash The Future of Finance

Yasuyuki Fuchita, Richard J. Herring, and Robert E. Litan, Editors