Behavioral Finance JAFFW2008

Volume 18, No. 2 Fall/Winter 2008

Editors:
Betty J. Simkins Ramesh P. Rao Charles W. Smithson

Academic Contributions
Behavioral Finance: Quo Vadis? Werner De Bondt, Gulnur Muradoglu, Hersh Shefrin, and Sotiris K. Staikouras The Effects of Institutional Risk Control on Trader Behavior Ryan Garvey and Fei Wu Why Do People Trade? Anne Dorn, Daniel Dorn, and Paul Sengmueller The Long-Term Value of Trade Informativeness Michel Rakotomavo Shareholder Theory-How Opponents and Proponents both Get It Wrong Morris G. Danielson, Jean L. Heck, and David R. Shaffer Student Managed Investment Funds: An International Perspective Edward C. Lawrence

Academic Assoc. Editors:
Edward I. Altman Kenneth A. Borokhovich Christine A. Brown Jennifer S. Conrad Javier Estrada Mark J. Flannery Gerald D. Gay Stuart I. Greenbaum Allaudeen S. Hameed Andrea J. Heuson Takato Hiraki Brian M. Lucey Sotiris K. Staikouras Laura T. Starks David A. Walker Ralph A. Walkling Samuel C. Weaver Lawrence W. Licon Martin R. Young

Practitioner Contribution
Behavioral Basis of the Financial Crisis Joseph V. Rizzi

Roundtable
University of Rochester Roundtable on Bankruptcy and Bailouts: The Case of the US Auto Industry Panelists: Thomas Jackson, Charles Hughes, James Brickley, Joel Tabas, and Clifford Smith Moderator: Mark Zupan

Practitioner Assoc. Editors:
Niso Abuaf Donald Chew Mike Edleson John Fraser Gene Guill Andrew J. Kalotay Ira G. Kawaller Joseph V. Rizzi D. Sykes Wilford

Interview
Pioneers in Finance: Vernon Smith Interview Terrance Odean and Betty J. Simkins

Case Study
The 2008 Federal Intervention to Stabilize Fannie Mae and Freddie Mac W. Scott Frame

Book Reviews
Book Review: Ending the Management Illusion: How to Drive Business Results Using the Principles of Behavioral Finance By Hersh Shefrin Andrea Heuson Book Review: The Venturesome Economy by Amar Bhidé Colby Wright

Financial Puzzles
Stewart C. Myers

THE FINANCIAL MANAGEMENT ASSOCIATION INTERNATIONAL OFFICERS—DIRECTORS—EDITORS
President Douglas R. Emery University of Miami 2008-2009 Secretary/Treasurer Ajay Patel Wake Forest University 2002-2012 Chairman, Finman Corporation Jennifer Conrad UNC-Chapel Hill 2008-2011 Vice President-Program G. Andrew Karolyi The Ohio State University 2009-Reno, Nevada Vice President-Financial Education Robert Parrino University of Texas-Austin 2008-2010 Vice President-Global Services Alexander J. Triantis University of Maryland 2007-2009 Vice President-Practitioner Services O. Rawley Thomas LifeCycle Returns 2008-2011 Editors, Survey and Synthesis Series John Martin Baylor University James Schallheim University of Utah 2004-2010 Editor, Financial Management William G. Christie Vanderbilt University 2006-2011 Editors, FMA Online Executive Editor Betty J. Simkins Oklahoma State University 2005-2009 Editors John Finnerty Fordham University Mark Flannery University of Florida Sheridan Titman University of Texas at Austin 2005-2009 Editors, Journal of Applied Finance Betty J. Simkins & Ramesh P. Rao Oklahoma State University Charles W. Smithson Rutter Associates 2007-2010

Journal of Applied Finance (ISSN 1534-6668) is published by the Financial Management Association International, an affiliate of the Finman Corporation. It is published semi-annually. The Editors and the Association assume no responsibility for the views expressed by the authors. Membership dues in the Association include a one-year subscription to the journal. Membership fees: New Professional $100, Renewal Professional $70, New Sustaining $125, and Renewal Sustaining $95. An application form is available inside this issue. JAF subscriptions for libraries are available. Contact Financial Management Association International, University of South Florida, College of Business Administration, Suite 3331, Tampa, FL 33620-5500, Telephone: (813) 974-2084 for further information. Memberships, Subscriptions and Address Changes: Write Financial Management Association International, University of South Florida, College of Business Administration, Suite 3331, Tampa, FL 33620-5500. Manuscripts: Electronically submit your submission form and a PDF file at www.fma.org. A submission fee is required for evaluation of each manuscript: $200 for non-FMA members, $130 for doctoral students who are not FMA members, and $100 for FMA members (U.S. dollars). The non-member submission fees include an FMA membership for the submitting author. Style information for manuscripts is located on the inside back cover of this journal. Permission to Quote or Republish: Blanket permission is granted to any individual wishing to use articles appearing in Journal of Applied Finance for educational (university classroom) purposes. Written permission from the Financial Management Association International or the Editor is not required. To make any other requests for permission to quote or republish, write to Financial Management Association International, University of South Florida, College of Business Administration, Suite 3331, Tampa FL 33620-5500. Telephone: (813) 974-2084; Fax: (813) 974-3318; Email: [email protected]; Website: http:// www.fma.org Copyright © 2008 Financial Management Association International, an affiliate of the Finman Corporation. Printed by Dartmouth Printing Company, Hanover, NH. Printed in the U.S.A.

Journal of Applied Finance
Volume 18 Number 2 Fall/Winter 2008

EDITORS
Betty J. Simkins Oklahoma State University Ramesh P. Rao Oklahoma State University Charles Smithson Rutter Associates

ASSISTANT EDITOR
Heidi Carter Oklahoma State University

ASSOCIATE EDITORS Academic
Edward I. Altman New York University Kenneth A. Borokhovich Cleveland State University Christine A. Brown University of Melbourne, Australia Jennifer S. Conrad University of North Carolina Javier Estrada IESE Business School Barcelona, Spain Mark J. Flannery University of Florida Gerald D. Gay Georgia State University Stuart I. Greenbaum Washington University, St. Louis Allaudeen S. Hameed National University of Singapore Andrea J. Heuson University of Miami Takato Hiraki Kwansei Gakuin University, Japan Brian M. Lucey Trinity College, Dublin Sotiris K. Staikouras Cass Business School, London Laura T. Starks University of Texas at Austin David A. Walker Georgetown University Ralph A. Walkling Drexel University Samuel C. Weaver Lehigh University Lawrence W. Licon Arizona State University Martin R. Young Massey University, New Zealand

Practitioner
Niso Abuaf Independent Consultant Donald Chew Morgan Stanley Mike Edleson Morgan Stanley John Fraser Hydro One Gene Guill Deutsche Bank Andrew J. Kalotay Andrew Kalotay Associates, Inc. Ira G. Kawaller Kawaller & Co. Joseph Rizzi CapGen Financial D. Sykes Wilford EAQ Partners; The Citadel

SPONSORS
Oklahoma State University University of South Florida

JAF REFEREES
JAF would like to thank all the referees who have reviewed manuscripts since the last issue. We appreciate the efforts of our reviewers for responding as soon as possible and for providing constructive comments for the authors. Tom Aabo Aarhus School of Business James J. Angel Georgetown University Thomas M. Arnold University of Richmond Chenchu Bathala Cleveland State University TK Bhattacharya Cameron University Kenneth A. Borokhovich Cleveland State University Helen M. Bowers University of Delaware Christine A. Brown University of Melbourne Kelly R. Brunarski Miami University Antonio Camara Oklahoma State University David A. Carter Oklahoma State University Don Chance Louisiana State University Donald H. Chew, Jr. Morgan Stanley Jennifer S. Conrad University of North Carolina Arnald R. Cowan Iowa State University Frank D’Souza Loyola College Maryland Michael Edleson Morgan Stanley Javier Estrada IESE Business School Michael G. Ferri George Mason University John R.S. Fraser Hydro One, Inc. Gabriele Galati De Nederlandsche Bank Jacqueline L. Garner Drexel University Gerald D. Gay Georgia State University Stuart L. Gillan Texas Tech University Radha Gopalan Washington University Stuart I. Greenbaum Washington University Yilmaz Guney University of Hull Benton E. Gup University of Alabama Allaudeen Hameed National University of Singapore Joel T. Harper Oklahoma State University Scott E. Hein Texas Tech University Andrea J. Heuson University of Miami Jonathan M. Karpoff University of Washington Eric Kelley University of Arizona Sivarama Krishnan University of Central Oklahoma David R. Lange Auburn University Montgomery Wendell L. Licon Arizona State University John D. Martin Baylor University K.C. Ma Stetson University Cathy Niden LECG, Inc. Thomas J. O’Brien University of Connecticut Tim Opler Torreya Partners Christos Pantzalis University of South Florida Janet Payne Texas State University San Marcos Ivilina Popova Texas State University San Marcos Jack S. Rader Financial Management Association International Daniel A. Rogers Portland State University Kasper Roszbach Riks Bank W. Gary Simpson Oklahoma State University Charles Smithson Rutter Associates Sotiris Staikoros Cass Business School, London Mathijs van Dijk Eramus University David A. Walker Georgetown University Larry Wall Federal Reserve Bank of Atlanta Samuel C. Weaver Lehigh University Melissa A. Williams University of Houston — Clear Lake

Journal of Applied Finance
Volume 18 Number 2 Fall/Winter 2008

Academic Contributions
7 22 37 51 62 67 Behavioral Finance: Quo Vadis? The Effects of Institutional Risk Control on Trader Behavior Why Do People Trade? The Long-Term Value of Trade Informativeness Shareholder Theory-How Opponents and Proponents both Get It Wrong Student Managed Investment Funds: An International Perspective Werner De Bondt, Gulnur Muradoglu, Hersh Shefrin, and Sotiris K. Staikouras Ryan Garvey and Fei Wu Anne Dorn, Daniel Dorn, and Paul Sengmueller Michel Rakotomavo Morris G. Danielson, Jean L. Heck, and David R. Shaffer Edward C. Lawrence

Practitioner Contribution
84 Behavioral Basis of the Financial Crisis Joseph V. Rizzi

Roundtable
97 University of Rochester Roundtable on Bankruptcy and Bailouts: The Case of the US Auto Industry Panelists: Thomas Jackson, Charles Hughes, James Brickley, Joel Tabas, and Clifford Smith Moderator: Mark Zupan

Interview
116 Pioneers in Finance: Vernon Smith Interview Terrance Odean and Betty J. Simkins

Case Study
124 The 2008 Federal Intervention to Stabilize Fannie Mae and Freddie Mac W. Scott Frame

Book Reviews
137 139 Book Review: Ending the Management Illusion: How to Drive Business Results Using the Principles of Behavioral Finance By Hersh Shefrin Book Review: The Venturesome Economy by Amar Bhidé Andrea Heuson Colby Wright

Financial Puzzles
142 Stewart C. Myers

4

Letter from the Editors
This is our second issue as editors of the Journal of Applied Finance (JAF). We were delighted to note that the first issue was received very well by our readers and we wish to thank all who sent us their feedback. We received positive feedback on the new layout of JAF with sections on academic and practitioner contributions, roundtable discussions, case and clinical studies, interviews, surveys, and book reviews. We will strive to keep these same features consistent across issues. Our continued success though depends on our ability to attract submissions in each of these categories, especially roundtables, surveys, case and clinical studies, and interviews. We would like to encourage our readers to be actively engaged in these types of submissions. As your editors, we are more than happy to work with you to develop submissions in these categories. We would also like to appeal to our practitioner community to consider writing for JAF and are willing to do what we can to identify academics that they can partner with.

In This Issue
As we mentioned in our first issue, one of our goals is to have one or two themes for each issue. In this issue our focus is on behavioral finance. Our lead article (Behavioral Finance: Quo Vadis?) provides the reader an overarching view of behavioral finance from its inception to the current state and beyond. It is based on a panel discussion held at the FMA-Europe meetings in Prague, 2008. The panelists included Werner De Bondt, Gulnur Muradoglu, Hersh Shefrin, and Sotiris Staikouras, who also authored the piece. Our academic contributions also include several articles on trading behavior: “The Long-Term Value of Trade Informativeness” by Michel Rakotomavo, “The Effects of Institutional Risk Control on Trader Behavior” by Ryan Garvey and Fei Wu; and “Why Do People Trade?” by Anne Dorn, Daniel Dorn, and Paul Sengmueller. There is also a provocative article on shareholder theory by Morris G. Danielson, Jean L. Heck, and David R. Shaffer (“Shareholder TheoryHow Opponents and Proponents both Get It Wrong”). In addition, our academic and practitioner readers will also find the article on student management investment funds by Ed Lawrence to be of interest (Student Managed Investment Funds: An International Perspective) The behavioral finance theme is continued in our practitioner contribution by Joseph Rizzi titled “Behavioral Basis of the Financial Crisis.” This article also continues a theme from our first issue on the subprime crisis. The behavioral finance theme continues in our “Pioneers of Finance” interview feature. Professor Vernon Smith, recipient of the 2002 Nobel Prize in Economics. In this interview, Professor Smith shares his perception of how experimental economics and behavioral economics are related. He goes on to provide some insights from his research on speculative bubbles in experimental markets that help us understand the recent bubble is US residential real estate, The interview was conducted by Terry Odean and Betty Simkins. We thank Terry for his contribution. Our readers will also enjoy our case/clinical study contribution by Scott Frame titled “The 2008 Federal Intervention to Stabilize Fannie Mae and Freddie Mac”. As a Federal Reserve insider, Scott Frame provides insights into the Fannie and Freddie debacles that only an insider can provide. The case study also continues the subprime crisis theme from our first issue. Our roundtable features a time topic: Bankruptcy and Bailouts: The Case of the US Auto Industry. The panelists feature several prominent auto industry executives and faculty members from the University of Rochester. We thank the University of Rochester for sponsoring the roundtable, Mark Zupan for moderating, and Don Chew for editing it. Our book review section features two books. One is titled Ending the Management Illusion: How to Drive Business Results Using the Principles of Behavioral Finance written by Hersh Shefrin, one of the pioneers in behavioral finance. The review was written by Andrea Heuson. The second is by Amar Bhidé, another prominent author who is an authority on strategy. The book is titled The Venturesome Economy and is reviewed by Colby Wright. This issue concludes with the solution to the Financial Puzzle that appeared in the Fall/Winter 2007 issue, along with two new puzzles by Stu Myers.

4

5 In Closing
In closing, we would like to highlight themes/topics we are considering for future issues. These include valuation, corporate restructuring, and dividends and share buybacks. We welcome your contributions and any suggestions you may have for JAF. Sincerely,

Ramesh P. Rao Paul C. Wise Chair Oklahoma State University Email: [email protected]

Betty J. Simkins Williams Cos. Professor of Business Oklahoma State University [email protected]

Charles Smithson Founding Partner Rutter Associates LLC [email protected]

6

Behavioral Finance: Quo Vadis?
Werner De Bondt, Gulnur Muradoglu, Hersh Shefrin, and Sotiris K. Staikouras

Behavioral finance endeavors to bridge the gap between finance and psychology. Now an established field, behavioral finance studies investor decision processes which in turn shed light on anomalies, i.e., departures from neoclassical finance theory. This paper is the summary of a panel discussion. It begins by reviewing the foundations of finance and it ends with a discussion of the future of behavioral finance and a self-critique. We describe the move from the standard view that financial decision making is rational to a behavioral approach based on judgmental heuristics, biases, mental frames, and new theories of choice under risk. A new class of asset pricing models, which adds behavioral elements to the standard framework, is proposed.

Proponents of behavioral finance argue that poorly informed and unsophisticated investors might lead financial markets to be inefficient. The debate between neoclassical and behavioral finance is wide ranging, and sometimes explains differences in policy recommendations on such issues as financial regulation, corporate governance, or the privatization of social security. It had immediate impact worldwide including emerging markets (Muradoglu, 1989). Behavioral finance emerged as a field in the early 1980s with contributions by, among others, David Dreman, Robert Shiller, Hersh Shefrin, Meir Statman, Werner De Bondt and

Werner De Bondt is a Professor of Finance at DePaul University in Chicago, IL. Gulnur Muradoglu is a Professor of Finance at Cass Business School in London, UK. Hersh Shefrin is a Professor of Finance at Santa Clara University in Santa Clara, CA. Sotiris K. Staikouras is a Senior Lecturer in Finance at Cass Business School in London, UK.

Richard Thaler. Soon, this small group of financial economists was meeting regularly with psychologists — including Paul Andreassen, Daniel Kahneman, and Amos Tversky — at the Russell Sage Foundation in New York. Five or six years later, the National Bureau of Economic Research began organizing semi-annual meetings. From its beginnings as a fringe movement, behavioral finance moved to a middle-ofthe-road movement, with spillover effects on marketing, management, experimental economics, game theory, political science and law. Now behavioral finance is poised to replace neoclassical finance as the dominant paradigm of the discipline. Traditionally, economists model behavior in terms of rational individual decision-makers who make optimal use of all available information. There is ample evidence that the rationality assumption is unrealistic. The path-breaking work of Herbert Simon, Tversky and Kahneman, Lola Lopes, and others on bounded rationality, judgmental heuristics, biases, mental frames, prospect theory, and SP/A theory has provided new foundations for financial economics. Behavioral finance studies the nature and quality of financial judgments and choices made by individual economic agents, and examines what the consequences are for financial markets and institutions. Investment portfolios are frequently distorted, with consequent excess volatility in stock and bond prices. Examples include the stock market crash of 1987, the bubble in Japan during the 1980s, the demise of Long-Term Capital Management, the Asian crisis of 1997, the dot-com bubble, and the financial crisis of 2008. Most everyone agrees that it is problematical to discuss these dramatic episodes without reference to investor psychology. The term “behavioral finance” has a variety of meanings. Our paper aims to provide an over-arching view of the field. It is a summary of a panel discussion. The paper is written for a wide spectrum of readers, including financial practitioners. It begins by examining the current state of finance, reviews some fundamental questions, and then

7

8
introduces behavioral concepts. “Behavioral finance: Quo Vadis?” Sections I and II review modern and behavioral finance, respectively. Section III briefly delves into the efficient markets literature. Section IV discusses key building blocks of the behavioral approach. Section V explores some new ideas in behavioral asset pricing and behavioral corporate finance. Section VI provides a self-critique. Section VII concludes.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

I. What Is Finance?
Let us start by defining finance. Even though the real economy and finance are linked, we usually make a distinction between the two. The real economy is where goods and services are produced and consumed, and where wealth is created. The world of finance is mostly seen as a sideshow. Even so, finance serves important functions such as the payment system, the pooling and transferring of funds, saving and investing, contract design, organizational architecture, and risk management. Anyone who contemplates the functions of finance, and the financial institutions involved in them (e.g., the banking system; insurance companies; money management firms; pension funds; rating agencies, and so on), soon realizes that the central unifying concept is asset valuation. Certainly, the theory of value, and comparisons of price and value, is what much of finance is about. Of course, valuation also impacts the decisions investors make about the composition of their portfolios and the decisions which managers make about the sources and uses of funds in their firms. Modern (or neoclassical) finance is the paradigm that has governed thinking in academic finance since the late 1950s. It flows from a philosophical tradition (the 18th century Enlightenment) that aims to reconstruct society with individual rational action as its centerpiece. Modern finance is built on two pillars. The first pillar is the concept of “beautiful people”, defined as logical, autonomous agents characterized by expected utility maximization (over time), risk aversion, Bayesian updating, and rational expectations. The second pillar is the concept of “beautiful markets” i.e. depending on the problem-at-hand, perfect, liquid, competitive, complete markets. Based on these two concepts as well as the mutual adjustment of demand and supply (plus an assortment of auxiliary assumptions), various asset pricing theorems are derived. In equilibrium, all agents reach their optimum. Investment portfolios are mean-variance efficient. Only systematic non-diversifiable risk is priced. There are no opportunities left for rational arbitrage. Conditional on what is known about the future, price equals value. What is the role of institutional factors such as market organization, regulatory framework, tax systems etc. in neoclassical finance? To a first approximation, there is none.

Rational agents work around institutional frictions and thereby render them immaterial to market outcomes. Of course, the process may take time. Merton Miller made this type of institutional arbitrage a favorite lecture theme. He spoke about institutions as potential distortions, though ultimately neutral mutations. Miller’s comments were often formulated in the context of regulatory barriers to financial innovation, but the link with the Miller-Modigliani theorems and the work of Ronald Coase is obvious. Robert Merton’s views are similar. His writings say that the basic functions of finance are the same, always and everywhere. What does change is the technological and regulatory environment. That is why banking in 2008 is different from banking in 1908, and why banking in Switzerland is different from banking in Egypt. How do modern finance theorists plead their case? They mostly reason in a logically deductive way starting from axioms that have a priori normative appeal.1 In the past, modern finance theorists rarely administered surveys (Muragdoglu, 1989) and they did not run experiments, although this is starting to change(Muragdoglu, Salih, and Mercan, 2005). Still, many financial economists believe that the swaying power of data cannot match the power of logic.

II. What Is Behavioral Finance?
Behavioral finance does not assume rational agents or frictionless markets. It suggests that the institutional environment is vitally important. The starting point is bounded rationality. Paul Slovic (1972) writes that “a full understanding of human limitations will ultimately benefit the decision-maker more than will naive faith in the infallibility of his intellect.” That economic and financial intuition is fragile may clash with our aspirations for mankind, but it looks more plausible than the opposite view that investors and advisors (as well as bankers and corporate managers) know perfectly well what to do. Behavioral finance is the study of how psychology impacts financial decisions in households, markets and organizations. The main question is: What do people do and how do they do it? The research methods are mostly (but not exclusively) inductive. Behavioral researchers collect “facts” about individual behavior (based on experiments, surveys, field studies, etc.) and organize them into a number of “superfacts.” The psychology of decision-making can be explored in various ways. A quarter-century ago, most effort went into cognition. Consider, for instance, the heuristics and biases literature pioneered by Tversky and Kahneman (1974) and Kahneman and Tversky (1979). Their main focus was on

The normative approach asks how decision-makers logically should act while the positive approach looks at how decisions are truly made.
1

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

9

questions such as: How do people think? How do they decide? difference? For an answer, we look to the decision process Current work continues to draw on cognitive research. In plus the well-known fact that people tend to stick with the addition, it studies emotion (mood; affect) and social status-quo. In case of a fatal car accident in the U.K., the law psychology (especially herding behavior). assumes –-unless the driver signs his license to the contraryWhat has been learned? The central insights of behavioral – that his bodily organs will not be donated. In Belgium, the finance are described in Barberis and Thaler (2003), Daniel default solution is the opposite, i.e. the driver’s organs are et al. (2002), De Bondt (2002, 2005, 2008a), Dreman (1995), donated. Note that in either country all it takes to modify the Shefrin (2001a, 2002) and Thaler (1993).2 There are three default is a signature.4 classes of findings. First, there Why is behavioral research is a catalog of biases, i.e., often so convincing? One Behavioral finance is based on predictable mistakes such as reason is that “good” behavioral overconfidence in judgment, research depends on support three main building blocks, wishful thinking, from multiple sources. For namely sentiment, behavioral procrastination, myopia, etc. instance, laboratory research Intuition is fragile. Note that it permits any reader who doubts preferences, and limits to is not alleged that financial the results to replicate the arbitrage. intuition is broken, only that it experiment “at home.” Further, can break. Specific errors many studies rely on surveys or depend on context, but are systematic nonetheless. The observe individual behavior (e.g., trading records) in a natural research examines psychological mechanisms which environment (e.g., Odean, 1998, 1999). Lastly, behavioral illuminate how the human mind works. It also explains why researchers also make use of conventional market-level price financial judgment is fallible. and volume data. This “one-two-three punch,” we believe, The second class of findings relates to the speculative provides a discipline to behavioral theorizing that is far dynamics of asset prices in global financial markets. Here, superior to what is typical for research in modern finance. the main insight is that the systematic errors of unsophisticated Decision anomalies (in the laboratory), matched with investors (“noise traders”) create profit opportunities for anomalies in the behavior of individual agents (in a natural experts, even if noise traders create a great deal of risk. environment), matched with market anomalies (when social Investor sentiment matters. Widely-shared misconceptions interaction allows fine-tuning) produce a powerful body of (that may be self-reinforcing) cause transient price bubbles, evidence. Take, for example, investor overreaction. Certainly, large and small. Certainly, rational arbitrage matters too but, experiments teach us that subjects do not update beliefs in since most people’s investment horizons are short, arbitrage Bayesian fashion (De Bondt, 1993, Muradoglu, 2002). does not wipe out inefficiencies. Second, when asked, investors tell us that they like to buy The third class of findings has to do with how decision past winner stocks but that they stay away from past losers. processes shape decision outcomes.3 Here too, the study of Regardless of what investors say, their trading records confirm fiascoes is informative, since it guides us to decision process the bias.5 Third, at the market level, we find predictable variables that are critical. Numerous specific applications of reversals in share prices (De Bondt and Thaler, 1985). The this finding appear in Nudge, a book authored by Richard laboratory, financial behavior, and market results appear to Thaler and Cass Sunstein (2008). One striking example has be connected. to do with organ donation (Johnson and Goldstein, 2003). The U.K. participation rate in organ donation is approximately III. Price and Value 15% whereas in Belgium it is over 95%. What explains this Milton Friedman (1953) and Eugene Fama (1965) argue that, even though naive investors may push security prices away from intrinsic values, more sophisticated traders will

These works lay emphasis on investment and asset pricing. However, Shefrin (2005) focuses on behavioral corporate finance. Apart from agency and asymmetric information problems, there are behavioral costs that obstruct the corporate value maximization process.
2

This type of research is especially relevant to the study of organizations. Everyday we learn more about committee decision-making (e.g. boards), the role of top managers in the creation of corporate wealth, and the pros and cons of bureaucratic formalities and red tape. As president of the American Finance Association, Michael Jensen asked that we break open the black box called the firm. Behavioral finance is contributing to that effort.
3

Our economist friends emphasize incentives. We ask them: What incentive scheme may achieve the same outcome (95% participation) that a seemingly minor adjustment in the decision process produces effortlessly?
4

Ironically, investors are more likely to hang on to losers than to winners if the changes in value occurred while the stocks were part of their portfolio (Shefrin and Statman, 1985).
5

10
find it worthwhile to correct any mispricing. In other words, competitive rational arbitrage guarantees that, at all times, the market valuation of any security reflects what is —and what can be— known about its future cash flows and the opportunity cost of capital. Based on market efficiency, finance academics have made two main assumptions about security valuation. First, securities have an intrinsic value based on their fundamentals; and second, their prices are not predictable on the basis of publicly available information. Among others, Fama (1965) argued that the competitive activity of arbitrageurs will bring security prices into line with fundamentals. Thus, the arbitrage activity of rational traders will prevail (over irrationals) as long as securities have close substitutes. Over the decades, this perspective, the efficient markets hypothesis, has been examined by many scholars. Behavioral finance has provided evidence which contradicts the notion of efficient markets. An example is the case of “Siamese twins” stocks (Rosenthal and Young, 1990; Froot and Dabora, 1999). Consider the share price movement of Royal Dutch/Shell Group, where Royal Dutch stock trades in the US/Netherlands and Shell stock trades in the U.K. The two companies’ original merging interests were on a 60:40 basis for Royal Dutch and Shell respectively. Thus a ratio of 1.5 (price of Royal Dutch relative to Shell) should have been achieved in order for the prices to reflect fundamentals. Froot and Dabora (1999) and Lamont and Thaler (2003) find that the relative price ratio ranges from 15% overvalued to 35% undervalued. This contradicts “the law of one price.” In relation to these stocks, there is also evidence that noise trader risk is a significant impediment to arbitrage (Scruggs, 2007). The efficiency of security prices has also been challenged by Graham (1949), Nicholson (1968), Basu (1977), Dreman (1977, 1980), and many others who believe that stocks with low price-to-earnings (PE) ratios are undervalued and stocks with high PE ratios are overvalued. Investors, these authors suggest, are overly pessimistic about the prospects of low PE stocks. Since the crowd avoids them, investing in low PE stocks is a profitable contrarian strategy.6 De Bondt and Thaler (1985) extend this idea with their analysis of investor overreaction and with the finding of predictable price reversals for long-term winner and loser stocks. Poterba and Summers (1988) obtain analogous reversals for national stock price indexes. There are other widely documented phenomena which are difficult to reconcile with efficient markets. Consider the following examples:

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

• Price volatility that is not linked to news: Cutler et al.
(1991) show that during periods with “no” major news announcements equity prices experience some of their largest one-day moves. A vivid example was the 22.6% drop in the Dow Jones Industrial Average on October 22, 1987. Roll (1984, 1988) offers systematic evidence of market volatility, not associated with information arrival. • Excess volatility: Keynes (1936, pp. 153-4) observes how “day-to-day fluctuations in the profits of existing investments …tend to have an altogether excessive, and even absurd, influence on the market.” This comment anticipates Shiller’s (1981, 1993) work on equity volatility. There, it is suggested that fluctuations in economic fundamentals alone (e.g., dividends) cannot possibly account for the observed aggregate price movements. • Earnings momentum: Stock prices “underreact” to annual and quarterly announcements of corporate earnings causing a post-announcement drift in returns, markedly for firms with low institutional shareholdings (Bartov et al., 2000). Bernard and Thomas (1989, 1990) were among the first to establish this effect, but the research goes back to Ball and Brown (1968).7 • Price momentum: For holding periods up to one year, Jegadeesh and Titman (1993, 2001) and others show trends in share prices of individual stocks, i.e., past winner stocks remain winners, and past losers remain losers.8 Yet, beyond one year, momentum is often followed by reversals. European and emerging markets exhibit similar patterns (Rouwenhorst, 1998, 1999; Muradoglu, 2000). Small firms feature more momentum than large firms (Jegadeesh and Titman, 1993; Grinblatt and Moskowitz, 1999; Lee and Swaminathan, 2000). Price momentum may be due to positive feedback trading. That is, when large increases in stock prices pull in new investors, the inflow of funds causes prices to rise further. It is probable that the phenomenon is also partly explained by earnings momentum, investor underreaction, and the gradual dissemination of news. Grinblatt and Han (2005) and Frazzini (2006) suggest that momentum can be explained by the disposition effect, a concept introduced by Shefrin and Statman (1985) whereby investors sell winners too early and hold losers for too long. • Equity premium puzzle: Historically, the spread between the return on equities and fixed income US government

Corporate news that is not directly related to earnings also predicts returns. See, e.g., Michaely et al. (1995) on dividends or Ikenberry et al. (1995) on share price repurchases. For a critique of these findings, see Fama (1998).
7

Other price-scaled ratios, e.g., the book-to-price ratio, also forecast stock returns. See, e.g., De Bondt and Thaler (1987) and Fama and French (1992).
6

Trends are also visible in stock indexes of US industries and investment styles, and in stock indexes of foreign equity markets. See Chen and De Bondt (2004) and De Bondt (2008b) for details.
8

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

11

securities has exceeded 6%. It is difficult to reconcile the magnitude of this premium with modern asset pricing theory (Mehra and Prescott, 1985) since it implies that the representative investor is exceedingly risk-averse. • Size and calendar effects: Small firms earn anomalous high returns. There is also ample literature on calendar effects. For example, there are curious patterns in equity returns related to weekends, the turn of the month, and the turn of the year (Siegel, 1998; Keim, 1983, 1986; Reinganum, 1983; Roll, 1983). The main point of the above examples is that business fundamentals alone do not explain the structure and dynamics of asset prices. Behavioral finance offers promising, plausible alternative explanations for some of these phenomena. In the next section, we describe some of the key psychological building blocks of the behavioral framework.

IV. Key Building Blocks
Behavioral finance is based on three main building blocks, namely sentiment, behavioral preferences, and limits to arbitrage. By sentiment is meant investor error. Errors originate at the level of the individual but can manifest themselves at the level of the market. Behavioral preferences capture attitudes about risk and return which do not conform with the principles of expected utility theory. In neoclassical finance, rational information traders exploit the behavioral inconsistencies of irrational noise traders, and in so doing lead prices to be efficient. Proponents of behavioral finance suggest that there are limits to the process of arbitrage, and as a result prices need not be efficient. We next describe each of these building blocks in greater detail. Psychology shows that people’s beliefs are often predictably in error. In many cases, the source of the problem is cognitive. That is, the problem is a function of how people think. Some psychological mechanisms have been modeled as heuristic rules of thumb. By and large, heuristics perform well but, sometimes, they lead to systematic error. A few biases in beliefs are described below. • Anchoring is a form of bias where beliefs rely heavily on one piece of information, perhaps because it is was available first, and are not sufficiently adjusted afterward. For instance, investor forecasts may anchor on the price at which they bought a security (De Bondt, 1993; Muradoglu and Onkal, 1994). “Conservatism” is closely related. Investors may place excessive weight on past information relative to new information, i.e., they underreact. • Representativeness is overreliance on stereotypes. Investors who regard recent time-series trends as representative of an underlying process are vulnerable to extrapolation bias. The “law of small numbers” is a related

bias whereby people behave as if the statistical properties of small samples must conform to the properties of large samples. Investor overreaction is partly rooted in representativeness. The “gambler’s fallacy” is also connected to representativeness but leads investors to make unwarranted predictions of reversal. • Availability bias means that investors overweigh information that is easily accessible, e.g., that is easily recalled from memory or that corresponds to a future scenario that is easy to imagine. People are likely to remember events that receive a lot of attention by the media and this influences their behavior (see, e.g., Barber and Odean, forthcoming). • Overconfidence implies that individuals overvalue their knowledge or abilities. It has many consequences. For instance, overconfidence may lead investors to underestimate risk or to overestimate their ability to beat the market. Overconfidence bias may also cause excessive trading. Daniel et al. (1998, 2001) suggest that investors suffer from a combination of overconfidence and self-attribution bias, i.e., people attribute success to their own skills, but blame failure on bad luck. Investor preferences constitute the second key element of financial models. In this regard, there are several behaviorallybased preference frameworks. The best known is prospect theory, developed by Kahneman and Tversky (1979) to describe the manner in which people systematically violate the axioms of expected utility theory. Prospect theory differs from expected utility theory in that probabilities are substituted by decision weights, and the value function is defined over gains and losses, not final wealth.9 Other behavioral preference frameworks include SP/A theory, change of process theory, regret theory, affect theory, and self-control theory. The following list describes some of the most important features of behavioral preferences: • Loss aversion portrays investors’ reluctance to realize losses. Tversky and Kahneman (1992) argue that people weight losses twice as much as gains of a similar magnitude. Unlike what is assumed in neoclassical finance, loss averse investors may be inconsistent towards risk. People may prefer

Fellner (1961) introduces the concept of decision weight to explain ambiguity aversion. Kahneman and Tversky (1979) state: “In prospect theory, the value of each outcome is multiplied by a decision weight. Decision weights are inferred from choices between prospects much as subjective probabilities are inferred from preferences in the Ramsey-Savage approach. However, decision weights are not probabilities: they do not obey the probability axioms and they should not be interpreted as measures of degree or belief.”
9

12

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

to avoid risk in order to protect existing wealth, yet may superior decision (ex post). Regret helps to explain the assume risk in order to avoid sure losses.10 dividend puzzle if, ex ante, investors want to avoid the regret • Mental accounting refers to how people categorize and of having sold shares that later went up in price. Such regrets evaluate financial outcomes (Henderson and Peterson, 1992). may also encourage investors to hold on to loser stocks Shefrin and Thaler (1988) assume that people categorize (Shefrin and Statman, 1985). Koening (1999) argues that wealth in three mental accounts: current income, current investors will bet on good assets, in order to avoid regret, which in turn could possibly wealth, and future income. It trigger some sort of herding is furthermore assumed that Sentiment impacts the prices of all behavior. the propensity to consume is Finally, limited arbitrage greatest from the current assets, and drives the difference plays a crucial role in income account and smallest between what behavioral and behavioral asset pricing. To from the future-income repeat, a basic tenet of modern account. One consequence is neoclassical finance tell us about the finance is that arbitrageurs the tendency to treat a new force prices to converge to relationship between risk and risk separately from existing their true fundamental values. risks, usually called narrow return. 11 Yet, research has uncovered a framing. Narrow framing series of financial market poses dangers. Investors phenomena that do not conform to the notion that full arbitrage may act as if they are risk averse in some of their choices but risk seeking in other choices. Shefrin and Statman (2000) is always carried out. For this reason, behavioral asset pricing develop behavioral portfolio theory in single and multiple models focus on the limits that arbitrageurs face in attempting mental account versions (SMA and MMA). In the SMA to exploit mispricing. Markets are not frictionless because of version, investors integrate their portfolios into a single mental transaction costs, taxes, margin payments, etc. Therefore, the account; in the MMA version, investors prefer securities with actions of noise traders (i.e., traders with biased beliefs, not non-normal, asymmetric distributions that combine downside based on fundamental information) may cause prices to be inefficient. As a result, arbitrage can be risky (Shleifer, 2000). protection (in the form of a floor) with upside potential. • Myopic loss aversion combines time horizon-based Mispricing has been the focus of many studies, e.g., Cornell framing and loss aversion. Investors are more averse to risk and Liu (2001), Schill and Zhou (2001), or Mitchell et al. when their time horizon is short than when it is long (Haigh (2002). and List, 2005). Benartzi and Thaler (1995) argue that the size of the equity premium suggests that investors weigh V. Behavioral Analogues to Neoclassical losses twice as much as gains, and that they evaluate their APV and SDF-based Pricing portfolios on an annual basis. • Self-control refers to the degree to which people can Asset pricing theory and corporate finance are in the control their impulses. Thaler and Shefrin (1981) analyze how process of becoming behavioralized. At the moment, the people exhibit self-control with respect to saving behavior. behavioral approach is somewhat piecemeal, whereas the Shefrin and Statman (1984) develop a theory of dividends neoclassical approach is more coherent and integrated. based on this idea, where mainly elderly investors have a Shefrin (2005, 2008a,b) argues that in the future finance will preference for dividends. Shefrin and Statman (1985) refer combine the best of neoclassical and behavioral elements, to self-control when they explain how investors deal with the thereby presenting a coherent, integrated framework for impulse to hold onto losing investments for too long (see describing how markets are impacted by psychological Lease et al., 1976, for empirical evidence). phenomena. • Regret aversion stipulates that investors may wish to avoid Behavioral asset pricing emphasizes that asset prices reflect losses for which they can easily imagine having made a investor sentiment, broadly understood as erroneous beliefs about future cash flows and risks (Baker and Wurgler, 2007). Sentiment impacts the prices of all assets, and drives the 10 In The Theory of Moral Sentiments, Adam Smith (1759) says that “we difference between what behavioral and neoclassical finance suffer more when we fall from a better to a worse situation than we ever tell us about the relationship between risk and return. In this enjoy when we rise from a worse to a better.” Smith’s observation captures regard, consider the global financial crisis that began in 2008. the modern notion of loss aversion. Academics, media, and policy makers have all contributed 11 In the traditional approach, investors judge a new gamble via its to the question of what caused the crisis. In a New York Times
contribution to total wealth.

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

13

article, Lohr (2008) discussed the failure of financial engineering to incorporate the human element. Notably, the behavioral stochastic discount factor (SDF) approach developed by Shefrin incorporates the human factor into financial engineering. Lohr says that Wall Street analysts did use risk models that correctly predicted how the market for subprime mortgage backed securities would be impacted by a decline in real estate prices. However, analysts attached too low a probability of a major decline in real estate prices. This type of situation is typical of events that take place in a behavioral SDF model, where investors collectively commit errors in their judgments of probabilities, thereby leading some derivatives and their underlying assets to be mispriced. One of the most important points made in behavioral corporate finance is that although the principles taught in traditional corporate finance have great value, psychological obstacles may prevent organizations from putting them into practice (Shefrin, 2005). Many normative aspects of traditional corporate finance remain intact. Yet, they need to be augmented so that there is a narrowing in the gap between what academics preach and what managers do. Tomorrow’s managers should understand why people, including themselves, make mistakes, and how as managers they should deal with market inefficiencies.12 The new approach should be specific, not general, and focus on how to make decisions about capital budgeting, capital structure, mergers and acquisitions, payout policy, and corporate governance. In this regard, Shefrin (2008a,b) introduces the concept of “behavioral adjusted present value.” He begins with traditional adjusted present value (which combines net present value and financing side effects) but adds a component to capture the effects of inefficient prices. Shefrin (2008a, b) suggests that an appropriate starting point for discussing the asset pricing paradigm transition is the book written by John Cochrane (2005). Cochrane’s excellent work is built around the concept of a stochastic discount factor. His approach offers a unified treatment. In particular, the capital asset pricing model (CAPM), FamaFrench multifactor model, and models for the yield curve and option prices all appear as special cases of a general SDF framework. For example, the CAPM corresponds to the special case when the SDF is a linear function of the growth rate of aggregate consumption in the economy. The weakness of the neoclassical SDF approach is that its underlying assumptions are behaviorally unrealistic. Although an extensive discussion is beyond the scope of the present study, a point worth addressing is whether

behavioral assumptions alter the basic neoclassical relationship between the SDF and mean-variance frontier. They do not. What they do is alter the shape of the SDF and the ingredients of mean-variance portfolios. In neoclassical theory, the SDF is monotone declining. However, Aït-Sahalia and Lo (2000) and Rosenberg and Engle (2002) find that, during the first half of the 1990s, the SDF features an oscillating shape that supports the predictions based on behavioral assumptions. Moreover, using survey expectations data, Shefrin (2005, 2008) predicted that the shape of the SDF would change during 2001-2004, with a decline in the left portion displayed in Figure 1. Notably, Barrone-Adesi et al. (2008) report that during 2002-2004 the left portion of the SDF does indeed feature a flat shape.13 Mean-variance analysis is very useful for bringing out the implications of behavioral phenomena for the pricing of all assets. To see how different behavioral and neoclassical mean-variance portfolios can be, consider figure 1. This figure contrasts the equilibrium returns to two mean-variance portfolios, one neoclassical and the other behavioral, as functions of aggregate consumption growth in the economy. The return to a neoclassical mean-variance portfolio is essentially linear, and corresponds to the return from combining the risk-free security and the market portfolios. In contrast, the return to a behavioral mean-variance portfolio oscillates with economic growth, reflecting the impact of investor sentiment. The construction of efficient portfolios under the neoclassical paradigm is done by combining an investment in the risk-free asset and the market portfolio. The theoretical outcome of such combination is known as the two-fund separation theorem (Tobin, 1985).14 Behavioral mean-variance portfolios satisfy the two-fund separation theorem. However, the risky asset used to construct behavioral mean-variance portfolios features the use of derivatives. It is a well-established fact that investors require compensation to assume risk. Risk can take any form in financial markets but, in broad terms, the neoclassical framework focuses on fundamental risk. The behavioral approach adds sentiment risk. Therefore, behavioral risk premiums serve as compensation for bearing both sentiment and fundamental risks. Behavioral risk premiums, like their neoclassical counterparts, will be associated with betas and factor pricing models. To illustrate this point, consider figure 2. This figure displays a mean-variance return pattern whose shape is that of an inverse U. Notably, such a shape is implied

Behavioral corporate finance emphasizes organizational heuristics and biases. Such heuristics and biases were endemic to financial firms involved in the global financial crisis that began in 2008.
12

If investors underestimate the probability of extreme negative events, which is part of the “black swan” phenomenon emphasized by Taleb (2006), then the SDF will typically be upward sloping in its left tail.
13

In the case of leveraged portfolios, the theorem still holds but a negative position with respect to the risk-free asset is held.
14

14
FIGURE 1

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Contrasting the return to a neoclassical mean-variance portfolio and the return to a behavioral mean-variance portfolio, as functions of aggregate consumption growth in the economy.
Mean-Variance Return (Gross) 110%

105%

100%

95%

Neoclassical Efficient MV Portfolio Return

90%

Behavioral MV Portfolio Return
85%

80%

75%

97%

99%

96%

101%

103%

104%

Consumption Growth Rate g (Gross)

FIGURE 2
Special case of Figure 1, when behavioral mean-variance return function has the shape of an inverse-U. This figure also shows that the neoclassical mean-variance return is approximately linear. In the CAPM, the mean-variance function is exactly linear.
Gross Return to Mean-variance Portfolio: Behavioral Mean-Variance Return vs Efficient Market Mean-Variance Return
1.03

1.02

Neoclassical MV Portfolio Return
1.01

Mean-variance Return

1

0.99

Behavioral MV Portfolio Return
0.98

0.97

Return to a Combination of the Market Portfolio and Risk-free Security

0.96

0.95

95.82%

96.64%

97.48%

98.31%

99.16%

100.01%

100.87%

101.74%

102.61%

103.49%

104.38%

105.28%

Consumption Growth Rate g (Gross)

106.19%

106%

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

15

by the work of Dittmar (2002). When the inverse U shape is regret theory, self-control theory, and affect theory.15 Indeed, quadratic, the risk premium for any security can be expressed a series of recent works has identified the limitations of as a function of two factors, the return to the market portfolio prospect theory in explaining the behavior of real world and the squared return to the market portfolio. Models investors.16 There are multiple behavioral explanations for involving squared returns to the market portfolio are momentum, not all mutually consistent.17 In this regard, many associated with the analysis of coskewness. The work of behavioral asset pricing models are eclectic and ad hoc. Some Barone-Adesi and Talwar (1983), Harvey and Siddique models rely on the assumption that prices are set by a (2000), and Barone-Adesi et al. (2004) indicates that representative behavioral investor, even though aggregation coskewness is important in the determination of risk theory indicates that such an assumption is unwarranted. As premiums. Much of the for the winner-loser effect, explanatory power of size, there is no clear explanation as One of the most important points book-to-market equity and to why reversals only appear momentum plausibly derives to occur in January. made in behavioral corporate from coskewness. To be sure, behavioral finance is that although the finance is a work in progress. It is unfinished. Indeed, at the principles taught in traditional VI. Strengths and present time, many researchers Weaknesses corporate finance have great value, refer to “behavioral finance” to describe their work18 but there psychological obstacles may Behavioral research has is no common accepted four major strengths. First, it definition of what it is. prevent organizations from putting has proven itself to be Perhaps, this is not an issue in them into practice. productive. For example, it the long-term. After all, the has led to a series of new main goal of behavioral empirical findings. Examples finance is to behavioralize finance, not to create a separate include over- and underreaction in share prices, the new issues field of scientific study. and stock repurchase puzzles, and the role the stock split The second weakness can be described by analogy. Just as effect. Second, with its focus on the impediments to optimal a study of the economic function of payments and settlements decision-making, behavioral finance brings a pragmatic cannot tell us much about the practical organization of the approach to the study of financial decisions. For instance, payment system (cash vs. credit cards etc.), in the same way, insights from behavioral finance help our understanding of undivided focus on psychological mechanisms (e.g. impulses how to structure the relationship between a firm’s investors and predispositions, or psychophysics) does not allow an and its managers. Certainly, the behavioral approach suits adequate interpretation of economic and financial events. An the professional business school which aims to educate individual is much more than a biological organism; (s)he is managers and to improve their expertise. Third, behavioral also a person, a social-historical creation. Reality is socially finance potentially brings a new type of discipline to social constructed. Philosophers often compare man’s conduct to science research. Discipline fundamentally implies triangulation i.e. the synthesis of data from multiple sources. (“Finance you can believe in” requires more than 15This list is hardly exhaustive. Investors also have preferences which mathematical proof.) The final strength of behavioral finance include issues that go beyond returns, an example of which is socially is simply that it is a stimulating field of scholarship. People responsible investing. See Statman (2008). and money: What can fascinate more? Perhaps the appeal of 16See Hens and Vlcek (2005), Barberis and Xiong (forthcoming), and behavioral finance is that it is social science, but with strong Shefrin (2008) emphasis on both the social and the science. 17 There are at least four separate theories to explain why markets exhibit Behavioral finance also has weaknesses. As mentioned in short-term momentum but long-term reversals. Some psychological the previous section, it lacks the unified theoretical core of explanations, such as Barberis, Shleifer, and Vishny (1998) emphasize neoclassical finance, and can be lacking in discipline. For underreaction. Other psychological explanations, such as Daniel, example, there is no single preference framework to Hirshleifer, and Subrahmanyam (2001) emphasize overreaction. Grinblatt accommodate the features in prospect theory, SP/A theory, and Han (2005) emphasize the disposition effect.
18 Hong and Stein (1998) develop a behavioral model in which some investors rely on fundamental analysis and other investors rely on technical analysis. However, there are no specific psychological elements in their model.

16
that of a stage actor. People enact roles. Their motives, outlook and self-image are shaped by what is expected from them in society. Hence, research in behavioral finance should examine the tangible content of people’s thought processes.19 Evidently, this issue cannot be resolved without reference to social, cultural and historical factors. We need to look more into the content, structure and style of intuitive economic stories. For example, how do Swiss citizens (who in majority rent) think about home ownership, and likewise how do Americans? In general, what sorts of economic arguments (true or false) sound plausible to investors, persuade them and motivate their actions? Investment bankers, client relationship and financial marketing managers, among others, would be interested in answers to these questions. Yet, so far, behavioral finance has little to say. Third, behavioral finance must move beyond the narrow micro-level study of typical “mistakes.” If not, too much behavior remains unintelligible. Yes, US data suggest that CEOs, entrepreneurs and investors tend towards unrealistic optimism – an error with perilous consequences. But, one may ask, what causes over-optimism? Is it context-specific? Does it stem from past personal success? Or is it an incontestable part of the American character? A more fundamental critique is to pose the related question: What is a mistake? Economists take a hard line. Error, they say, is strictly about the contrast between actions that are taken and actions that rationally should be taken in accordance with an individual agent’s costs and benefits. Economists’ chief concern is efficiency. However, the concept of error is elastic. James March and Chip Heath (1994) draw a useful distinction between the economic “logic of consequences” and the more broadly applicable “logic of appropriateness.” Consider, for instance, someone who breaks the rules of etiquette. His norm violations may be embarrassing, perhaps inexcusable, but may also make little practical difference. Still, collective beliefs and norms often make all the difference. For example, aside from efficiency, there are other criteria of economic and financial organization such as sustainable development or equity and fairness. These may be “protected values,” i.e., people reject all trade-offs for money. Finally, there is a disconnect between the emphasis in behavioral research on human frailties and the reality that in many corners of the globe people lead a pretty good life. Why are we collectively so strong, yet as individuals so weak? Why does societal rationality transcend individual rationality?

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Orthodox economic theory places the pinnacle of rationality in the brains of individual people whose self-interest drives market prices.20 It blames social evils on dysfunctional incentives and disarray, mainly in corporate bureaucracies and government. The truth may be nearly the opposite. Rationality and well-being derive from organization, spontaneous or deliberate. Why are institutions so crucially important? The reason is that everyone in society depends on everyone else. We sell 99% of what we produce, we buy 99% of what we consume, and we lead better lives for it. Incessant technological progress, product and service standardization, and economic organization are central. The secret is encapsulated by the motto of the 1933 Chicago World’s Fair: “Science finds, industry applies, man conforms.” Technological artifacts make us smart for several reasons (Norman, 1993). First, technology greatly extends man’s cognitive capabilities. Because we forget, we use a notepad or we access the Internet. Second, technology is coupled with labor specialization. For example, experts make decisions (e.g. in relation to the nation’s supply of electricity) that tens of thousands are incapable of making for themselves. Third, technology embodies knowledge. Few of us know exactly how the watches on our wrists function. Fortunately, we do not need to know. It is enough that we are able to read the time.21 Finally, technological artifacts often allow cheap replication. So, good products or ideas spread quickly. However, people and machines have to work together.. Technology can be easy or difficult to use. Similarly, administrative organization can be effective or ineffective. Smart technology and organization are human-centered (Reason, 1990). In the short run, this is a matter of design, i.e., of pragmatic behavioral research. Over longer periods, it is the outcome of an evolutionary process. To ask about the “logic” of American corporate law or the dashboard of an automobile is a bit like asking who designed the French language, to what purpose and under which specifications. That in modern society the balance between individual and institutional forces has shifted often gets on our nerves. We lament that man must “conform,” that personal freedom is lost when either law regulates what we do or large corporations – e.g. because of network externalities – control our choice options. Yet, man is limited by his brainpower, habits, and conception of purpose. Organization produces

It is difficult to interpret human action without knowing first how people think about a problem. An extravagant illustration, far removed from finance, has to do with the September 2001 attacks in New York. The questions that we would ask in relation to these evil acts are as follows: How did the perpetrators comprehend the world, and how did they understand their self-interest so that they wanted to be suicide-pilots?
19

20 Austrian, institutional and evolutionary economics do not. These economists espouse the private enterprise system but call attention to the fact that its assumed virtues (innovativeness, responsiveness, administrative parsimony) have no solid basis in microeconomic theory.

Occasionally, however, society forgets why some systems or technologies were designed the way they were, and this can be very costly. Recall the Y2K problem.
21

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

17

predictability. This is fundamental. Rules and regulations coordinate society while reducing the individual’s need to think.22 Of course, financial technology is often customerfriendly and performs brilliantly. Take, for instance, the ATMmachine. Still, it is easy to come up with counter-examples. Retirement saving plans and asset allocation tools can be made more effective. The US mortgage debt crisis of 2007-2008 is a gigantic drama from which, one can only hope, the industry will learn. The global wave of financial deregulation that allowed unparalleled growth in the use of complex derivatives may produce even more spectacular failures since quantitative risk models disregard rare events and try to model what arguably cannot be modeled. In every instance, the solution of these problems starts with the recognition that people are human. What is required is “financial ergonomics,” a discipline that engineers financial products and services according to human needs and that optimizes well-being and overall system performance. Behavioral finance holds the potential to create much value for society but it also has a great deal of work to do.

VII. Conclusion
Over the last few decades, our understanding of finance has increased a great deal, yet there are countless questions

begging for answers. On the whole, financial decision making processes in households, markets and organizations remain a grey area waiting for behavioral researchers to shed light on it. A major paradigm shift is underway. Chances are that “the new paradigm” will combine neoclassical and behavioral elements. It will replace unrealistic, heroic assumptions about the optimality of individual behavior with descriptive insights tested in laboratory experiments. Asset pricing theory, we hope, will combine a new realism in assumptions with methods and techniques first developed in neoclassical finance. (Behavioral mean-variance portfolios may explain risk premiums. A unified SDF framework may also provide the basis for behavioral explanations of option pricing, the term structure of interest rates, and other asset prices.) Finally, and more broadly, history requires that economic and financial systems are continually updated, and that they are intelligently reconstructed to meet social changes and to take advantage of technological progress. It is clear that, if academicians are to succeed in understanding financial institutions and actors, and if the agents themselves, as well as policy-makers, want to make wise decisions, they must take into account the true nature of people, that is to say their imperfections and bounded rationality.

References
Aït-Sahalia, Y. and A. Lo, 2000, “Nonparametric Risk Management and Implied Risk Aversion,” Journal of Econometrics 94, 9-51. Baker, M. and J. Wurgler, 2007, “Investor Sentiment in the Stock Market,” Journal of Economic Perspectives 21 (No. 2), 129-151. Barber, B. and T. Odean, 2007, “All that Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors,” Review of Financial Studies, forthcoming. Barberis, N.A. Shleifer, and R. Vishny, 1998, “A Model of Investor Sentiment,” Journal of Financial Economics 49 (No. 3), 307-344. Barberis, N. and R. Thaler, 2003, “A Survey of Behavioral Finance,” in Ed. by G.M. Constandinides, M. Harris and R. Stulz, Handbook of the Economics of Finance Elsevier. Barberis, N. and W. Xiong, 2009, “What Drives the Disposition Effect? An Analysis of a Long-Standing Preference-Based Explanation,” Journal of Finance 64 (No. 2), 751-784. Barone-Adesi, G. and P. Talwar, 1983, “Market Models and Heteroscedasticity of Residual Security Returns,” Journal of Business and Economic Statistics 1 (No. 2), 163-168. Barone-Adesi, G., P. Gagliardini and G. Urga, 2004, “Testing Asset Pricing Models with Coskewness,” Journal of Business and Economic Statistics 22, 474-485.

Behavioral finance also offers a framework for understanding financial market regulation. See Shefrin and Statman (1993b).
22

18
Barone-Adesi, G., R. Engle and L. Mancini, 2008, “A GARCH Option Pricing Model with Filtered Historical Simulation,” Review of Financial Studies 21(No. 3), 12231258. Basu, S., 1977, “Investment Performance of Common Stocks in Relation to their Price-Earnings Ratios: A Test of the Efficient Market Hypothesis,” Journal of Finance 32 (No. 3), 663–682. Benartzi, S. and R. Thaler, 1995, “Myopic Loss Aversion and the Equity Premium Puzzle,” Quarterly Journal of Economics 110 (No. 1), 75-92. Bernard, V., 1992, “Stock Price Reactions to Earnings Announcements,” In Advances in Behavioral Finance, Edited by R.H. Thaler, Russell Sage Foundation. Bernard, V. and J.K. Thomas, 1989, “Post-earnings Announcement Drift: Delayed Price Response or Risk Premium?” Journal of Accounting Research 27 (Supplement), 1-48. Bernard, V. and J.K. Thomas, 1990, “Evidence that Stock Prices do not Fully Reflect the Implications of Current Earnings for Future Earnings,” Journal of Accounting and Economics 13, 305–340. Bartov, E., I. Krinsky and S. Radhakrishnan, 2000, “Investor Sophistication and Patterns in Stock Returns after Earnings Announcements,” Accounting Review 75, 43–63. Black, F., 1986, “Noise,” Journal of Finance 41 (No. 3), 529-543. Chen, H.-L. and W. De Bondt, 2004, “Style Momentum Within the S&P 500 Index,” Journal of Empirical Finance 11(No. 4), 483-507. Cochrane, J., 2005, Asset Pricing, Princeton University Press. Cornell, B., and Q. Liu, 2001, “The Parent Company Puzzle: When is the Whole Worth Less Than One of its Parts?” Journal of Corporate Finance 7, 341-366. Cutler, D., J. Poterba, and L. Summers, 1991, “Speculative Dynamics,” Review of Economic Studies 58 (No. 3), 529– 546. Daniel, K.D., D. Hirshleifer and A. Subrahmanyam, 1998, “Investor Psychology and Security Market Under- and Over-Reactions,” Journal of Finance 53 (No. 6), 18391886. Daniel, K.D., D. Hirshleifer and A. Subrahmanyam, 2001, “Overconfidence, Arbitrage, and Equilibrium Asset Pricing,” Journal of Finance 56 (No. 3), 921-965.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Daniel, K.D., D. Hirshleifer and S.H. Teoh, 2002, “Investors Psychology in Capital Markets: Evidence and Policy Implications”, Journal of Monetary Economics 49 (No. 1), 139-209. De Bondt, W.F.M., 1993, “Betting on Trends: Intuitive Forecasts of Financial Risk and Return,” International Journal of Forecasting 9 (No. 3), 355–371. De Bondt, W.F.M., 2002, “Bubble Psychology,” in W. Hunter and G. Kaufman (eds.), Asset Price Bubbles: Implications for Monetary, Regulatory, and International Policies, MIT Press. De Bondt, W.F.M., 2005, The Psychology of World Equity Markets, Edward Elgar. De Bondt, W.F.M., 2008a, “Stock Prices: Insights from Behavioral Finance,” in Alan Lewis (ed.), The Cambridge Handbook of Psychology and Economic Behavior, Cambridge University Press. De Bondt, W.F.M., 2008b, “Investor Psychology and International Equity Markets,” Revue Economique et Sociale 66 (No. 4), 107-121. De Bondt, W.F.M. and R. Thaler, 1985, “Does the Stock Market Overreact?” Journal of Finance 40 (No. 3), 793808. De Bondt, W.F.M. and R. Thaler, 1987, “Further Evidence on Investor Overreaction and Stock Market Seasonality,” Journal of Finance 42 (No. 3), 557-581. De Bondt, W.F.M. and R. Thaler, 1990, “Do Security Analysts Overreact?” American Economic Review 80 (No.2), 52– 57. Dittmar, R., 2002, “Nonlinear Pricing Kernels, Kurtosis Preference, and Evidence from the Cross Section of Equity Returns,” Journal of Finance 57 (No. 1), 369-403. Dreman, D.N., 1977, Psychology and the Stock Market, Amacom. Dreman, D.N., 1980, The New Contrarian Investment Strategy, Random House. Dreman, D.N., 1995, “Exploiting Behavioral Finance: Portfolio Strategy and Construction,” In Behavioral Finance and Decision Theory in Investment Management, Edited by Arnold S. Wood, AIMR. Fama, E.F., 1965, “The Behavior of Stock Market Prices,” The Journal of Business 38 (No. 1), 34–106.

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

19

Fama, E.F., 1998, “Market Efficiency, Long-Term Returns, and Behavioral Finance,” Journal of Financial Economics 49 (No. 3), 283–306. Fama, E.F. and K. French, 1992, “The Cross-Section of Expected Stock Returns,” Journal of Finance 47 (No. 2), 427–465. Fellner, W., 1961, “Distortion of Subjective Probabilities as a Reaction to Uncertainty,” Quarterly Journal of Economics 75, 670-690. Frazzini, A., 2006, “The Disposition Effect and Underreaction to News,” Journal of Finance 41 (No. 6), 2017-2046 Friedman, M., 1953, “The Case for Flexible Exchange Rates,” Essays in Positive Economics, University of Chicago Press. Froot, K. and E. Dabora, 1999, “How are Stock Prices Affected by the Location of Trade?” Journal of Financial Economics 53 (No. 2), 189-216. Graham, B., 1949, The Intelligent Investor: A Book of Practical Counsel, Harper. Grinblatt, M., and T. Moskowitz, 1999, “The Cross-Section of Expected Returns and its Relation to Past Returns: New Evidence,” UCLA Anderson School, Working Paper No. 24-99. Grinblatt, M., S. Titman and R. Wermers, 1995, “Momentum Investment Strategies, Portfolio Performance and Herding: A Study of Mutual Funds Behavior,” The American Economic Review 85, 1088-1105. Han, B. and M. Grinblatt,. 2005, “Prospect Theory, Mental Accounting and Momentum,” Journal of Financial Economics 78 (No. 2), 311-333. Haigh, M.S. and J.A. List, 2005, “Do Professional Traders Exhibit Myopic Loss Aversion? An Experimental Analysis,” Journal of Finance 60 (No. 1), 523-534. Harvey, C. and A. Siddique, 2000, “Conditional Skewness in Asset Pricing Tests,” Journal of Finance 55 (No. 3), 12631295. Henderson, P. and P. Peterson, 1992, “Mental Accounting and Categorization,” Organizational Behavior and Human Decision Processes 51 (No. 1), 92-117. Hens, T. and M. Vlcek, 2005, “Does Prospect Theory Explain the Disposition Effect?” University of Zurich working paper.

Ikenberry, D., J. Lakonishok, and T. Vermaelen, 1995, “Market Underreaction to Open Market Share Repurchases,” Journal of Financial Economics 39 (No. 2-3), 181–208. Jegadeesh, N. and S. Titman, 1993, “Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency,” Journal of Finance 48 (No. 1), 65-91. Jegadeesh, N. and S. Titman, 2001, “Profitability of Momentum Strategies: An Evaluation of Alternative Explanations,” Journal of Finance 56 (No. 2), 699–720. Johnson, E.J. and D. Goldstein, 2003, “Do Defaults Save Lives?” Science 302 (No. 5649), 1338-1339. Kahneman, D. and A. Tversky, 1979, “Prospect Theory: An Analysis of Decision Making under Risk,” Econometrica 47, 263-291. Keim, D., 1983, “Size-Related Anomalies and Stock Return Seasonality: Further Empirical Evidence,” Journal of Financial Economics 12, 13-32. Keim, D., 1986, “Dividend Yields and the January Effect,” Journal of Portfolio Management 12 (Winter), 54-60. Keynes, M., 1964, “The General Theory of Unemployment, Interest and Money,” Harcourt Brace Jovanovich, reprint of the 1936 edition. Koening, J., 1999, “Behavioral Finance: Examining Thought Processes for Better Investing,” Trust and Investments 69, 17-23. Lamont, O.A. and R.H. Thaler, 2003, “Can the Market Add and Subtract? Mispricing in Tech Stock Carve-outs,” Journal of Political Economy 111 (No. 2), 227-268. Lease, R.C., W.G. Lewellen, and G.G. Schlarbaum, 1976, “Market Segmentation,” Financial Analysts Journal 32, 53-60. Lee, C.M. and B. Swaminathan, 2000, “Price Momentum and Trading Volume,” Journal of Finance 55 (No. 5), 2017–2069. LeRoy, S. and R.D. Porter, 1981, “Stock Price Volatility: A Test Based on Implied Variance Bounds,” Econometrica 49, 97-113. Lohr, S., 2008, “In Modeling Risk, the Human Factor Was Left Out,” New York Times, November 5.

20
Lopes, L., 1987, “Between Hope and Fear: The Psychology of Risk,” Advances in Experimental Social Psychology 20, 255-295. March, J. and C. Heath, 1994, A Primer on Decision Making: How Decisions Happen, Free Press. Michaely, R., R.H. Thaler, and K. Womack, 1995, “Price Reactions to Dividend Initiations and Omissions: Overreaction or Drift?” Journal of Finance 50 (No. 2), 573–608. Mitchell, M., T. Pulvino, and E. Stafford, 2002, “Limited Arbitrage in Equity Markets,” Journal. of Finance 57 (No. 2), 551-584. Muradoglu, G., 1989, “Factors Influencing Stock Demand in Turkey,” Unpublished thesis, Bogazici University, Istanbul, Turkey. Muradoglu, G., 2000, “Turkish Stock Market: Anomalies and Profit Opportunities,” In Security Market Imperfections in Worldwide Equity Markets, Edited by D. Keim, and W. Ziemba, Cambridge University Press. Muradoglu, G., 2002, “Portfolio Managers and Novices’ Forecasts of Risk and Return: Are There Predictable Forecast Errors?” Journal of Forecasting 21 (No. 6), 395416. Muradoglu, G,. A. Salih, and M.Mercan, 2005, “A Behavioral Approach To Efficient Portfolio Formation” The Journal of Behavioural Finance 6 (No. 4), 202-212. Muradoglu, G. and D. Onkal, 1994, “An Exploratory Analysis of Portfolio Managers’ Probabilistic Forecasts of Stock Prices,” Journal of Forecasting 13 (No. 7), 565–578. Nicholson, F., 1968, “Price Ratios in Relation to Investment Results,” Financial Analysts Journal 24 (No. 1), 105-109. Norman, D.A., 1993, Things That Make Us Smart, Perseus Books. Odean, T., 1998, “Are Investors Reluctant to Realize Their Losses?”Journal of Finance 53 (No. 5), 1775-1798. Odean, T., 1999, “Do Investors Trade Too Much?” American Economic Review 89 (No. 5), 1279-1298. Poterba, J.M. and L.H. Summers, 1988, “Mean Reversion in Stock Returns: Evidence and Implications,” Journal of Financial Economics 22, 27–59.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Reason, J., Human Error, Cambridge University Press, Cambridge, 1990. Reinganum, M., 1983, “The Anomalous Stock Market Behavior of Small Firms in January: Empirical Tests for Tax-Loss Selling Effects,” Journal of Financial Economics 12, 89-104. Roll, R., 1983, “The Turn-of-the-Year Effect and the Return Premia of Small Firms,” Journal of Portfolio Management 9 (Winter), 18-28. Roll, R., 1984, “Orange Juice and Weather,” American Economic Review 74 (No. 5), 861–880. Roll, R., 1988, “R2,” Journal of Finance 43 (No. 3), 541– 566. Rosenberg, J. and R. Engle, 2002, “Empirical Pricing Kernels,” Journal of Financial Economics 64 (No. 3), 341372. Rosenthal, L., and C. Young, 1990, “The Seemingly Anomalous Price Behavior of Royal Dutch/Shell and Unilever N.V./PLC,” Journal of Financial Economics 26 No. 1), 123–142. Rouwenhorst, K.G., 1998, “International Momentum Strategies,” Journal of Finance 53 (No. 1), 267–284. Rouwenhorst, K.G., 1999, “Local Return Factors and Turnover in Emerging Stock Markets,” Journal of Finance 54 (No. 4), 1439–1464. Scruggs, J.T., 2007, “Noise Trader Risk: Evidence from the Siamese twins,” Journal of Financial Markets 10 (No. 1), 76-105. Shefrin, H., 2001a, Behavioral Finance, Edward Elgar. Shefrin, H., 2001b, “Behavioral Corporate Finance,” Journal of Applied Corporate Finance 14 (No. 3), 113-124. Shefrin, H., 2002, Beyond Greed and Fear, Oxford University Press. Shefrin, H., 2005, Behavioral Corporate Finance, McGrawHill. Shefrin, H., 2008a, A Behavioral Approach to Asset Pricing, Elsevier Academic Press (2nd Ed.). Shefrin, H., 2008b, “Risk and Return in Behavioral SDFBased Asset Pricing Models,” Journal of Investment Management 6 (No. 3), 1-19.

DE BONDT, MURADOGLU, SHEFRIN, & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS?

21

Shefrin, H. and M. Statman, 1984, “Explaining Investor Preference for Cash Dividends,” Journal of Financial Economics 13, 253-282. Shefrin, H. and M. Statman, 1985, “The Disposition to Sell Winners too Early and Ride Losers too Long: Theory and Evidence,” Journal of Finance 40 (No. 3), 777-790. Shefrin, H. and M. Statman, 1993a, “Behavioral Aspects of the Design and Marketing of Financial Products,” Financial Management 22 (No. 2), 123-134. Shefrin, H. and M. Statman, 1993b, “Ethics, Fairness and Efficiency in Financial Markets,” Financial Analysts Journal 49 (No. 6), 21-29. Shefrin, H. and M. Statman, 2000, “Behavioral Portfolio Theory,” Journal of Financial and Quantitative Analysis 35 (No. 2), 127-151. Shefrin, H. and R.H. Thaler, 1988, “The Behavioral LifeCycle Hypothesis,” Economic Inquiry 26 (No. 4), 609643. Schill, M.J. and C. Zhou, 2001, “Pricing an Emerging Industry: Evidence from Internet Subsidiary Carve-outs,” Financial Management 30 (No. 3), 5-34. Shiller, R.J., 1981, “Do Stock Market Prices Move too Much to be Justified by Subsequent Changes in Dividents?” American Economic Review 71, 421–436. Shiller, R.J., 1993, Market Volatility, MIT Press. Shleifer, A., 2000, Inefficient Markets, Oxford University Press. Siegel, J.J., 1998, Stocks for the Long Run, McGraw-Hill.

Slovic, P., 1972, “Psychological Study of Human Judgment: Implications for Investment Decision Making,” Journal of Finance 27 (No. 4), 779-799. Smith, A., 1759, The Theory of Moral Sentiments, Liberty Fund (published in 1982). Statman, M., 2008, “The Expressive Nature of Socially Responsible Investors,” Journal of Financial Planning, (February). Taleb, N., 2007, The Black Swan: The Impact of the Highly Improbable, New York, Random House. Thaler, R., 1993, Advances in Behavioral Finance, Russell Sage Foundation. Thaler, R. and H. Shefrin, 1981, “An Economic Theory of Self Control,” Journal of Political Economy 89 (No. 2), 392-410. Thaler, R.H. and C.R. Sunstein, 2008, Nudge: Improving Decisions about Health, Wealth, and Happiness, Yale University Press. Tobin, J., 1958, “Liquidity Preference as Behavior Towards Risk,” Review of Economic Studies 25 (No. 67), 65-86. Tversky, A. and D. Kahneman, 1974, “Judgment under Uncertainty: Heuristics and Biases,” Science 185 (No. 4157), 1124-1131. Tversky, A. and D. Kahneman, 1992, “Advances in Prospect Theory: Cumulative Representation of Uncertainty,” Journal of Risk and Uncertainty 5 (No. 4), 297-323.

22

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

The Effects of Institutional Risk Control on Trader Behavior
Ryan Garvey and Fei Wu

We examine how institutional risk control mechanisms influence proprietary stock trader behavior. When traders are forced to liquidate their inventory at a pre-designated time, they often hold onto their losing trades until the very last moment. We find that the difference between losing and winning round-trip holding times systematically widens leading up to an inventory liquidation deadline and trading becomes less driven by trading practices and more induced by the firm’s control mechanism as the deadline draws near. When trade price is heavily controlled yet trade size isn’t, we find that the difference between losing and winning roundtrip holding times systematically widens with trade size. This result suggests traders increase their risk-taking in areas where institutional control mechanisms are weaker. Our findings highlight the difficult balancing act firms face with getting market professionals to realize their losses without impeding their trading strategies.

 A considerable amount of research has uncovered behavioral biases among financial market participants.1 In order to circumvent these biases and help employees make
Ryan Garvey is an Associate Professor of Finance at Duquesne University in Pittsburgh, PA. Fei Wu is a Senior lecturer in Finance at Massey University in Palmerston North, New Zealand. We would like to thank the Editor, Ramesh Rao, and an anonymous referee for helpful comments and suggestions on a prior draft. We are also grateful to executives at the US Securities Firm for providing proprietary data. See, for example, Odean (1998), Grinblatt and Keloharju (2001), and Coval and Shumway (2005).
1

better decisions, financial institutions implement risk control mechanisms with their employees who make trade decisions with institutional capital. The contributions of this study are: 1) to examine how effective control mechanisms are at mitigating psychological trading biases, and 2) to examine how employees respond to control mechanisms. While our results are useful for firms implementing or planning to implement risk control mechanisms, our results also provide a step forward for the academic literature. Much of the academic literature has been devoted to uncovering behavioral biases in various settings and among different types of market participants. Uncovering these biases is, of course, a necessary first step. Yet, some trading biases are well known and firms have been grappling with ways to control them for decades. Despite this, there is very little research that examines trader behavior in settings where prevention techniques are actively implemented by firms to get their employees to recognize and refrain from biases in their decisions.2 In our paper, we examine such a setting. We analyze how proprietary stock traders, who work on behalf of a National Securities Dealer, react to institutional control mechanisms that are primarily intended to get them to realize their trading losses. It is well known that market professionals often have difficulty coming to terms with their losses. Consequently, they have a tendency to hold their losing trades too long because they want to recover from their losses. This desire to get even is quite persuasive in financial market settings, and it is inevitably ingrained in many of the everyday decisions that traders make. Indeed, some of the greatest
Statman and Caldwell (1987) discuss risk control and behavioral biases in the context of capital budgeting decisions.
2

22

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

23

trading losses of all time have occurred because traders were their positions by a predetermined time (i.e. the end of the simply unwilling to take a loss, so they gambled in an attempt trading day) and our main focus is on how traders holding to recover from their loss.3 times and performance vary across the day leading up to the In order to help traders come to terms with their losses, inventory liquidation deadline. our sample firm implemented several control mechanisms. We find that the difference between losing and winning The most binding of these control mechanisms was that they round-trip holding times systematically rises throughout the required traders to liquidate day and that it rises to its their inventory by the end of highest level just prior to the We analyze how proprietary stock the trading day in order to inventory liquidation deadline. ensure loss realization. The Moreover, trading performance traders, who work on behalf of a firm implemented other significantly declines as the National Securities Dealer, react to control mechanisms including liquidation deadline draws institutional control mechanisms an emphasis on price control. near. Our sample traders often They implemented training have difficulty realizing certain that are primarily intended to get sessions that often stressed the losses and they have a tendency them to realize their trading losses. dangers of holding losses too to hold onto them until the very long. For example, the firm last moment. While the firm’s cites in their training manual that a trader’s inability to take a efforts do not statistically eliminate a trader’s tendency to loss is the number one reason why traders fail. And they hold losing trades longer than winning trades, they clearly employed a trading manager who closely monitored trading do have an influence on trader behavior. activity throughout the day. The firm even hired an on-site Inventory liquidation requirements ensure losses get 4 psychologist who was readily available to meet with traders. realized, but firms (traders) also rely on price control Despite all of these control measures, traders still appear mechanisms to do the job. While trade price is often heavily to have difficulty coming to terms with their losses. We find controlled in institutional trading settings, trade size usually that, on average, traders hold their losing trades significantly is not. Institutional market participants trade in large trade longer than their winning trades, which is consistent with the sizes and their ability to execute these large trade sizes in behavior underlying the disposition effect (see Shefrin and their entirety is often driven by market conditions. Thus, Statman, 1985). These longer holding times coincide with professional traders need flexibility with respect to trade size. lower performance. While prior studies document that While we find that traders adhere to a highly disciplined professional traders have a tendency to hold their losing trades approach with respect to their exit prices, trade size longer than their winning trades (see, for example, Locke considerably varies and traders let their losses run longer on and Mann (2005) and Garvey and Murphy (2004)), the larger size trades. Consequently, they are more unprofitable professional traders observed in prior studies were not when they trade in larger trade sizes. required to close out of all of their positions by a These traders were constantly being drilled on the dangers 5 predetermined time. In our setting, traders are forced to exit of holding losses too long, but they tended to hold onto their losses despite the warning. Their tendency to hold losses for longer periods of time resulted in lower performance. If the 3 firm did not require traders to close out of their positions by One of the more notable cases, or largest losses resulting from this behavior, occurred with Nicholas Leeson. Mr. Leeson incurred over $1.4 billion in the end of the day, or if they did not implement any control trading losses in 1995, which led to the demise of his employer, 232 yearmeasures, presumably losses would be held for even longer old Barings PLC. periods of time and performance would be a lot worse. This 4 We sat in on the firms training sessions for traders, reviewed trader manuals, had several discussions with management and traders, and is why it is so important for financial firms to implement observed traders trade so that we could better prepare this paper. control mechanisms. 5 Our results highlight the complexities involved with For example, Garvey and Murphy (2004) examine proprietary stock traders who mainly offset their positions intraday, but the traders can and do hold implementing optimal risk control mechanisms to circumvent positions overnight. Locke and Mann (2005) assume commodity traders traders’ aversion to realizing losses. If firms implement end each day flat when determining trader gains and losses. However, the authors do not report, or are not aware of, any mandatory time period in control mechanisms that are too stringent, they are likely to
which traders are required to liquidate their inventory.

24
conflict with traders’ overall strategies and objectives because, as our research shows, trader behavior is heavily influenced by control mechanisms. On the other hand, if firms do nothing they open themselves up for considerable risks. Thus, our findings imply that firms do need to implement control mechanisms, but they need to be very careful with how they enforce this because traders align their strategies with control mechanisms. Because psychological biases are so intertwined in many people’s decision-making processes, firms are unlikely to eliminate these biases, but they can lessen the damage caused by them.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

I. Related Research
Kahneman and Tversky’s (1979) prospect theory provides a descriptive framework for decision-making under risk. A central theme in their research is the role of loss. Much of the behavioral finance literature focuses around how market participants make decisions when they are confronted with the prospect of a loss. Shefrin and Statman (1985) were the first to apply prospect theory to a financial market setting. They also placed prospect theory in a wider theoretical framework that includes mental accounting, regret aversion, and self-control. These factors together help explain theoretically why traders have a tendency to hold their losing trades much longer than their winning trades, a behavior that is commonly known as the “disposition effect”. Traders, who exhibit behavior that is consistent with the disposition effect, think about stock purchases within separate mental accounts (see Thaler, 1985) then apply prospect theory decision rules to each mental account. The disposition effect has proven to be quite pervasive in US markets.6 For example, research shows that individual investors (e.g., Barber and Odean (1999) and Odean (1998)), mutual fund managers (e.g., Frazzini (2006) and Scherbina and Jin (2005)), and professional traders (e.g., Garvey and Murphy (2004) and Heisler (1994)) all exhibit signs of this behavior. Some of the more recent studies identify individual trader characteristics that are correlated with the disposition effect. For example, Dhar and Zhu (2007) find that individuals’ income level, occupational status, etc. are important factors in indicating who is more susceptible to the disposition effect.

While most researchers examine traders’ unwillingness to take a loss through holding times and focus on decisionmaking in a single period setting, some other studies have looked elsewhere. They examine a trader’s reluctance to realize losses using other risk measures and focus on decisionmaking in a multi-period setting. For example, Coval and Shumway (2005) and Garvey et al. (2007) find traders who have experienced prior morning losses engage in subsequent afternoon risk-taking as measured through increased trading activity, larger size trading, etc. These findings are consistent with the same behavioral tendency that leads traders to hold their losing trades too long. Like much of the prior research, we examine trader resistance to loss realization through holding-time decisions, and we examine trader holding-time decisions in a singleperiod setting. Our motivation is not to examine if traders suffer from the get even behavior that underlies the disposition effect, but rather our motivation is to examine how (if) traders respond to institutional control mechanisms to prevent this behavior.

II. Data
Our data originates from a National Securities Dealer. The firm had several trading operations and our focus is on the firm’s proprietary trading operation for US equities. The data covers June 3, 2002 through May 30, 2003. During this oneyear period, the US stock markets were open for 251 days. In total, the 150 traders combined to execute 2.5 billion shares through 1.3 million transactions on 693 securities. For every transaction, the data reveals the identity of the trader, the execution time, the type of trade (marketable versus limit order), the action taken (buy, sell, short, and cover), the volume, the price, the market where the order was sent, the contra party on the trade (if given), the location of the trader (the traders were located in five branch offices), and various other trade execution information. Each trader’s sole objective was to generate intraday trading profits utilizing firm capital. Consequently, the traders trade often and they also trade in large trade sizes. The average trader executes 75 per day and the average executed trade size is for 1,925 shares. This average trade size is more than three times the average trade size in US equity markets.7 Trading activity is concentrated in certain stocks (mostly
The average trade size for NYSE (Nasdaq) stocks was 488 (579) shares in 2003 (Source: NYSE and Nasdaq data).
7

Researchers have also found strong evidence of the disposition effect in Finland (e.g. Grinblatt and Keloharju, 2001), Israel (e.g. Shapira and Venezia, 2001), China (e.g. Feng and Seasholes, 2005), Taiwan (e.g. Barber et al., 2007) and other countries.
6

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

25

Nasdaq) on certain days, often accounting for a sizeable most intriguing aspect of this particular setting is the firm’s portion of a particular stock’s trading volume.8 While the efforts to get traders to realize their trading losses. The firm traders trade often, they also set market prices often. is not alone with respect to their efforts in this regard. Getting Approximately 65% of their trades provide liquidity, while traders to take losses is a common problem that securities 35% of their trades take liquidity. The traders set market firms and their risk managers constantly grapple with. prices in the various trading venues which US equities trade in.9 III. Empirical Results To get an idea of the Traders hold losses longer than gains, trading intensity difference A. Methodology but these holding time patterns do not between institutional and The objective of our study retail market participants, remain constant throughout the day. is to directly examine how consider a sample of retail The difference in holding times traders react to institutional brokerage accounts control mechanisms that are studied by Barber and between losing and winning round10 primarily intended to get Odean (2000). They trips systematically rises throughout them to accept their losses. analyze the trading Specifically, we measure the the day and it dramatically increases behavior and performance effects of institutional control of retail market in the moments just prior to the firm’s mechanisms on traders’ participants, who trade holding-time decisions. mandatory close-out period. through a US discount Round-trip performance and brokerage firm. In total, holding-time measures are not included in the raw transaction 66,465 households execute 1,969,701 stock trades over a sixyear period ending in December, 1996. Our 150 traders data. In order to determine the gains and losses for each execute 1,316,334 stock trades over just a one-year period. round-trip, we use an intraday round-trip matching procedure Institutional market participants dominate the trading similar to the one used in Garvey and Murphy (2005). We landscape in US equity markets, yet much of the academic pair off opening trades with the subsequent closing trade(s) literature examining trader decision-making in equity markets in the same day. The traders did not always open and close is focused on retail market participants. While some studies positions with two trades. A trader could combine a closing do examine institutional trading (see, more recently, Conrad, transaction with an opening transaction, or they could lay off Johnson, and Wahal, 2002), this literature largely focuses on part of an open position. Regardless of whether trades transaction costs, and prior studies do not examine how opened, closed, or open and closed a position simultaneously, we searched forward in time each day until the opening institutional traders respond to risk control mechanisms. Our sample traders received continual training on various position was closed out, and we kept track of execution times, trading strategies from the firms’ management, yet they had accumulated inventory, and corresponding prices paid or considerable freedom with selecting stocks to trade. The received. The matching procedure creates 730,417 roundcompensation of the traders is solely tied to their trading trip trades from the 1.3 million trades. We then calculate out performance. Thus, the traders had a clear incentive to the corresponding holding time for each round-trip. In order maximize their trading performance. For our purposes, the to do this, we calculate the holding time between the intraday opening and closing transaction(s). If a trader accumulates inventory before they eventually close out of their position, 8 we use a weighted average between the various opening The only stocks traded every day were Sun Microsystems (SUNW) and JDS Uniphase (JDSU). The traders accounted for 1.5% and 3.3% of the positions.11
annual share volume of SUNW and JDSU respectively. SUNW and JDSU are two of the more actively traded US stocks.
9

The traders often traded on The Island ECN, which reported its trades through the Cincinnati Stock Exchange. See Nguyen, Van Ness, and Van Ness (2004) for a discussion on the reporting of Island trades on the Cincinnati Stock Exchange.
10

11

The data has been used in several studies in finance literature.

For example, suppose a trader opens up a 2,000 share position of Yahoo at 10:30:00 a.m. and then purchases another 4,000 shares of Yahoo at 10:30:10 a.m. If the next trade were a sell of 6,000 shares of Yahoo at 10:30:20 a.m., the holding time on the round trip trade is 13.33 seconds (1/3 * 20 seconds + 2/3 * 10 seconds).

26
From the matching procedure, there are 290,248 winning round-trips (gross round-trip trading profits above $0), 209,271 losing round-trips (gross round-trip trading profits below $0), and 230,898 break-even round-trips (gross roundtrip trading profits equal to zero). The frequency of breakeven round-trips highlights how focused these traders are on their trade purchase price (the reference point). The traders do not hold their open positions for long. And when they enter into a position on one side of the market, they generally seek to quickly offset their position by trading on the opposite side of the market. For example, the mean holding time per round-trip is 780 seconds and the median holding time per round-trip is 205 seconds. The sample period we observe was a difficult time to trade US equities. Many securities firms were reporting steep losses on their equity trading desks.12 Our firm was not immune to these difficult trading conditions, yet the traders did experience many profitable (and unprofitable) round-trips which enable us to examine their holding time decisions in both the domain of gain and loss.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

B. Do Institutional Risk Control Mechanisms Eliminate Behavioral Biases?
Table I provides information on the overall holding time difference between winning and losing round-trip trades. Despite the firm’s efforts to get traders to realize their losses in a timely fashion, traders hold their losing round-trip trades considerably longer than their winning round-trip trades (note that the firm’s control mechanisms were in place over our entire sample period). On average, traders hold their losing round-trip trades for 1,274 seconds and their winning roundtrip trades for 568 seconds. The difference of 706 seconds is statistically different from zero at the 1% level. The magnitude of the overall holding time difference is rather surprising given the firm’s continual efforts to get traders to realize their trading losses. It would be interesting to see how this result would change if the firm did not engage in any efforts to get traders to realize their losses. Presumably, traders would hold their losses for even longer periods of time. In order to check the robustness of our initial result, we examine the holding time differences for each individual trader. This allows us to see if certain traders are driving our
For example, the revenues of broker dealers with a Nasdaq market making operation fell over 70% during 2000-2004 (GAO, 2005). Our firm had a Nasdaq market making operation.
12

overall result. On a mean holding time basis, 145 out of 150 traders hold their losing round-trip trades longer than their winning round-trip trades (135 differences are statistically significant). On a median holding time basis, 146 out of 150 traders hold their losing round-trip trades longer than their winning round-trip trades (139 differences are statistically significant). The individual trader results coincide with the aggregate trader results.13 In our setting, holding positions for longer periods of time is generally undesirable. The objective of the traders is to rapidly enter and exit positions in order to profit from small price changes. The trader’s information is short-lived, so when traders’ open positions are held for extended periods of time, it is a good indicator that the position has moved against the trader and they are primarily holding it in order to recover from the loss. In Figure 1, we segregate the roundtrip profits into five holding time categories. Round-trip trading profits systematically decline with longer holding times. When traders hold their trades under (over) five minutes they are profitable (unprofitable). Trader profits are highest when they hold their trades for under one minute, and trader profits are lowest when they hold their trades for more than 15 minutes.14 These results highlight the trader’s short-term strategies. They also show how important holding times are for a trader’s success. A professional trader’s decision to hold trades for slightly longer periods of time could make the difference between being profitable or unprofitable on an overall basis.

C. How Does Inventory Control Influence Trader Behavior?
While the firm’s efforts do not appear to eliminate biases from traders’ decisions, their risk control measures do have an influence on trader behavior. This influence is quite strong in the moments just prior to the inventory liquidation deadline, which highlights how deadline effects coincide with a trader’s resistance to realize losses. Table II and Figure 2 provide information on round-trip holding times across the day. Traders hold losses longer than gains, but these holding time patterns do not remain constant throughout the day. The difference in holding times between losing and winning round-

The average round-trip holding time among traders ranges from 122 to 3,176 seconds.
13

The average round-trip gain is $13.70 and the average round-trip loss is $19.23. The absolute trading profit difference is statistically significant from zero at the 1% level.
14

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

27

Table I. Round-Trip Holding Times
This table reports the mean and median holding times for winning and losing round-trip trades. Results are reported on an aggregate basis and on an individual basis. The results are based on the trading records of 150 proprietary stock traders, who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period. The traders only traded US equities. Winning (losing) round-trips have a gross trading profit above (below) zero and break-even round-trips are omitted. Holding times are calculated in seconds.

Panel A. Aggregate Results
Mean Holding Time
Losing round-trips Winning round-trips Difference 1,274 568 706*

Median Holding Time
377 166 141*

Panel B. Individual Results
Mean
Number of traders who hold their losses longer than their gains Number of traders who hold their losses longer than their gains *Significant at the 0.10 level. 145 135*

Median
146 139*

Figure 1. Performance and Holding Times
This figure displays the average round-trip trading profit for five holding time categories. The mean (median) holding time is 780 (205) seconds. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period.

Average Round-trip Profit and Holding Times
$6.00 $4.00 $2.00 $0.00 ($2.00) ($4.00) ($6.00) Less than 1-2 1 minute minutes 2-5 minutes 5-15 More than minutes 15 minutes

28
Table II. Trading by Time of Day

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

This table reports traders’ trading activity and the number (percentage) of round-trips exceeding the overall mean holding time (13 minutes) for each half-hour period. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period. The 1,428 round-trip trades executed before the open (9:30 a.m.) or after the close (4:00 p.m.) are omitted.

Intraday Time

Number of Round-Trips

Percentage of Round-Trips

Number of RoundTrips above Mean Holding Time
2,093 9,680 12,128 13,415 13,485 13,027 13,094 12,777 12,390 12,604 13,124 13,051 16,625

Percentage of Round-Trips above Mean Holding Time
3.4% 12.8% 18.2% 21.7% 23.7% 25.8% 28.0% 27.1% 25.3% 24.4% 24.6% 24.6% 31.0%

9:30-10:00 a.m. 10:00-10:30 a.m. 10:30-11:00 a.m. 11:00-11:30 a.m. 11:30-12:00 a.m. 12:00-12:30 p.m. 12:30-1:00 p.m. 1:00-1:30 p.m. 1:30-2:00 p.m. 2:00-2:30 p.m. 2:30-3:00 p.m. 3:00-3:30 p.m. 3:30-4:00 p.m.

62,267 75,921 66,816 61,874 56,787 50,537 46,848 47,207 49,055 51,597 53,419 53,107 53,554

8.5% 10.4% 9.1% 8.5% 7.8% 6.9% 6.4% 6.5% 6.7% 7.1% 7.3% 7.3% 7.3%

Figure 2. Winning vs. Losing Round-trip Holding Times by Time of Day
This figure displays the average holding time for winning and losing round-trips throughout the trading day. Winning round-trips have a gross trading profit above zero while losing round-trips have a gross trading profit below zero. Holding times are calculated in seconds. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period.

3500 3000 2500

Average Winning vs. Losing Holding Times: Time of Day

Seconds

2000 1500 1000 500 0

. . . . . . . . . . . . . a.m 0 a.m 0 a.m 0 a.m 0 a.m 0 p.m 0 p.m 0 p.m 0 p.m 0 p.m 0 p.m 0 p.m 0 p.m 0 0:0 0:3 1:0 1:3 2:0 2:3 1:0 1:3 2:0 2:3 3:0 3:3 4:0 - 1 0 -1 0 -1 0 -1 0 -1 0 -1 :30 - :00- :30- :00- :30- :00- :300 1 1 2 2 3 3 9:3 10 :0 10 :3 11 :0 11 :3 12 :0 12
Winning round-trips Losing round-trips

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

29

trips systematically rises throughout the day and it dramatically increases in the moments just prior to the firm’s mandatory close-out period. The results indicate that traders often hold losses up until the very last moment before they are forced to realize them. The trader’s behavior is not a desirable reaction to what the firm is trying to accomplish, but the liquidation requirement appears necessary. Without this control mechanism in place, these traders’ reluctance to take losses would have most likely resulted in larger trading losses. In Table III, we report the gross round-trip trading profits for each half hour of the trading day. Trading profits are highest in the initial opening period, or the period which is furthest away from the liquidation deadline, and statistically different from zero at the 1% level. Trading profits steadily decline until noon. Trading profits are positive after 12:00 p.m., but they steadily decline again until the close of trading. The sharp decline in traders’ profits in the moments just before the close, along with the holding time patterns, indicates that trading is significantly driven by the firm’s control mechanism. The average round-trip profit in the last 30 minutes is -$2.56, whereas the average round-trip profit at other intraday times is $0.18. The average round-trip trading profit in the closing 30 minutes is 1,522% less than the average round-trip trading profit at other intraday times. The end-ofday trading losses can be broken down further. While the traders lost $136,934 in the final 30-minute period, approximately 63% of this occurred in the final 5 minutes of the trading day (note that they lost money in each 5 minute interval in the final 30 minute period). The magnitude of the losses that occurred in each five minute interval is highlighted in Figure 3. Trading profits most likely continue to decline until noon because traders often break for lunch. When traders leave their trading terminals, they usually close out of their positions. For many traders, the lunch period serves as another period for realizing losses. However, the midday close-out period is not binding like the end-of-the-day period is, and losses are generally held for shorter periods of time leading up to the midday period. This is why losses are far more pronounced at the close of the day than they are at the middle of the day. While the end of day closeout period induces trade and forces traders to realize their losses, it is interesting to theorize about what would occur if the firm did not have this control mechanism in place. If the firm allowed traders to hold their positions overnight, would this give traders greater flexibility

with implementing their trading strategies and subsequently improve performance? Or, would removing the liquidation deadline result in traders holding their losses for significantly longer periods of time resulting in catastrophic losses? While it is not possible to definitively answer these questions from our available data, we compute a hypothetical performance measure assuming some trades were held overnight. Because trading in the very last moments of the day appears highly driven by the firm’s control mechanism, we recalculate trading profits for positions closed out in the final 15 minutes of the trading day. Contrary to using the round-trip trade price at the end of the day to determine profits, we assume traders closed their positions at the opening price on the following (trading) day. In order to do this, we obtained opening price data on sample stocks traded from the Center for Research in Security Prices (CRSP) database and recalculated trading profits for positions closed out in the final minutes of the day. There are 24,332 round-trip observations, and the average closing position is for 1,674 shares. Under the adjusted closing price, the average round-trip trading profit declines from -$3.95 to -$4.17 (note that a few anomalous observations are dropped from the analysis). While the firm’s liquidation requirement imposes a constraint on traders, most of the losing positions traders were forced to realize at the end of the day continued to decline in value into the next day of trading. Our firm’s inventory liquidation requirement is not unique. Many securities firms require their market making, proprietary, arbitrage, and other types of traders to end the day flat, or they significantly restrict traders ability to accumulate inventory from day to day.15 If traders at other firms exhibit behavioral tendencies similar to these traders, than our results provide insight into some factors that drive intraday order flow patterns. Researchers have long known that intraday trading activity in US equity markets exhibits a U-shaped pattern across the main trading hours (9:30 a.m. to 4:00 p.m.). The Admati and Pfleiderer (1988) theory has served as a prominent explanation for these intraday volume patterns. While our traders’ intraday trading activity closely resembles a U-shape pattern, there are competing factors
There are many self-employed (retail) traders who adhere to this rule on a self-imposed basis. In the US, these retail day traders typically trade through direct access brokers. Bear Stearns finds that the more active traders (25+ trades per day) who trade through direct access firms account for approximately 40% of Nasdaq/NYSE trading volume (Goldberg and Lupercio, 2004). Thus, retail and institutional traders, who often end the trading day flat, account for a very large percentage of overall daily trading volume in US equity markets.
15

30
Table III. Performance by Time of Day

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

This table reports trading performance, average round-trip trade size and average round-trip holding time for each half-hour period. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period. The 1,428 round-trip trades executed before the open (9:30 a.m.) or after the close (4:00 p.m.) are omitted.

Intraday Time

Total Trading Profits
$93,730 $34,227 $13,245 $12,488 -$11,718 $22,187 $12,656 $2,846 $5,004 -$11,416 -$19,853 -$32,833 -$136,934

Average RoundTrip Trade Size

Average RoundTrip Trading Profit
$1.51*** $0.45*** $0.20* $0.20* -$0.21* $0.44*** $0.27*** $0.06 $0.10 -$0.22* -$0.37*** -$0.62*** -$2.56***

Average Holding Time per Round-Trip

9:30-10:00 a.m. 10:00-10:30 a.m. 10:30-11:00 a.m. 11:00-11:30 a.m. 11:30-12:00 a.m. 12:00-12:30 p.m. 12:30-1:00 p.m. 1:00-1:30 p.m. 1:30-2:00 p.m. 2:00-2:30 p.m. 2:30-3:00 p.m. 3:00-3:30 p.m. 3:30-4:00 p.m.

1,787 1,747 1,763 1,771 1,782 1,757 1,740 1,749 1,713 1,655 1,674 1,698 1,717

187 341 475 591 695 805 922 982 906 924 973 1,084 1,645

***Significant at the 0.01 level. *Significant at the 0.10 level.

Figure 3. Trading Losses in the Final 30 Minutes
This figure reports the percentage of the overall amount that was lost in the final 30 minutes of the trading day (see Table 3) across each five-minute category. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period.

Percentage of Losses in Closing Minutes
70% 60% 50% 40% 30% 20% 10% 0%
. . . . . . p.m p.m p.m p.m p.m p.m :55 :35 :45 :40 :50 :00 0-3 0-3 5-4 5-3 0-3 5-3 3:3 3:5 3:4 3:5 3:4 3:3

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

31

inducing this pattern across the day. Trading at the open, on the firm trained the traders to adhere to a disciplined exit average, appears to be motivated by short-term information strategy (e.g., use of stop loss mechanisms) to ensure loss (i.e. gross round-trip trading profits are statistically different realization. Stop loss mechanisms can be employed either from zero). Trading activity rises in the second half of the explicitly (attached with the opening order) or through a selfday, but this time the rise in trading activity corresponds with imposed rule. The firm attempted to monitor trader exit prices a decrease in performance. The rise in trading activity in the through the trading manager and the traders uniformly exited second half of the day, on average, seems driven more by the most, but not all, of their positions within a very tight pricing firm’s risk control mechanism rather than their traders’ normal range. In Table IV, we report the distribution of round-trip trading practices. price changes for both winning While our findings might be and losing round-trips. The While losing round-trips are useful for providing insight into median price change for both held considerably longer than intraday order flow patterns, winning and losing round-trips is they are also potentially useful one cent.17 winning round-trips for each in understanding why trading The traders prefer trading in trade size category, the activity levels may predictably larger trade sizes in order to rise or decline on certain days, maximize their trading profits, difference systematically widens at certain times of the year, or but they are often forced to trade with trade size. in response to certain situations. in smaller trade sizes due to Our results provide direct factors beyond their control. For evidence on how trader behavior varies within a set trading example, suppose a trader submits a 5,000 share limit order time horizon and when a deadline exists. We would expect at the underlying best bid price. If an incoming 1,000 share similar loss-averse behavioral patterns to occur over various order executes against the trader’s order, but then the market time horizon settings in which there is some type of deadline price moves sharply away from the trader’s bid price, the being imposed or self-imposed on a decision-maker (e.g., a trader is left with 80% of the original order unfilled and will fund manager, trader, retail investor, etc.), such as with a be forced to reassess execution strategy. performance evaluation period, compliance or audit period, While trade sizes vary with underlying market conditions, a government tax period, a maintenance margin level, etc. traders (firms) do not usually reset price control mechanisms Economics literature has found evidence of this “wait until in accordance with trade size (i.e. on a percentage basis). In the very last moment” approach in other settings (e.g., with institutional trading settings such as ours, the emphasis is on bargaining negotiations), and our results provide some disciplined trading and adhering to a specific and well defined evidence on how psychological trading biases and deadline trading strategy. Our traders are trained to enter and exit effects interact in a financial market setting. There has been their positions within a very tight price range and they very little direct research on this front because researchers typically offset their positions within one or two cents. If studying trader behavior in financial market settings often these traders were to constantly reset their exit prices on a lack data on traders’ time horizons or the time horizon set by trade size percentage basis, this would create a much less the employee’s institution.16 Thus, it is difficult to measure disciplined approach to trading, and it would give them how trader behavior varies over a trading time horizon if the incentives to deviate from their normal trading practices. The traders are not trained to capture large price changes. Instead, trading time horizon is not truly known. they are trained to capture small price changes while trading D. How Does Price Control Influence Trader frequently on both sides of the market. Behavior? When price is heavily controlled and traders are given greater leeway with respect to trade size, this leaves firms While the end-of-day inventory liquidation requirement is vulnerable to heightened risk-taking with larger size trades. the most binding control mechanism the firm has in place, The existing stock price in relation to the opening stock price
16 For example, Benartzi and Thaler (1995) assume that the average investors holding time period is one year. This assumption has been applied in other research settings (e.g. Odean, 1998).

17

Recently adopted Regulation National Market System (Reg. NMS) eliminated sub-penny pricing for securities priced above $0.99.

32

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Table IV. Round-Trip Price Changes and Trade Size Distributions
This table reports the distribution for winning round-trip price changes, losing round-trip price changes, and executed trade sizes. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from June 2002 through May 2003. The traders executed 1.3 million trades which resulted in 730,417 round-trip trades over the 251 day sample period.

Price Change on Gains Mean Percentiles
10th 20th 30th 40th 50th 60th 70th 80th 90th $0.0020 $0.0060 $0.0080 $0.0100 $0.0100 $0.0100 $0.0100 $0.0130 $0.0200 290,248 $0.0126

Price Change on Losses
$0.0173

Trade Size
1,925

$0.0010 $0.0020 $0.0084 $0.0100 $0.0100 $0.0100 $0.0200 $0.0200 $0.0400 209,271

100 300 500 900 1,000 1,600 2,000 3,000 5,000 1,316,334

Obs.

determines whether a trade is for a capital gain or loss, but trade size, along with trade price, determines the magnitude of a trading gain or loss. When traders enter into larger trades and the price moves against them, the magnitude of their trading losses will increase and according to decision-making theory, they will have an increasing desire to get even. We segregate round-trip trade sizes into five trade size categories: 1) trade sizes less than 250 shares, 2) trade sizes greater than 249 shares and less than 1,000 shares, 3) trade sizes greater than 999 shares and less than 2,000 shares, 4) trade sizes greater than 1,999 shares and less than 3,000 shares, and 5) trade sizes greater than 2,999 shares. The overall holding time results for each trade size category, and for gains and losses, are reported in Figure 4A. While losing round-trips are held considerably longer than winning roundtrips for each trade size category, the difference systematically widens with trade size. We suspect this pattern is the result of traders moving deeper into the red with their larger trade sizes. As losses surmount and traders move further away from the break point, they will have an increasing desire to gamble (hold trades longer) in order to get back to the break even point. Yet, control mechanisms are not in place or are much weaker to stop this undesirable behavior because the

emphasis is on disciplined trading with respect to uniform price control. In Panel A of Table V, we examine overall performance and trade size. The absolute difference between the average trading gain and loss correspondingly widens with trade size. In general, we expect holding times to rise with larger trade sizes because it is more challenging to execute larger trades than smaller trades. However, this does not explain the widening gap between losing and winning round-trips with respect to trade size. Most of the trading losses can be attributed to trading in larger trade sizes (3,000 or more shares), where the loss-gain holding time difference is most pronounced. This suggests that traders’ decision to ride their losses longer with larger trade sizes is costly. We check the robustness of our trade size results by controlling differences in liquidity across the stocks traded. How liquid a stock is can affect both holding times and trading profits. We expect, on average, holding times to be lower on more liquid stocks. And, on average, we expect the price impact (if any) incurred executing a trade to be smaller on more liquid stocks. Variations in price impacts will be reflected in trading profits. In order to control liquidity differences across stocks, we sort the stocks traded into two

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

33

Figure 4. Winning vs. Losing Round-trip Holding Times by Trade Size
This figure displays the average holding time for winning and losing round-trips based on trade size (shares). Winning round-trips have a gross trading profit above zero while losing round-trips have a gross trading profit below zero. Holding times are calculated in seconds. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades on 693 securities which resulted in 730,417 roundtrip trades over the 251 day sample period. In Fig. 4B and 4C, holding times results are reported separately for trades occurring on liquid stocks and illiquid stocks. Liquid (illiquid) stocks are stocks with a turnover ratio in the top (bottom) 50% percentile of our sample. Stock turnover ratios are calculated using CRSP by averaging daily stock turnover (volume/shares outstanding) over the sample period.
Fig. 4A

Fig. 4B

Fig. 4C

34
Table V. Performance and Trade Size

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

This table reports round-trip performance results segregated by five trade size categories. The results are based on the trading records of 150 proprietary stock traders who traded the capital of a National Securities Dealer from May 2002 through June 2003. The traders executed 1.3 million trades on 693 securities which resulted in 730,417 round-trip trades over the 251 day sample period. In Panel B & C, performance results are reported separately for trades occurring on liquid stocks and illiquid stocks. Liquid (illiquid) stocks are stocks with a turnover ratio in the top (bottom) 50% percentile of our sample. Stock turnover ratios are calculated using CRSP by averaging daily stock turnover (volume/shares outstanding) over the sample period.
Panel A. All Stocks Trade Size (Shares) Number of RoundTrips Avg. RoundTrip Gain Avg. RoundTrip Loss Absolute RoundTrip Profit Difference Total Trading Profits

<250 250,<1000 1000,<2000 2000,<3000 3000
Trade Size (Shares)

154,469 182,982 154,496 89,044 149,426
Number of RoundTrips

$1.66 $7.62 $14.47 $23.05 $38.78
Avg. RoundTrip Gain

-$2.69 -$11.01 -$19.46 -$28.40 -$44.51
Avg. RoundTrip Loss

$1.03*** $3.39*** $4.99*** $5.35*** $5.73***
Absolute RoundTrip Profit Difference

-$18,652.07 $102,517.69 $18,803.97 $41,808.65 -$189,269.11
Total Trading Profits

Panel B. Liquid Stocks

<250 250,<1000 1000,<2000 2000,<3000 3000
Trade Size (Shares)

139,271 163,969 138,732 79,803 130,802
Number of RoundTrips

$1.54 $7.00 $14.22 $22.85 $39.05
Avg. RoundTrip Gain

-$2.50 -$9.99 -$18.93 -$28.09 -$45.04
Avg. RoundTrip Loss

$0.96*** $3.00*** $4.71*** $5.25*** $5.99***
Absolute RoundTrip Profit Difference

-$14,747.42 $92,034.00 $27,821.93 $45,967.47 -$129,172.52
Total Trading Profits

Panel C. Illiquid Stocks

<250 250,<1000 1000,<2000 2000,<3000 3000

12,221 14,684 12,018 6,613 12,104

$2.22 $9.66 $16.65 $26.00 $36.26

-$3.69 -$15.12 -$24.46 -$33.24 -$39.13

$1.47*** $5.46*** $7.82*** $7.25*** $2.88**

-$3,771.24 $3,057.33 $8,629.62 $2,453.53 -$29,340.84

***Significant at the 0.01 level. **Significant at the 0.05 level.

groups (liquid vs. illiquid stocks) based on their average daily turnover ratios (volume / shares outstanding) over our one year sample period. Volume and share outstanding data is obtained from the CRSP database. We are able to use matching trade data from CRSP for more than 97% of the trading activity in our data. We compute holding time differences and performance differences for both liquid stocks and illiquid stocks according to our trade size classifications. The holding time results are reported in Figures 4B and 4C and the performance results are reported in Panels B and C of Table V. As expected, most trading activity occurs on liquid stocks and holding times

are much lower on liquid stocks. For both liquid stocks and illiquid stocks, losing round-trips are held considerably longer than winning round-trips for each trade size category and the difference systematically widens with trade size. In general, the absolute difference between the average trading gain and loss correspondingly widens with trade size too. Although there is a sharp drop in the performance difference for illiquid stocks under the largest trade size category, there are very few large trade size observations on illiquid stocks. The trade size results highlight the need to assess institutional market participants’ resistance to loss realization (and its associate costs) at the individual trade level. Our

GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR

35

results also highlight the need to devise control mechanisms, among other things, on a situational trade basis. For example, a firm may analyze the trading decisions of its fund managers and find that overall, their fund managers do not exhibit a tendency to avoid realizing their losses. Consequently, the firm might feel less of a need to implement control mechanisms to prevent traders from taking excessive risks when they are confronted with the prospect of a loss. While the fund managers may not exhibit a tendency to avoid realizing their losses on an overall basis, they may exhibit a tendency to do so with their larger holdings, which would pose a significant (preventable) risk that is not easily detectable though casual analyses of the overall trading data.

IV. Conclusion
One of the more well known psychological tendencies that permeates Wall Street trading desks is the traders’ aversion to realizing losses. Traders have a tendency to hold their losing trades too long because they are predisposed to get even with their losses. By all accounts, this behavior is undesirable and can be quite costly. Securities firms are well aware of the costs that arise with this behavior and they often implement risk control mechanisms to prevent (limit) it from occurring. In our paper, we examine whether such measures actually work and how they influence proprietary stock trader behavior. Despite our sample firm’s efforts to get traders more comfortable with realizing their trading losses through training, managerial oversight, trader access to a licensed psychologist, discipline price control, and inventory liquidation, we find that professional traders still have difficulties accepting their losses. For example, traders hold

losing trades more than twice as long as winning trades and these longer holding times coincide with lower trading profits. While our results highlight how difficult it is for institutions to rid psychological biases from the traders’ decisions, our results also highlight the complexities involved with implementing efficient control mechanisms to get traders to realize their losses. When firms force traders to liquidate their inventory and realize their losses, professional traders respond by holding their losses up until the very last moment. When firms heavily focus on disciplined trading and uniform price control to ensure loss realization, professional traders respond by holding their losses longer on larger size trades. Clearly, these are not desirable behavioral responses to the firms underlying objective. However, if financial institutions impose stricter control mechanisms to get their traders to realize their losses sooner, the additional trading constraints will likely begin to start conflicting with the traders’ overall strategies and trading practices. On the other hand, if control mechanisms put in place are too lax, losing trades will be held for longer periods of time and losses will surmount. Securities firms implement control mechanisms to improve performance and reduce risk, but their efforts to get employees to accept their losses has much broader implications. Our results show that institutional risk control can have a strong influence on trader behavior. Future studies, which provide institution detail on the design of control mechanisms being used at other financial institutions and how employees respond to them, would be insightful for both creating optimal risk control mechanisms and also for determining their overall effects in the marketplace.

36
References
Admati, A.R. and P. Pfleiderer, 1988, “A Theory of Intraday Patterns: Volume and Price Variability,” Review of Financial Studies 1 (No. 1), 3-40. Barber, B.M. and T. Odean, 1999, “The Courage of Misguided Convictions: The Trading Behavior of Individual Investors,” Financial Analysts Journal 56 (No. 6), 41-55. Barber, B.M. and T. Odean, 2000, “Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,” Journal of Finance 55 (No. 2), 773-806. Barber, B., Y. Lee, Y. Lu, and T. Odean, 2007, “Is the Aggregate Investor Reluctant to Realize Losses: Evidence from Taiwan,” European Financial Management, 13 (No. 3), 371-388. Benartzi, S. and R. Thaler, 1995, “Myopic Loss Aversion and the Equity Premium Puzzle,” Quarterly Journal of Economics 110 (No. 1), 73-92. Conrad, J., K.M. Johnson, and S. Wahal, 2002, “The Trading of Institutional Investors: Theory and Evidence,” Journal of Applied Finance 12 (No. 1), 7-14. Coval, J. and T. Shumway, 2005, “Do Behavioral Biases Affect Prices?” Journal of Finance 60 (No. 1), 1-34. Dhar, R. and N. Zhu, 2007, “Up Close and Personal: An Individual Level Analysis of the Disposition Effect,” Management Science, forthcoming. Feng, L. and M. Seasholes, 2005, Do Investor Sophistication and Trading Experience Eliminate Behavioral Biases in Financial Markets?” Review of Finance 9 (No. 3), 305-351. Frazzini, A., 2006, “The Disposition Effect and Underreaction to News,” Journal of Finance 61 (No. 4), 2017-2046. Garvey, R. and A. Murphy, 2004, “Are Professional Traders Too Slow to Realize Their Losses?” Financial Analysts Journal 60 (No. 4), 35-43. Garvey, R. and A. Murphy, 2005, “The Profitability of Active Stock Traders,” Journal of Applied Finance 15 (No. 2), 93-100. Garvey, R., A. Murphy, and F. Wu, 2007, “Do Losses Linger? Evidence from Proprietary Stock Traders,” Journal of Portfolio Management 33 (No. 4), 75-83. Goldberg, D.C. and A. Lupercio, 2004, “Cruising at 30,000: SemiPro Numbers Level Off, but Trading Volumes Rise,” Bear Stearns Company Report, August.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Government Accountability Office (GAO), 2005, “Decimal Pricing has Contributed to Lower Trading Costs and a More Challenging Trading Environment,” GAO Report, May. Full report available online at: http://www.gao.gov/new.items/d05535.pdf. Grinblatt, M. and M. Keloharju, 2001, “What Makes Investors Trade?” Journal of Finance 2 (No. 2), 589-616. Heisler, J., 1994, “Loss Aversion in a Futures Market: An Empirical Test,” Review of Futures Markets 13 (No. 3), 793-822. Locke, P.R. and S.C. Mann, 2005, “Professional Trader Discipline and Trade Disposition,” Journal of Financial Economics, 76 (No. 2), 401-44. Nguyen, V.T., B.F. Van Ness, and R.A. Van Ness, 2004, “The Reporting of Island Trades on the Cincinnati Stock Exchange,” Journal of Applied Finance, 14 (No. 2), 30-39. Odean, T., 1998, “Are Investors Reluctant to Realize Their Losses?” Journal of Finance 53 (No. 5), 1775-1798. Scherbina, A. and L. Jin, 2005, “Change is Good or The Disposition Effect among Mutual Fund Managers,” Working Paper, Harvard Business School. Shapira, Z. and I. Venezia, 2001, “Patterns of Behavior of Professionally Managed and Independent Investors,” Journal of Banking and Finance 25 (No. 8), 1573-1587. Shefrin, H. and M. Statman, 1985, “The Disposition to Sell Winners Too Early and Ride Losers Too Long: Theory and Evidence,” Journal of Finance 40 (No. 3), 777-790. Statman, M. and D. Caldwell, 1987, “Applying Behavioral Finance to Capital Budgeting: Project Terminations,” Financial Management 16 (No. 4), 7-15. Thaler, R., 1985, “Mental Accounting and Consumer Choice,” Marketing Science 4 (No. 3), 199-214.

Why Do People Trade?1
Anne Dorn, Daniel Dorn, and Paul Sengmueller

Besides trading to save, manage risk, and speculate, people trade simply because they find it entertaining. In a survey of 1,300 German discount brokerage clients, respondents who indicate that they “enjoy investing” and “enjoy risky propositions” trade twice as much as their peers. In contrast, standard motives for trading such as saving and rebalancing explain little of the variation in trading activity across investors. Entertainment appears to be a straightforward explanation for why some people trade much more than others and why active traders underperform their peers after transaction costs.

important social function of incorporating information into asset prices. It also (viewed from certain quarters more importantly) provides a major source of revenue for securities firms. In this paper, we explore reasons investors trade that go beyond assembling a portfolio with the best return-risk profile. We argue that in exchange for this major source of revenue, securities firms are also allowing traders to enjoy themselves in the act of trading. Some investors view trading as a hobby, and hand over their large trading fees as happily as an audiophile pays top dollar for the latest in speaker technology. Other investors, we find, are playing the market in a very literal sense, racking up trading costs like a casino patron sliding his chips across the table. We make these characterizations in an attempt to explain three stylized facts about trading that have caught the attention of researchers and practitioners alike: 1. Trading volume in financial markets is high. For example, stocks in the US change hands roughly once per year. 2. Trading volume is concentrated among a small number of market participants. In a well-cited study of US discount brokerage clients, Barber and Odean (2000) report that the most active investors turn over their portfolio several times per year. In contrast, a substantial fraction of US individual investors with a brokerage account do not trade at all in a given year, according to recent waves of the US Survey of Consumer Finances. 3. Traders underperform buy-and-hold investors. Barber and Odean (2000) report that the most active investors underperform the least active investors by several percentage points per year and that the performance differential is essentially due to trading costs. Standard economic theory appears to be at odds with these facts. The rational investor assumed by standard theory is only interested in the return and risk attributes of his portfolio,

Trading in financial markets is an important economic activity. Trades are necessary to get into and out of the market, to put unneeded cash into the market, and to convert back into cash when the money is wanted. They are also needed to move money around within the market, to exchange one asset for another, to manage risk, and to exploit information about future price movements. All this trading performs the

Anne Dorn is currently unaffiliated. Daniel Dorn is an Assistant Professor of Finance at Drexel University in Philadelphia, PA. Paul Sengmueller is an Assistant Professor of Finance at CentER-Tilburg University in Tilburg, the Netherlands. This contribution is based on the article “Trading as Entertainment?” by Daniel Dorn and Paul Sengmueller which is forthcoming in Management Science. The copyright on “Trading as Entertainment?” is held by INFORMS.
1

37

38
and only trades if the benefits from trading justify the costs. In general, return maximizing rational investors should be very reluctant to trade with one other. The intuition is simple. Trading is a zero-sum game. If rational investor A offers to trade with rational investor B, B should be suspicious that A knows something about the future price that B does not. Milgrom and Stokey (1982) and Tirole (1982) formalize this intuition and show that indeed, rational investors should refuse to speculatively trade with each other. The inadequacy of standard theory opens the door for the behavioral approach. De Bondt and Thaler (1995) call the observed trading volume in financial markets “perhaps the single most embarrassing fact to the standard finance paradigm.” The leading answer of the behavioral camp to the question “Why do people trade?” is overconfidence, prominently advocated by Odean (1998). Essentially, overconfidence allows both parties to a trade to believe that they will win the zero-sum game of trading. For all its intuitive appeal, the overconfidence hypothesis has received only mixed empirical support. Barber and Odean (2001) report that male US discount brokerage clients trade more than their female counterparts and interpret this finding as consistent with the overconfidence hypothesis. Grinblatt and Keloharju (2008) use self-confidence assessments from a psychological test administered by the Finnish military to infer overconfidence of male Finnish investors. They report that the univariate correlation between the self-confidence score and trading activity is close to zero. Other things equal, however, the proxy for overconfidence derived from the selfconfidence assessments is significantly positively related to the number of trades, though not to portfolio turnover. Glaser and Weber (2003) use a questionnaire to elicit nine proxies for overconfidence in a sample of German discount brokerage customers and relate the proxies to actual portfolio turnover. None of the proxies help explain cross-sectional variation in portfolio turnover. In trading experiments with students from different countries, Deaves, Lüders, and Luo (2004), and Biais, Hilton, Mazurier, and Pouget, (2005) report little or no relation between proxies for overconfidence and observed trading activity. This paper explores a different explanation of why people trade, anticipated by Black (1986) who notes that “[w]e may need to introduce direct utility of trading to explain the existence of speculative markets.” For people who trade because they like to do so, the monetary cost of trading is offset by non-pecuniary benefits from researching, executing, talking about, anticipating the outcome of, or experiencing the outcome of a trade. Motives for entertainment trading can be classified in three distinct groups: recreation, sensation seeking, and an aspiration for riches. Recreational trading can be motivated by a feeling of

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

accomplishment (similar to a homeowner who decides to do it himself rather than hiring a contractor), camaraderie (among members of an investment club, for example), or it can emerge as a by-product of following the financial markets as a hobby (like a technophile who likes to read reviews of the latest gadgets, and is then tempted to go out and buy them). Perceiving investing as a diversion rather than a chore, hobby investors have less of a psychological hurdle to overcome when executing changes to their portfolio, directly lowering their marginal cost of trading. By actively following the financial markets, they also expose themselves to more trading signals and should hence be expected to trade more than their peers.2 Entertainment trading can also be motivated by sensation seeking in the financial domain. According to Zuckerman (1994), “Sensation seeking is a trait defined by the seeking of varied, novel, complex, and intense sensations and experiences, and the willingness to take [...] financial risks for the sake of such experience.” In a (perhaps subconscious) quest for arousal, sensation seekers look for both intensity and novelty in experience. An undiversified portfolio of volatile stocks exposes its holder to intense stimuli in the form of extreme returns. Exposure to such stimuli by itself may trigger trading as argued by Dorn and Huberman (2007). In addition, as pointed out by Grinblatt and Keloharju (2008), sensation seekers in the financial domain may value the act of trading in and of itself because a trade — a new bet — affords the desired novelty of experience. Grinblatt and Keloharju (2008) use traffic violations to proxy for thrill seeking behavior; they report that variation in the number of speeding tickets explains variation in trading activity in a large sample of Finnish investors. Alternatively, trading can be motivated by an aspiration for riches as suggested by Statman (2002). A trade can be seen as a bet that carries a “dream value,” that is, the joy of imagining what a handsome payoff will buy. Such aspirations have been used to explain lottery participation (Conlisk, 1993) and exploited in advertising by retail brokers (Barber and Odean, 2002). Aspiration-driven investors should hold portfolios with volatile and positively skewed returns to increase the chance of reaching an aspiration level far above their current wealth (see Kumar, 2008). The exposure to trading stimuli in the form of extreme returns, coupled with an inherent impatience to reach their desired wealth level, may lead aspiration-driven investors to pick up and abandon trading ideas more quickly than their peers.

This argument is reminiscent of Merton (1987) who motivates his 1986 Presidential Address to the American Finance Association by the simple observation that an investor needs to know about a stock before he can trade it.
2

DORN, DORN, & SENGMUELLER — WHY DO PEOPLE TRADE?

39

The paper examines the hypothesis that entertainment In addition to the standard trade attributes, the records include motives drive trading by combining survey responses and a channel variable that indicates whether the order was placed transaction records for a sample of more than 1,000 clients over the phone, over the internet, or within an automatic at one of the top three discount brokers in Germany. The investment or withdrawal plan that exist for dozens of survey offers responses to statements that elicit whether a individual stocks and mutual funds. Such plans allow respondent enjoys investing and statements that have been investors to gradually build or reduce positions at four dates used to identify compulsive gamblers. The responses to these each month (similar to ShareBuilder in the US). statements serve as proxies In July and August 2000, for the entertainment benefits after the sample period, each Standard motives for trading such derived from trading. investor participated in a The main findings are as survey that elicited a wide as saving and rebalancing fail to follows: range of objective and explain both the level of observed 1. Standard motives for subjective investor attributes trading fail to explain much of detailed below. Table I trading activity and the variation the observed trading activity. summarizes the client Turnover due to savings, portfolios and trading activity of such activity across individuals. dissavings, liquidity, and during the sample period rebalancing considerations January 1995 to May 2000. accounts only for about one third of total turnover. Average monthly turnover, defined as one half the sum of 2. Standard motives for trading fail to explain why some the absolute values of purchases and sales during a given people trade much more than others. Turnover due to savings, month divided by the average portfolio value during that dissavings, liquidity, and rebalancing considerations varies month averaged first across time for each investor and then much less across investors than turnover that cannot be across investors, is 15%. In our turnover calculation, we justified by standard trading motives. consider purchases and sales of individual stocks, individual 3. Entertainment appears to be a major driver of portfolio bonds, mutual funds, options, and term deposits. Individual holdings. Entertainment-driven investors hold more stock trades account for 62%, fund trades account for 18%, concentrated portfolios, riskier portfolio components, and and option trades account for 15% of the total trading volume portfolios with more positively skewed returns. during the sample period. The average portfolio size over 4. Entertainment appears to be a major driver of portfolio the entire account life is roughly 90,000 Deutsche Mark turnover, especially turnover that cannot be justified by [DEM] or 50,000 US dollars [USD] at the average USD/ standard trading motives. Entertainment-driven investors turn DEM exchange rate of 1.7 during the sample period. over their portfolio of stocks, bonds, funds, and options at We analyze the portfolios using a measure of concentration roughly twice the rate of their peers. known as the Herfindahl-Hirschmann Index (HHI).4 The 5. Proxies for overconfidence are at best weakly correlated median HHI of the stock and fund portfolios during the sample period is 31%; that is, the typical client holds the equivalent with trading activity. The remainder of the paper proceeds as follows: Section I of an equally-weighted portfolio of three individual positions. From the information provided by the client to the broker describes the data and the construction of the variables. Section II discusses the main findings in more detail. Section at account opening, we can infer the gender of all main account holders and the age of those who choose to report III concludes. their birth date. The typical respondent is male, young, and has held the account for three years. Judging from a survey I. The Data of Germans who hold stocks, either directly or through mutual funds (see Deutsches Aktieninstitut, 2000), our sample A. Brokerage Records investors are more predominantly male and younger than the The analysis is based on a complete history of daily transaction records in individual stocks, term deposits, bonds, mutual funds, and options obtained for a sample of 1,345 current and former clients at one of Germany’s three largest discount brokers between January 1, 1995 and May 31, 2000.3
The broker is labeled as a discount broker because no investment advice is given.
3

The HHI is defined as the sum of squared portfolio weights. A portfolio consisting of n equally-weighted stocks would have an HHI of 1/n. Stock mutual funds are assumed to consist of one hundred equally-weighted positions that do not overlap with other holdings of the investor. That is, the HHI of portfolio of an investor holding one stock mutual fund is 1% and that of an investor splitting his money equally between two stock mutual funds is 0.5%.
4

40
Table I: Summary Statistics

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Portfolio characteristics are calculated from the complete daily transaction history available for each of the 1,345 sample investors from the day when the account was opened until May 31, 2000 or the day when the account was closed, whichever comes first. Turnover in a given month is the sum of the absolute value of purchases and sales of stocks, bonds, mutual funds, and options divided by twice the higher of the portfolio value at the beginning or at the end of the month (to avoid extreme values). Average portfolio value is calculated at the end of every month across all individual stocks, funds, options, bonds, and term deposits in the client’s portfolio. During the sample period, one US Dollar [USD] corresponds to roughly Deutsche Mark [DEM] 1.7. The Herfindahl-Hirschmann Index (HHI) is calculated using only stocks and stock mutual funds for which Datastream offers a complete history of non-stale prices and returns. Higher values of the HHI indicate less diversification. “Gender” is a dummy variable that is one if the respondent reports to be male and zero otherwise (if missing, we replace the missing value with the gender recorded for the main account holder in the brokerage database). “Age” is the age of the respondent (if missing, we replace the missing value with the age recorded for the main account holder in the brokerage database). “College” is a dummy that is one if a respondent has a college degree and zero otherwise. “Self-employed” is a dummy that is one if the respondent reports to be self-employed and zero otherwise. “Income” is the self-reported gross annual income. “Wealth” is the self-reported total net worth (including all financial assets and real estate).

Portfolio Characteristics
Average monthly portfolio turnover Average portfolio value [DEM] Average Herfindahl-Hirschmann Index

Mean
15% 86,000 31%

Median
7% 38,000 25%

Investor Characteristics
Gender [fraction male] Age Education [fraction with college education] Self-employed Gross annual income [DEM} Net worth [DEM] 88% 39 70% 17% 93,000 373,000 36

88,000 325,000

typical German stock market participant. Relative to the population of German stock market participants, the sample investors also appear to be more highly educated and earn higher incomes (see Dorn and Huberman, 2005).

on a five-point scale ranging between (1) strongly disagree, (2) tend to disagree, (3) tend to agree, (4) strongly agree, and (5) don’t know: 1. I enjoy investing. 2. I enjoy risky propositions. 3. Games are only fun when money is involved. 4. In gambling, the fascination increases with the size of the bet. Agreement with statement one defines a hobby investor. Agreement with statements two to four identifies respondents who enjoy risky propositions, in general, and gambling, in particular; in fact statements three and four are taken from a study on identifying compulsive gamblers (Nadler, 1985). Hobbyists and gamblers appear to form distinct groups; the response to the first statement is only weakly correlated with the responses to the other statements. Statements two through four flag investors as gamblers more or less consistently; the pairwise correlation between the responses to statements two to four is quite high and reaches 0.46 between statements three and four. Table II summarizes objective demographic and socioeconomic attributes of investors grouped by their responses

B. Survey Variables
To gauge which investors likely derive non-pecuniary benefits from their trading activities, we use their self-reported attitudes towards investing and gambling gleaned from a survey administered in July and August 2000. The survey elicited information on the investors’ investment objectives, risk attitudes and perceptions, investment experience and knowledge, portfolio structure, and demographic and socioeconomic status. Dorn and Huberman (2005) describe the survey in detail. To pin down the importance of entertainment motives for different investors, we focus on the survey items that make an explicit reference to whether or not respondents enjoy dealing with their investments or enjoy gambling. This focus yields responses to a total of four statements (reproduced below in translation from the original German). The investors are asked to indicate their agreement with the four statements

DORN, DORN, & SENGMUELLER — WHY DO PEOPLE TRADE?

41

Table II: Characteristics of Entertainment-Driven Investors
Panels A through D characterize investors grouped by their responses to four survey statements designed to elicit whether the respondents enjoy investing or gambling with money. The investors are asked to indicate their agreement with the four statements on an ordinal scale of (1) strongly disagree, (2) tend to disagree, (3) tend to agree, (4) strongly agree. In Panel A, we have combined the categories (1) and (2) to “disagree” since only four respondents choose to “strongly disagree.” In Panel D, we have combined the categories (3) and (4) to “agree” since only thirty-eight respondents choose to “strongly agree.” “Nobs” is the number of respondents in each category. The demographic and socio-economic variables are defined as in Table I.

Nobs

Gender

Age

College

Selfemployed

Income

Wealth

Panel A - Statement 1: “I enjoy investing.”
Disagree Tend to agree Strongly agree 84 403 822 76% 87% 91%*** 40 41 41 73% 72% 69% 15% 17% 16% 90 94 94 262 358 396***

Panel B - Statement 2: “I enjoy risky propositions.”
Strongly disagree Tend to disagree Tend to agree Strongly agree 148 571 492 87 82% 88% 90% 95%*** 48 41 39 37*** 69% 70% 70% 75% 13% 16% 17% 21% 85 93 97 102** 421 385 357 330*

Panel C - Statement 3: “Games are only fun when money is involved.”
Strongly disagree Tend to disagree Tend to agree Strongly agree 470 487 277 71 87% 89% 92% 87% 41 41 40 38* 76% 66% 67% 62%** 14% 17% 19% 22% 91 94 97 84 387 381 351 282**

Panel D - Statement 4: ``In gambling, the fascination increases with the size of the bet."
Strongly disagree Tend to disagree Agree 674 396 199 89% 88% 90% 41 41 39** 72% 68% 65%* 16% 18% 18% 95 90 94 392 364 329**

***Significant at the 0.01 level **Significant at the 0.05 level. *Significant at the 0.10 level.

to the above statements. We exclude the few investors with missing responses and investors who respond with “don’t know” — out of a total of 1,345 respondents, the number of missing responses ranges from 10 (for statement three) to 15 (for statement one); the number of respondents who respond with “don’t know” ranges from 11 (for statement one) to 56 (for statement four). To be able to make meaningful statistical comparisons across groups, we group investors who “strongly disagree” with statement one together with those who “tend to disagree” as there are only four investors who “strongly disagree.” For the same reason, we combine the “strongly

agree” and “tend to agree” categories for statement four as only 38 investors “strongly agree.” Male investors and wealthier investors appear to enjoy dealing with investments more than their female and less wealthy counterparts. Those who enjoy games only when money is involved, in particular, tend to be younger, less well educated, and less wealthy. Although we have no direct information about whether our sample investors engage in gambling outside the stock market, it is interesting to note that younger age, a lower level of education, and less wealth have been linked to a higher propensity to participate in legal

42
forms of gambling in Germany (see Albers and Hübl, 1997), the UK (see Farrell and Walker, 1999), and the US (see Clotfelter and Cook, 1989).

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

normal as they are likely motivated by liquidity and savings considerations.6 Our trade classification likely overstates normal turnover, in part because we only observe part of the portfolio for some investors. For example, an investor might sell off a complete II. Main Results position of stock A in an unobserved account and invest the proceeds in stock B in the observed account because he It is our task in this section to show that the high trading expects stock B to outperform stock A; such a purchase would levels laid out in the stylized facts with which we began the be classified as a normal trade even though it is not driven by paper are due to investors savings, liquidity, or trading for entertainment rebalancing motives. motives. We will demonstrate Entertainment-driven investors Table III reports summary that that small group of statistics for normal and turn over their portfolio of stocks, investors who exhibit high excess turnover. Across the turnover and lagging returns bonds, funds, and options at sample respondents, the is composed of those average monthly total turnover roughly twice the rate of their individuals who find trading of 15% consists of 5% normal entertaining, those individuals turnover and 10% excess peers. whose survey responses label turnover — in other words, them as hobby or gambling only one third of the observed trading volume can be investors. Dorn and Sengmueller (2008) go on to document explained by savings, liquidity, and rebalancing motives. that these results are not driven by survey response bias, past The standard deviation of normal turnover across the returns, or small accounts. sample respondents is 4% as opposed to 29% for excess turnover. Therefore, the challenge in explaining the A. Standard Motives Inadequately Explain the heterogeneity in trading activity across investors appears to Observed Trading Activity lie in understanding excess turnover; investors appear to be fairly homogenous in their desire to trade due to savings, We divide the turnover we observe in our sample into two liquidity, or rebalancing motives. finer measures: normal turnover and excess turnover. Normal turnover consists of trading that can be explained by standard B. Cross-sectional Differences in Portfolio motives for trading such as savings, dissavings, liquidity, or Characteristics Are Consistent With the rebalancing considerations; excess turnover is the portion of Entertainment Hypothesis total turnover that cannot be explained by these motives. Similar to Barber and Odean (2002), we define an excess Table IV sets out the portfolio characteristics of investors, sale as a sale of a complete position of an individual stock, separated by their responses to our four statements. We find mutual fund, or option that is followed by one or more stock, fund, or option purchases within three weeks of the sale. We that those investors who are most excited by risk indeed hold define excess purchases as all stock, fund, and option more risky portfolios. Self-professed gamblers in our sample hold more purchases made within three weeks of an excess sale. All concentrated equity portfolios. For example, those who 5 other trades are classified as normal trades. In particular, all trades in term deposits and automatic investment and strongly agree with the statement “I enjoy risky propositions” withdrawal plans — plans that allow investors to gradually hold equity portfolios with an average HHI of 0.39 which build or reduce positions in dozens of stocks and funds at corresponds to an equally weighted position in two to three four predetermined dates per month — are classified as individual stocks; by contrast, their peers who strongly disagree with this statement hold the equivalent of an equally

We have explored variations of this definition in unreported robustness checks. For example, one could argue that the complete sale of an individual stock position followed by the purchase of a stock mutual fund constitutes a diversifying and hence normal trade. One could also argue that put purchases are used for portfolio insurance purposes. Since most trading occurs in individual stocks and call options, these variations in the definition of excess turnover have little effect on our results.
5

6

Barber and Odean (2002) use the terms “non-speculative” and “speculative” trades instead of normal and excess trades. Substantively, our classification differs from theirs in three ways. First, they restrict their analysis to trades in common stocks. Second, they require that sales be for a profit to rule out tax-loss motivated trading (capital gains from sales of financial securities are essentially not taxed in Germany). Third, they do not distinguish between savings plan and non-plan trades.

DORN, DORN, & SENGMUELLER — WHY DO PEOPLE TRADE?

43

Table III: Normal Turnover Versus Excess Turnover
Normal turnover in a given month is defined as one half the sum of the absolute values of normal purchases and normal sales during a given month divided by the average portfolio value during that month. Excess turnover is defined similarly, but using excess purchases and excess sales. An excess sale is defined as a sale of a complete position of an individual stock, mutual fund, or option that is followed by one or more stock, fund, or option purchases within three weeks of the sale. An excess purchase is defined as a stock, fund, or option purchase made within three weeks of an excess sale. All other trades are classified as normal trades. In particular, all trades in term deposits and automatic investment and withdrawal plans are classified as normal.

Mean
Average monthly portfolio turnover thereof: normal turnover excess turnover 15% 5% 10%

Std
32% 4% 29%

Median
7.4% 3.9% 2.9%

Table IV: Portfolio Choices of Entertainment-Driven Investors
HHI is the average Herfindahl-Hirschmann Index across the portfolios in the group, higher values indicate less diversification. Average component volatility (ACV) is the value-weighted average volatility of the portfolio components in an investor’s portfolio. Realized skewness is calculated from daily portfolio returns as in Chen et al. (2001). “Options” is the fraction of respondents in a group that have traded options at some point during the sample period. HHI, ACV, and skewness are calculated using only the individual stocks and stock mutual funds for which Datastream provides daily total return data.

Nobs

HHI

ACV

Realized Skewness 0.92 0.50 0.74

Options

Panel A - ``I enjoy investing."
Disagree Tend to agree Strongly agree 84 403 822 30% 30% 31% 42% 42% 45% 18% 29% 41%***

Panel B - ``I enjoy risky propositions."
Strongly disagree Tend to disagree Tend to agree Strongly agree 148 571 492 87 26% 28% 33% 39%*** 39% 41% 48% 52%*** 0.45 0.45 0.91 1.32*** 22% 29% 46% 51%***

Panel C - ``Games are only fun when money is involved."
Strongly disagree Tend to disagree Tend to agree Strongly agree 470 487 277 71 28% 30% 34% 38%*** 42% 43% 48% 53%*** 0.63 0.52 0.87 1.07 30% 34% 46% 52%***

Panel D - ``In gambling, the fascination increases with the size of the bet."
Strongly disagree Tend to disagree Agree 674 396 199 28% 32% 37%*** 42% 45% 50%*** 0.53 0.77 0.96** 31% 42% 47%***

***Significant at the 0.01 level **Significant at the 0.05 level.

44

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

involved” turn over their portfolios at an average monthly weighted portfolio of four stocks. Not only are the portfolios of gamblers more concentrated, rate of 24% — twice the rate of those who strongly disagree but they also consist of individually riskier securities. For with the statement (see Figure 3). example, the average component volatility — the valueIf differences in trading activity were indeed driven by weighted average of the annualized volatility of the stock entertainment, one would expect such differences to manifest portfolio components — of investors who strongly agree with themselves in terms of excess turnover; that is, turnover the statement “I enjoy risky propositions” averages 52% unlikely due to savings, liquidity, or rebalancing relative to 39% for investors considerations. Indeed, who strongly disagree with Figures 1-4 also show that Entertainment as a motive for trading this statement. virtually the entire difference Consistent with gamblers in total turnover between is distinct from overconfidence. preferring skewness (see those who enjoy investing or Proxies for overconfidence are at best Golec and Tamarkin, 1998), gambling and their peers is people classified as due to the higher excess weakly correlated with trading gamblers in our data set hold turnover of the activity. portfolios of stocks and entertainment-driven mutual funds that exhibit investors. For example, more positively skewed returns. We exclude holdings of investors who strongly agree with “Games are only fun when individual bonds and options when calculating portfolio money is involved” exhibit normal turnover rates averaging statistics, in part because of a lack of high-frequency price 5% — similar to the average normal turnover of 4.5% of data. However, options holdings and trades also point to their peers who strongly disagree with the statement. entertainment-motivated investors preferring securities with However, the average excess turnover rate of the selfpositively skewed payoffs. For example, half of the investors professed gamblers, 19%, is almost thrice the corresponding who strongly agree with the statement “I enjoy risky rate of their peers (see Figure 3). propositions” trade options during our sample period; in In Dorn and Sengmueller (2008) we investigate whether contrast, only one out of five investors who strongly disagree these correlations between turnover and entertainment with this statement also trade options. motives hold up under multivariate statistical analysis. And indeed we find that even after controlling for gender, age, education, income, employment status, and wealth, investors C. Cross-sectional Differences in Turnover who enjoy investing or gambling trade more than those who Are Consistent With the Entertainment enjoy neither, and investors who enjoy both trade the most of Hypothesis all. These results suggest that investors appear to derive Trading is costly. The typical respondent in the paper’s pleasure from trading both as a pastime and as a form of sample spends 0.5% of his self-reported gross annual income on trading commissions. The main hypothesis entertained gambling. Figure 5 illustrates that respondents who enjoy here is that some investors derive non-pecuniary benefits from investing (that is, they strongly agree with the statement “I researching, executing, talking about, anticipating, or enjoy investing”) but not gambling (that is, they disagree or experiencing the outcome of a trade. These benefits help strongly disagree with the statement “Games are only fun when money is involved”), and those who enjoy gambling offset the cost of trading. We group the survey respondents by their responses to each but not investing trade more than their peers who enjoy neither of the four entertainment statements. Figures 1-4 illustrate investing nor gambling; those who enjoy both investing and the equally-weighted average monthly turnover rates for the gambling trade the most. members of each group. Investors who report enjoying investing also trade more aggressively than their peers. Figure 1 shows that investors who strongly agree with “I enjoy investing” exhibit an average monthly turnover of 17% — significantly higher than the average turnover rate of 10% for the investors who disagree with the statement. Similar turnover patterns are obtained for investors grouped by their responses to statements that elicit the investor’s affinity to gambling (see Figures 2-4). For example, investors who strongly agree with “Games are only fun when money is

D. Differences in Overconfidence Fail to Explain Turnover Differences
Overconfidence might explain the paper’s results if overconfident investors report enjoying trading because they enjoy doing what they wrongly perceive themselves to be good at. Alternatively, entertainment might amplify the effects of overconfidence or vice versa. The wealth of survey responses allows us to construct three

DORN, DORN, & SENGMUELLER — WHY DO PEOPLE TRADE?

45

Figure 1: Turnover as a Function of Enjoyment of Investing
18% 16% 14% 12% 10% 8% 6% 4% 2% 0% Tend to disagree or strongly disagree Tend to agree Strongly agree

Total turnover Normal turnover Excess turnover

Agreement with "I enjoy investing."

Figure 2: Turnover as a Function of Enjoyment of Risky Propositions
45% 40%

Total turnover
35% 30% 25% 20% 15% . 10% 5% 0% Strongly disagree Tend to disagree Tend to agree Strongly agree

Normal turnover Excess turnover

Agreement with "I enjoy risky propositions."

46

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Figure 3: Turnover as a Function of Affinity for Gambling (I)
30%

25%

Total turnover Normal turnover Excess turnover

20%

15%

10%

.

5%

0% Strongly disagree Tend to disagree Tend to agree Strongly agree

Agreement with "Games are only fun when money is involved."

Figure 4: Turnover as a Function of Affinity for Gambling (II)
30%

25%

Total turnover Normal turnover

20%

Excess turnover

15%

10%

5%

0% Strongly disagree Tend to disagree Tend to agree or strongly agree

Agreement with "In gambling, the fascination increases with the size of the bet."

DORN, DORN, & SENGMUELLER — WHY DO PEOPLE TRADE?

47

Figure 5: Excess Turnover of Investors Sorted by Enjoyment of Investing and Gambling
35%

30%

Do not enjoy gambling Enjoy gambling

25%

20%

15%

10%

5%

0% Do not enjoy investing Enjoy investing

proxies that capture different aspects of overconfidence: the tendency to overestimate one’s knowledge, the tendency to overly attribute successes to skill in conjunction with past returns (known as the self-enhancing attribution bias), and the erroneous expectation of being able to affect chance outcomes (known as the illusion of control; see also Barber and Odean, 2002; Daniel et al., 1998; and Gervais and Odean, 2001). We use the investor’s agreement with the statement “I’m much better informed than the average investor” as a proxy for the tendency to overestimate one’s knowledge, or relative knowledge. To estimate the self-enhancing attribution bias, we consider the extent to which survey participants agree with the statement “My past investment successes were, above all, due to my specific skills.” To construct a proxy for the illusion of control, we compute an aggregate score using the investors’ responses to four statements: 1. When I make plans, I am certain that they will work out. 2. I always know the status of my personal finances. 3. I am in control of my personal finances. 4. I control and am fully responsible for the results of my investment decisions.

In regressions reported in Dorn and Sengmueller (2008), we find that none of the overconfidence proxies is significantly related to excess turnover. Moreover, the significance of our hobby and gambler proxies is maintained even while controlling for overconfidence. We also investigated how our hobbyist and gambler designations interact with overconfidence. The results of interacting the investor’s agreement with “I enjoy investing” with the three overconfidence proxies are shown in Figures 6-8. To simplify the presentation, and to ensure that the resulting groups consist of enough members, we created binary entertainment and overconfidence proxies. In general, no additional insights were found in this interaction, overconfident hobby or gambler investors turned over their portfolios at similar rates as underconfident investors with the same hobby or gambling affinities.7

III. Conclusion
Some investors derive enjoyment from trading which offset
In Figure 6, it appears that the underconfident who do not enjoy investing trade less than their overconfident peers. However, the difference in trading activity is not statistically significant.
7

48

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Figure 6: Excess Turnover of Investors Sorted by Enjoyment of Investing and Relative Knowledge
35%

30%

Less knowledge than average investor More knowledge than average investor

25%

20%

15%

10%

5%

0% Do not enjoy investing Enjoy investing

Figure 7: Excess Turnover of Investors Sorted by Enjoyment of Investing and Self-Attribution of Success
30%

25%

Low self-attribution of success High self-attribution of success

20%

15%

10%

5%

0% Do not enjoy investing Enjoy investing

DORN, DORN, & SENGMUELLER — WHY DO PEOPLE TRADE?

49

Figure 8: Excess Turnover of Investors Sorted by Enjoyment of Investing and Self-Control
30%

25%

Low self-control High self-control

20%

15%

10%

5%

0% Do not enjoy investing Enjoy investing

the costs of churning. Like lottery players who buy tickets with negative expected values, entertainment-driven investors trade even though trading diminishes the expected monetary payoff of their portfolio. Consistent with this conjecture, variation in the self-reported enjoyment of investing and gambling explains variation in trading intensity even after controlling for competing explanations such as overconfidence. The most entertainment-driven investors trade about twice as much as those who fail to take pleasure in gambling or investing. Relying solely on transaction records (that is,

independently of the survey responses), we estimate that more than half of the observed portfolio turnover is excess turnover — turnover in excess of what can be justified by standard trading motives such as savings/dissavings, liquidity, and rebalancing. Most of the variation in trading activity across individuals is variation in excess turnover. Variation in excess turnover is highly correlated with our proxies for nonpecuniary benefits derived from trading. In sum, entertainment trading appears to be quantitatively important — at least for this sample of discount brokerage customers during the late 1990s.

50 References
Albers, N. and L. Hübl, 1997, “Gambling Market and Individual Patterns of Gambling in Germany,” Journal of Gambling Studies 13 (No.2), 125-144. Barber, B. and T. Odean, 2000, “Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,” Journal of Finance 55 (No.2), 773-806. Barber, B. and T. Odean, 2001, “Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment,” Quarterly Journal of Economics 116 (No. 1), 261-292. Barber, B. and T. Odean, 2002, “Online Investors: Do the Slow Die First?” Review of Financial Studies 15 (No. 2), 455-487. Biais, B., D. Hilton, K. Mazurier, and S. Pouget, 2005, “Judgmental Overconfidence, Self-monitoring and Trading Performance in an Experimental Financial Market,” Review of Economic Studies 72 (No. 2), 287-312. Black, F., 1986, “Noise,” Journal of Finance 41 (No. 3), 529-543. Chen, J., H. Hong, and J.C. Stein, 2001, “Forecasting Crashes: Trading Volume, Past Returns and Conditional Skewness in Stock Prices,” Journal of Financial Economics 61, 345-381. Clotfelter, C. T. and P. J. Cook, 1989, Selling Hope: State Lotteries in America, Harvard University Press, Cambridge, MA. Conlisk, J., 1993, “The Utility of Gambling,” Journal of Risk and Uncertainty 6, 255-275. Daniel, K.D., D. Hirshleifer, and A. Subrahmanyam, 1998, “Investor Psychology and Security Market Under- and Overreactions,” Journal of Finance 53 (No. 6), 1839-1886. Deaves, R., E. Lüders, and G. Y. Luo, 2004, “An Experimental Test of the Impact of Overconfidence and Gender on Trading Activity,” Working Paper. DeBondt, W.F.M. and R. H. Thaler, 1995, “Financial Decisionmaking in Markets and Firms: A Behavioral Perspective,” in R.A. Jarrow, V. Maksimovic and W.T. Ziemba (eds), Finance, Handbooks in Operations Research and Management Science, Vol. 9, North Holland, Amsterdam, chapter 13, 385-410. Deutsches Aktieninstitut, 2000, Factbook 1999, Frankfurt am Main. Dorn, D. and G. Huberman, 2005, “Talk and Action: What Individual Investors Say and What They Do, Review of Finance 9 (No. 4), 437-481.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Dorn, D. and G. Huberman, 2007, “Turnover and Volatility,” Drexel University Working Paper. Dorn, D. and P. Sengmueller, 2008, “Trading as Entertainment?” Management Science, Forthcoming. Farrell, L. and I. Walker, 1999, “The Welfare Effects of Lotto: Evidence from the UK,” Journal of Public Economics 72, 99120. Gervais, S. and T. Odean, 2001, Learning to Be Overconfident,” Review of Financial Studies 14 (No. 1), 1-27. Glaser, M. and M. Weber, 2003, “Overconfidence and Trading Volume,” University of Mannheim Working paper. Golec, J. and M. Tamarkin, 1998, “Bettors Love Skewness, Not Risk, at the Horse Track,” Journal of Political Economy 106 (No. 1), 205-225. Grinblatt, M. and M. Keloharju, 2008, “Sensation Seeking, Overconfidence, and Trading Activity,” Journal of Finance, Forthcoming. Kumar, A., 2008, “Who Gambles in the Stock Market?” Journal of Finance, Forthcoming. Merton, R.C., 1987, “A Simple Model of Capital Market Equilibrium With Incomplete Information,” Journal of Finance 42 (No. 3), 483-510. Milgrom, P. and N. Stokey, 1982, “Information, Trade and Common Knowledge,” Journal of Economic Theory 26 (No. 1), 17-27. Nadler, L., 1985, “The Epidemiology of Pathological Gambling: Critique of Existing Research and Alternative Strategies,” Journal of Gambling Behavior 1 (No. 1), 35-50. Odean, T., 1998, “Volume, Volatility, Price and Profit When All Traders are Above Average,” Journal of Finance 53 (No. 6), 1887-1934. Statman, M., 2002, “Lottery Players/Stock Traders,” Financial Analysts Journal 14-21. Tirole, J., 1982, “On the Possibility of Speculation under Rational Expectations,” Econometrica 50 (No. 5), 1163-1182. Zuckerman, M., 1994, Behavioral Expressions and Biosocial Bases of Sensation Seeking, Cambridge University Press.

The Long-Term Variation of Trade Informativeness
Michel Rakotomavo

This paper analyzes the time variation of the informativeness of trades for NYSE-listed stocks between 1998 and 2004. Trade informativeness is defined as the percentage of efficient price variance that is attributable to trades. The evidence suggests that trade informativeness was related to institutional buying and both uninformed and informed trading. The results indicate a positive relation between institutional buying and trade informativeness before Regulation Fair Disclosure and decimal pricing. After these events, the evidence is consistent with both a rise in uninformed trading and a fall in informed trading. Similar results are found for the period following the enactment of the Sarbanes-Oxley Act (SOA). While the decrease in informed trading may be a continuation of the decimalization effect, there is evidence pointing to a relation between uninformed trading and preceding-quarter institutional buying, a phenomenon that does not seem to be present before SOA.

Asymmetric information can have an important impact on microstructure, as illustrated by Whitcomb (2003). By decomposing the variance of changes in the efficient price into its trade-correlated and uncorrelated components, Hasbrouck (1991) proposed the ratio, trade informativeness, of the trade-correlated component to the total variance as a
Michel T.J. Rakotomavo is an Assistant Professor of International Business Administration at the American University of Paris in Paris, France. I thank an anonymous referee and Betty Simkins (Editor) for their constructive comments. Logistical support was provided by the Andrew Batinovich Trading Room and Research Center.

measure of the degree of asymmetric information (relative to total information) in the market for the security under study. Trade informativeness is defined as the percentage of efficient price variance that is attributable to trades (Hasbrouck (1991)). For example, this percentage for Boeing was, on average, 26.22 between the first quarter of 1998 and the last quarter of 2000. Therefore, about 26.22% of the public information on Boeing stock came from transactions on its shares during that period. While it is well known that some traders have superior information, 1 studies of the dynamics of asymmetric information among investors have focused on short time periods. For example, Zhao and Chung (2006) analyze the effect of NYSE decimal pricing on the probability of informed trading. They define November 1, 2000-January 28, 2001 as the pre-decimal period, and June 1, 2001-August 31, 2001 as the post-decimal period. They find that the post-decimal probability of informed trading is greater than its pre-decimal equivalent. Chakravarty, Van Ness, and Van Ness (2005) examine the effect of the same event on adverse selection costs. They cover the months of January and February, 2001. They conclude that the percentage adverse selection cost has increased and the dollar adverse selection cost has decreased. This paper complements the above studies by using a different information asymmetry metric, implementing trade informativeness, and focusing on a longer term (1998-2004) that includes the enactment of the Sarbanes-Oxley Act (SOA). Collver (2007) uses August 1, 1999-January 31, 2002 NYSE panel data and finds a significant decrease in daily trade informativeness after both the implementation of Regulation

see Golbe and Shranz (1994) and Karpoff and Lee (1991) for illustrations of informed traders.
1

51

52
Fair Disclosure (RFD) and the switch to decimal pricing.2 Collver assumes that the reduction is caused by less informed trading. This paper extends his finding by 1.) analyzing the temporal variation of trade informativeness in terms of microstructural variables, thus enabling a distinction between the various causes of such variation, and 2.) considering a longer term that includes the post-SOA period. For Boeing, the average quarterly trade informativeness of 26.22%, as previously mentioned, decreased to 16.77% after Regulation Fair Disclosure and decimal pricing, and went further down to 3.66% after SOA. The main results of this paper are illustrated with Boeing data in Section II. Section I develops the paper’s hypotheses on the temporal covariation between the informativeness of trades and some microstructure variables. Section II discusses the data and results. Section III concludes the paper.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Hypothesis 1: An increase in the level of institutional holdings, over time, implies an increase in the informativeness of trades, ceteris paribus. Hypothesis 2: An increase in the change of institutional holdings, over time, implies an increase in the informativeness of trades, ceteris paribus. Chiyachantana, Jain, Jiang, and Wood provide international “evidence of an increased use of order-breaking strategy” (2004, p.878) by institutions over time.3 This implies that institutions have decreased the size and increased the frequency of their trades. Therefore, if institutional buying affects the time variation of trade informativeness and institutions have an information advantage, then: Hypothesis 3: An increase in trade size, over time, implies a decrease in the informativeness of trades, ceteris paribus. Hypothesis 4: An increase in trade frequency, over time, implies an increase in the informativeness of trades, ceteris paribus. Both Nofsinger and Sias (1999) and Sias, Starks, and Titman (2006) report a positive correlation between changes in institutional holdings and contemporaneous stock returns. If institutional buying affects the time variation of trade informativeness, then: Hypothesis 5: An increase in price, over time, implies an increase in the informativeness of trades, ceteris paribus. The Kyle model (1985) suggests that a greater volatility may increase the profit of informed traders (therefore, it may attract more informed trading). Similarly, a greater market depth may increase profit. Therefore, if informed trading affects the time variation of trade informativeness, then: Hypothesis 6: An increase in the percent price range, over time, implies an increase in the informativeness of trades, ceteris paribus. Hypothesis 7: An increase in depth, over time, implies an increase in the informativeness of trades, ceteris paribus. Coughenour and Deli argue that “to the extent trading off of the NYSE represents purchased order flow and to the extent purchased order flow dries up during periods of increased informed trading, the percent of dollar volume executed at the NYSE could reflect the degree of informed trading.”4 Hence:

I. Hypotheses
Amihud and Li (2006) find that the abnormal returns at dividend change announcements is a decreasing function of institutional holdings. They also show that institutional investors use their information advantage to buy before dividend increases. Interestingly, both Nissim and Ziv (2001) and Garrett and Priestley (2000) find that only dividend increases are directly correlated with earnings. Nissim and Ziv (2001) note that managers may elect to take a “big bath” when faced with bad news by reflecting that news directly onto current earnings. Therefore, this paper hypothesizes that institutional buying may increase trade informativeness because of institutions’ information advantage. Both institutional holdings and changes in institutional holdings are used as proxies for the level of institutional buying in a stock for the reasons that follow. Sias, Starks, and Titman (2006) find that changes in institutional holdings, which represent net institutional demand, are correlated with contemporaneous stock returns because of information effects. This conclusion is consistent with holdings changes being a measure of informed buying intensity. This paper assumes that any residual institutional buying of a stock, which is not captured by net institutional demand, is correlated with the level of institutional holdings. The basis of this assumption is the herding behavior of institutions evidenced, for example, in Nofsinger and Sias (1999). Therefore, holdings levels would capture the intensity of any lagged herd net buying. Hence, if institutional buying affects the time variation of trade informativeness and institutions have an information advantage then:

see Jorgensen and Wingender (2004) for the evidence on the reaction of large corporations to RFD.
2

see Conrad, Johnson, and Wahal (2002) for a review of the theory and evidence on institutional trading.
3 4

See Coughenour and Deli (2002) p. 857.

RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS

53
.

Hypothesis 8: An increase in the percentage of NYSE executions, over time, implies an increase in the informativeness of trades, ceteris paribus.

+ f2 v2,t-1 + … + f11 v2,t-10

(2)

II. Data and Results

If w represents the innovation in the efficient price which is assumed to evolve as a random walk, trade informativeness (TINFO) is the ratio vw,x/vw, where: vw,x = (i=1,11 fi).var (v2,t). (i=1,11 fi T) ,

(3) Our sample includes 2,296 quarterly observations of 82 and randomly chosen firms for which 1998-2004 NYSE TAQ intraday trade (time, size and price) and quote (time, bid and vw = vw,x + (1+i=2,11 ei)2.var (v1,t) . (4) ask) data, as well as 1998-2004 Thomson Therefore, TINFO is a ratio Trade informativeness is defined as the Financial institutional where the efficient price shareholding data are variance attributable to trades percentage of efficient price variance that available. For a is divided by the full efficient comparison, Brooks is attributable to trades. For example, this price variance. Trade (1996) used a random informativeness is estimated percentage for Boeing was, on average, sample of 90 dividendfrom 1998-2004 NYSE TAQ paying stocks to study intraday data for each of the 26.22 between the first quarter of 1998 the variation of trade 82 firms and each of the 28 and the last quarter of 2000. Therefore, informativeness quarters between 1998 and around earnings and 2004, resulting in 2,296 about 26.22% of the public information d i v i d e n d values. on Boeing stock came from transactions announcements. More recently, Al-Suhaibani B. Other Variables on its shares during that period. and Kryzanowski (2000) used a sample Institutional holdings are of 56 stocks to study the informativeness of orders on the measured as the percentages of shares outstanding held by Saudi stock market. financial institutions, and are computed for each of the 82

A. Trade Informativeness
In Hasbrouck (1991), trades and price changes are modeled in a vector autoregression (VAR): rt = a1 rt-1 + … + a5 rt-5 + b0 xt + … + b5 xt-5 + v1t xt = c1 rt-1 + … + c5 rt-5 + d1 xt-1 + … + d5 xt-5 + v2t , (1) where rt is the mid-quote return (logarithm differentials), xt is the column vector of trade attributes, and t is the time of a transaction or a change of quote. Three trade attributes are considered: the sign of the trade, the signed trade size, and the signed squared trade size. A transaction that has a price above the prevailing quote midpoint is assigned a positive sign; the opposite holds for a negative sign. The return is set to zero if no quote revision follows a trade within 5 seconds. Transactions occurring within 5 seconds of each other without any intervening quote are aggregated. A quote posted within less than 5 seconds prior to a trade is resequenced. A moving average representation of the VAR in equation (1) is computed as follows: rt = v1,t + e2 v1,t-1 + … + e11 v1,t-10 + f1 v2,t

firms and each of the 28 quarters between 1998 and 2004. The data are from the Thomson Financial base. The change in institutional holdings in quarter q for each stock is the difference in institutional holdings between quarter q and quarter q-1. Price, trade size, price range (as a percentage of the minimum price of the day), trade frequency, volume, the percent of trades executed at NYSE, and depth are estimated daily for each stock before the quartely averages (of the daily values) are computed. This allows a comparison with the daily averages reported in the literature. The data are from the NYSE TAQ intraday trade and quote database.

C. Sample Description and Preliminary Results
Figure 1 shows the evolution of the median values of the previously mentioned variables from 1998 to 2004.5 Two notable patterns are the declining time trend for trade informativeness and the clear break in the depth data after 2001 Q1. These, and other patterns, will be investigated shortly. To aggregate values and test hypotheses, the sample
5

These values are available at http://ac.aup.edu/~mrakotomavo/JAFdata.htm.

54

Figure 1. Sample Statistics Between 1998 And 2004

TINFO is trade informativeness. IHLDG is institutional holdings. D IHLDG is change in institutional holdings. TRADEFREQ is daily trade frequency. PCNTNYSEEXEC is the percent of trades executed at NYSE.

Median of TINFO
.72 .06 .68 .04 1200 1000 .60 -.02 .56 -.04 .52 1998 1999 2000 2001 2002 2003 2004 1998 1999 2000 2001 2002 2003 2004 -.06 400 1998 1999 2000 2001 2002 2003 600 800 .00 .64 .02 1400 .08 1600

Median of IHLDG

Median of DIHLDG

Median of TRADESIZE

.5

.4

.3

.2

.1

.0

1998

1999

2000

2001

2002

2003

2004

2004

Median of TRADEFREQ
36 32 28 24 20 16 12 8 1998 1999 2000 2001 2002 2003 2004 1998 1999 1.20E+07 2000 2001 2002 2003 2004 1.40E+07 1.60E+07 1.80E+07 2.00E+07 2.20E+07 2.40E+07 2.60E+07

Median of DEPTH

Median of VOLUME
.89 .88 .87 .86 .85 .84 .83 .82 1998

Median of PCNTNYSEEXEC

1600

1400

1200

1000

800

600

400

200

0

1998

1999

2000

2001

2002

2003

2004

1999

2000

2001

2002

2003

2004

Median of PRICE
4.8 4.4 4.0 3.6 3.2 2.8 2.4 2.0 1998 1999 2000 2001 2002 2003 2004

Median of RANGE

44

40

36

32

28

24

20

16

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

1998

1999

2000

2001

2002

2003

2004

RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS

55

is divided into subperiods. Regulation Fair Disclosure (RFD) became effective on October 23, 2000. The NYSE’s decimalization was fully implemented on January 29, 2001. The Sarbanes-Oxley Act (SOA) was enacted on July 30, 2002. Therefore, three periods are considered:

to 1.87% in P3, and the time trend is still negative in P3. Therefore, SOA may have lowered trade informativeness. No other study seems to be available on this result, although Jain and Rezaee (2008) find an improvement in market liquidity after SOA. Institutional holdings have increased from a median of 1.) 1998 Q1-2000 Q4 (P1), before decimal pricing, RFD, 58.73% in P1 to 68.07% in P2. For comparison, Grinstein and SOA. and Michaely (2005) 2.) 2001 Q1-2002 Q2 report increasing median This paper presents evidence suggesting (P2), after decimal holdings for their sample, pricing and RFD, but culminating at 57.78% that the quarterly variation of the before SOA. for 1991-1996, before informativeness of trades for NYSE-listed P1. Similarly, Amihud 3.) 2002 Q3-2004 Q4 and Li (2006) have the (P3) after decimal stocks between 1998 and 2004 was related median value climbing at pricing, RFD and SOA. 54.47% in 1998 (the to institutional buying, uninformed The Jarque-Bera test beginning of P1) for their indicates a significant trading, and informed trading. The results sample. The median departure from normality change in these holdings of the panel data. This indicate a positive relation between seems to have stayed leads to the use of institutional buying and trade constant over time; in nonparametric tests particular, the P1 and P2 throughout the paper. informativeness before Regulation Fair values of .37% and .38% Panel A of Table I Disclosure and decimal pricing. After are comparable with the contains the periodic .35% average that Sias, median values and Panel these events, the evidence is consistent with Starks, and Titman B contains the periodic (2006) report for their both a rise in uninformed trading and a time trend (rank 1979 Q4-2000 Q4 correlation with a quarter fall in informed trading. sample. index running from 1 to The evidence in Panels 28) for each variable. The A and B suggests that trade size has decreased and trade hypothesis of equality of median values across each pair of frequency has increased after RFD/decimalization, and after periods is tested, using the Wilcoxon/Mann-Whitney, Median SOA. The same time variations are observed within each Chi-square, Kruskal-Wallis, and Van der Waerden statistics. period; most notably, trade size was decreasing and trade The hypothesis reports rejection when at least 3 out of the 4 frequency was increasing over time before these events . The statistics are significant at the 10%, or lower significance P1 median trade size of 1296.51 shares per day is consistent level. with the 1,500 and 1,596 group means reported in The median trade informativeness is 33.15% for P1. This Coughenour and Deli (2002) for their September-November value is comparable with Hasbrouck (1991)’s average 1997 sample. The P1 median trade frequency of 294.19 estimate of 34.3% for a sample of 177 firms on the NYSE transactions per day is consistent with the 282 average for 1989 Q1. Panel B suggests that trade informativeness (116+153+13) reported in Table 4 of Chakravarty, Van Ness, has not moved over time within P1. However, and Van Ness (2005) for their January 1, 2001-January 26, informativeness has decreased from 33.15% to 15.71% in 2001 pre-decimalization sample. The same table provides P2, after both RFD and decimalization, and the difference is evidence of an increase in the frequency of small trades that significant. Furthermore, its time trend has changed from is not offset by the observed decrease in the frequency of insignificant in P1 to negative in P2. These results are medium and large trades, after decimalization; the changes consistent with Collver (2007)’s finding of a significant in trade size and frequency between P1 and P2, shown in decrease in trade informativeness after both the Figure 1, are consistent with this evidence. However, another implementation of RFD and decimal pricing. They also agree study documenting a decrease in trade size and an increase with Chakravarty, Van Ness, and Van Ness (2005) who find in trade frequency after SOA could not be found. a reduction in dollar adverse selection after decimalization. The P1 median depth of 23.84 round lots is smaller than The same pattern is observed in P3, after SOA is enacted: the subsample averages of 36.32 and 38.30 in Table II of the median trade informativeness drops from 15.71% in P2

56

Table I. Descriptive Statistics

This table shows the descriptive statistics on the data used in the paper. The sample contains 2,296 quarterly observations, between 1998 and 2004, of 82 stocks listed on the NYSE. Trade informativeness, is the percentage of the efficient-price variance attributable to trade innovations. It is computed by following Hasbrouck (1991) and estimated for every quarter between 1998 and 2004 from NYSE TAQ intraday trade and quote data. The institutional holdings level is the number of shares held by financial institutions divided by the number of shares outstanding from the Thomson Financial database. The change in institutional holdings in quarter q for each stock is the difference in institutional holdings between quarter q and quarter q-1. Price, trade size, price range (as a percentage of the minimum price of the day), trade frequency, volume, the percent of trades executed at NYSE, and depth are estimated daily, for each stock, before the quartely averages are computed. The data are from the NYSE TAQ intraday trade and quote database. Panel A uses the Wilcoxon/Mann-Whitney, Median Chi-square, Kruskal-Wallis, and Van der Waerden tests of equality of median values across different periods. The hypothesis of equality is rejected when at least 3 out of the 4 statistics are significant at the 10% or lower level. The time index equals 1 at the start of each period. The ranking of values is repeated over all firms. 1998 Q1-2000 Q4 (P1) is a pre-decimal pricing/Regulation Fair Disclosure, pre-Sarbannes/Oxley period. 2001 Q1-2002 Q2 (P2) is a post-decimal pricing/Regulation Fair Disclosure, pre-Sarbannes/Oxley period. 2002 Q3-2004 Q4 (P3) is a post-decimal pricing/Regulation Fair Disclosure, postSarbannes/Oxley period.
Panel A. Median Values by Period
Institutional Holdings Change in Institutional Holdings 0.0053 0.0037 0.0038 643.12 1263.79 8.95 19.87 935.52 673.06 11.56 18.30 0.8720 0.8446 1296.51 294.19 23.84 17.06 0.8653 33.84 27.25 26.14 Trade Size Trade Frequency Depth (Round Lots) Volume ($Millions) NYSE Executions Price Range (%)

Period

Trade Informativeness

1998 Q1-2000 Q4 0.6149 0.6807

0.3315

0.5873

3.25 3.01 2.74

2001 Q1-2002 Q2

0.1571

2002 Q3-2004 Q4 Hypothesis No No No Yes Yes Yes No No No No No No No No No Yes No No

0.0187

P1=P2 P2=P3 P1=P3

No No No

Yes No No

No Yes No

No No No

Panel B. Rank Correlation with the Time Index by Periods
Change in Institutional Holdings -0.0481 0.1282*** 0.0948*** 0.0191 -0.1864*** -0.4995*** -0.2249*** -0.0869*** 0.3377*** 0.3623*** 0.3891*** 0.7862*** Trade Size Trade Frequency Depth Volume NYSE Executions Price Range

Period

Trade Informativeness

Institutional Holdings

0.0126

0.1160***

0.3345*** -0.3338*** 0.0783** -0.5944***

0.0373 0.0608 0.2714*** 0.2736***

-0.0908*** -0.2007*** -0.2514*** -0.2327***

-0.3512*** 0.0651 0.3843*** -0.1720***

0.4906*** -0.3765*** -0.5713*** -0.2219***

-0.2581***

0.0715

1998 Q1-2000 Q4 2001 Q1-2002 Q2 2002 Q3-2004 Q4 All periods

-0.2097***

0.3766***

-0.7210***

0.5219***

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

***Significant at the 0.01 level. **Significant at the 0.05 level.

RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS

57

Coughenour and Deli (2002). However, the significant drop to a median of 11.56 in P2 is consistent with Bessembinder (2003)’s finding of a depth reduction after decimalization. The evidence in Panel A points to an increase of depth within P1, followed by a decrease in P2. P3 shows a reversal of that trend, with an increase at a small rate. The price range has increased over time within P1, for this sample. Therefore, the P1 median range of 3.25% is consistent with the 2.66% reported in Table II of Coughenour and Deli (2002) for 1997. The decrease in range shown in Panels A and B, between P1 and P2, is in step with the decrease in return volatility documented by Bessembinder (2003) after decimalization. There seems to be a decrease also after SOA. The P1 volume of $17.06 million per day is higher than the group means of $11.99 million and $8.58 million found in Table I of Coughenour and Deli (2002). The evidence suggests that volume has not changed after RFD/ decimalization, but has increased after SOA. The median percent of trades executed at the NYSE in P1 is 86.53%. This figure is comparable with 84.01% and 85.05% subgroup averages for the 806 stocks sampled by Coughenour and Deli (2002). This percentage seems to have not changed after RFD/decimalization, but it appears to have decreased after SOA. The P1 median price of $33.84 is similar to the subgroup averages of $32.09 and $31.37 for the 806 stocks in Coughenour and Deli (2002). The median price level has gone down from P1 to P2, but does not seem to have changed from P2 to P3.

• range. • Percent of NYSE trade executions. • log(depth).
The use of log mirrors Coughenour and Deli (2002). The results are shown in the first column of Table II. The P2 variable has a negative and significant coefficient. This confirms the univariate result showing a decrease in trade informativeness after RFD/decimalization. It also agrees with Collver (2007)’s finding of a significant decrease in trade informativeness and Chakravarty, Van Ness, and Van Ness (2005)’s finding of a reduction in dollar adverse selection after decimalization. The P3 variable has also a negative and significant coefficient, again confirming the univariate result. This seems to indicate a drop in trade informativeness after the enactment of the Sarbanes-Oxley Act. Change in institutional holdings has a positive and significant coefficient. This supports the hypothesis that institutional buying leads to greater informativeness of trades. The Kyle (1985) finding that volatility is positively correlated with informed trading profit is strengthened by the positive coefficient for price range. The coefficient for trade frequency is negative and significant. This result does not support the institutional order-breaking hypothesis, which predicted a positive relation. Instead, when combined with the decrease in trade informativeness and the rise in trade frequency documented for the whole period in Panel B of Table I, it is consistent with a rise in uninformed trading between 1998 and 2004. No other coefficient is statistically significant. The previous results are further detailed by restricting the observations to each of the three periods (and removing the dummy variables). The results are shown in the last three columns of Table II. For P1, the coefficients for both level and change in institutional holdings are positive and significant. By hypotheses 1 and 2, this suggests that the time variation of trade informativeness before RFD/ decimalization has been related to institutional buying. For P2, the negative coefficient for trade frequency, the rise in trade frequency (see Panel B, Table I), and the decrease in trade informativeness (see Panel B, Table I) suggest that a rise in uninformed trading may have taken place after RFD/ decimalization. The coefficients for price level, range, and depth are positive and significant, indicating that the time variation of trade informativeness may also be related to the time variation of institutional trading (see hypothesis 5 on price level) and informed trading (see hypotheses 6 and 7 on price range and depth) after RFD/decimalization. Specifically, combined with the decline of both range and depth within that period as shown in Panel B of Table I and the decline in trade informativeness in P2, these results point to a decline in informed trading. The evidence for P2 is consistent with Chakravarty, Van Ness, and Van Ness (2005),

D. Multivariate Results
Because of departure from normality, the rank, not the value, of trade informativeness is used in the multivariate analyses that follow. Trade informativeness rank, which goes from 1 to 28, is computed for each stock. Ordered probit models are used throughout this section. All Jarque-Bera tests on residuals cannot reject the hypothesis of normality. The first model uses all observations and the following explanatory variables: • a period 2 (post-RFD/decimalization, pre-SOA) dummy variable. • a period 3 (post-RFD/decimalization/SOA) dummy variable. • institutional holdings. • change in institutional holdings. • log(trade size). • log(trade frequency). • log(price).

58

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Table II. 1998-2004 Ordered Probit Analysis of Trade Informativeness
This table shows the results of ordered-probit models of trade informativeness against the variables listed below. The sample contains 2,296 quarterly observations, between 1998 and 2004, of 82 stocks listed on the NYSE. Trade informativeness (TINFO), is the percentage of the efficient-price variance attributable to trade innovations. It is computed by following Hasbrouck (1991) and estimated for every quarter between 1998 and 2004 from NYSE TAQ intraday trade and quote data. Price, trade size, price range (as a percentage of the minimum price of the day), trade frequency, the percent of trades executed at NYSE, and depth are estimated daily, for each stock, before the quartely averages are computed. The data are from the NYSE TAQ intraday trade and quote database. The institutional holdings level is the number of shares held by financial institutions divided by the number of shares outstanding from the Thomson Financial database. The change in institutional holdings in quarter q for each stock is the difference in institutional holdings between quarter q and quarter q-1. Price, trade size, price range, trade frequency, volume, the percent of trades executed at NYSE, and depth are estimated daily, for each stock, before the quartely averages are computed. The data are from the NYSE TAQ intraday trade and quote database. The ranking of values is repeated over all firms. 1998 Q1-2000 Q4 (Period 1) is a pre-decimal pricing/Regulation Fair Disclosure, pre-Sarbannes/ Oxley period. 2001 Q1-2002 Q2 (Period 2) is a post-decimal pricing/Regulation Fair Disclosure, pre-Sarbannes/Oxley period. 2002 Q32004 Q4 (Period 3) is a post-decimal pricing/Regulation Fair Disclosure, post-Sarbannes/Oxley period. z-statistics are in parentheses.

Trade Informativeness Rank (All Periods)
Period 2 Dummy Variable Period 3 Dummy Variable Institutional Holdings -1.17*** (-14.22) -2.21*** (-19.94) -0.0335 (-0.2684) 0.8179** (2.04) 0.0219 (0.6160) -0.0711** (-2.21) 0.0574 (1.10) 0.0242** (2.39) 0.0081 (0.3259) 0.0580 (1.22)

Trade Informativeness Rank (Period 1)

Trade Informativeness Rank (Period 2)

Trade Informativeness Rank (Period 3)

0.5248*** (2.76) 0.9850* (1.73) -0.0166 (-0.3256) 0.0285 (0.5340) 0.0855 (0.9125) -0.0289 (-1.59) -0.0103 (-0.4097) 0.0373 (0.4941)

-0.4406 (-1.38) -0.0511 (-0.0553) 0.1171 (1.47) -0.2710*** (-3.73) 0.3328*** (2.93) 0.0410* (1.91) 0.4141 (0.9124) 0.3214** (2.52)

-0.7113*** (-3.02) 0.8107 (1.12) 0.0145 (0.1761) -0.0414 (-0.7020) -0.1686** (-2.07) 0.0588*** (3.79) 0.4749 (1.38) -0.1207 (-1.58)

Change in Institutional Holdings Ln(trade size)

Ln(trade frequency)

Ln(price)

Range (%)

NYSE Executions (%)

Ln(depth)

***Significant at the 0.01 level. **Significant at the 0.05 level. *Significant at the 0.10 level.

RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS

59

who find a reduction in dollar adverse selection cost after decimalization on the NYSE, an increase in the frequency of small trades (which they interpret to mean a greater participation by retail customers), smaller decreases in the frequency of medium and large trades (which they interpret to mean less institutional trading), and the strongest evidence of a decrease in adverse selection cost for trades of medium size. They suggest that institutions trade less because of lower liquidity supply (as evidenced by the smaller depths and smaller limit-order sizes), which may explain the reduction in adverse selection cost. For P3, the coefficient for price range is positive and significant, suggesting that the hypothesized link between informed trading and trade informativeness exists. Since both range and trade informativeness decreased during this period (see Panel B, Table I), this result is consistent with a decrease in informed trading after SOA. Both coefficients for price and institutional holdings are negative and significant. This result is inconsistent with hypotheses 1 (positive institutional holdings coefficient) and 5 (positive price coefficient), which

assume that institutional buying (including lagged herd buying captured by the level of institutional holdings) drives the time variation of trade informativeness. Instead, since both institutional holdings and price increased, while trade informativeness decreased, during that period, the negative coefficients are consistent with 1.) a rise in uninformed trading, and 2.) a relation between uninformed trading and lags of institutional buying after SOA. To investigate these hypotheses, a probit model, where various lags of change in institutional holdings are included as explanatory variables, is estimated for each of the three periods. Only the significant lags are reported in Table III. The results suggest that trade informativeness was not affected by lags of institutional buying in the first two periods. However, after SOA, there was a negative association between trade informativeness, and 1-quarter (and, at a weaker significance level, 2-quarter) lagged institutional buying. The negative coefficients support the hypotheses that there was a post-SOA increase in uniformed trading and that this uninformed trading was related to the 1-quarter lag of institutional buying.

The Case of Boeing
The following discussion is only meant to illustrate, not prove, the main results of the paper. As previously mentioned, the quarterly informativeness of trades for Boeing’s stock averaged 26.22% in 1998 Q1-2000 Q4 (P1), before decimal pricing, Regulation Fair Disclosure (RFD), and the Sarbanes-Oxley Act (SOA), 16.77% in 2001 Q1-2002 Q2 (P2), after decimal pricing and RFD, but before SOA, and 3.66% in 2002 Q3-2004 Q4 (P3) after decimal pricing, RFD and SOA. Therefore, trade informativeness for Boeing, like that of the average stock in this study, has decreased over time. In P1, the level of net institutional buying of Boeing shares per quarter went from an average of 0.79% of shares outstanding in the first half of the period to 1.59% in the second half. Meanwhile, Boeing’s trade informativeness went from 23.31% to 29.14%, thus illustrating the effect of institutional buying during this period. After decimal pricing and RFD, in P2, informed trading of Boeing shares became relatively less attractive, as quoted depth for Boeing stock went from an average of 17.33 round lots in the first half of P2, to 15.23 round lots in the second half (and the average was 57.32 round lots in P1). At the same time, the average number of daily trades increased from 2,471.15 to 3,366.26 (while the evidence indicates that trade size did not affect trade informativeness). As Boeing’s trade informativeness went from 21.68% to 11.86%, these figures imply that there was both an increase in the level of uninformed trading and a decrease in the level of informed trading of Boeing shares during this period. After SOA, in P3, the evidence indicates that the level of institutional holdings affected trade informativeness negatively. Since institutional holdings reflect past institutional buying, their level is a proxy for the amount of trades that are related to such past institutional actions, with a lag (the evidence in the paper points to a lag of one quarter). For Boeing, this level went from 61.65% in the first half of P3 to 63.17% in the second half. Since Boeing’s informativeness of trades dropped from 4.55% to 2.77%, the trading related to lagged institutional buying previously described was mostly uninformed. Concurrently, informed trading became relatively less attractive as the daily price range narrowed from 3.43% (of the minimum price of the day) to 2.25%, thus reducing the informed investor’s profit potential.

60

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Table III. 1998-2004 Ordered Probit Relation Between Trade Informativeness and Lagged Institutional Holdings Changes
This table shows the results of ordered-probit models of trade informativeness against lags of change in institutional holdings. The data definitions in Table II apply here. z-statistics are in parentheses.

Trade Informativeness Rank (Period 1)
Lag 1 of the Change in Institutional Holdings Lag 2 of the Change in Institutional Holdings Ln(trade size) -0.6821 (-1.04) 0.4435 (0.6939) 0.0128 (0.2434) 0.0083 (0.1431) 0.1681* (1.70) -0.0236 (-1.25) -0.0118 (-0.4622) 0.0555 (0.6806)

Trade Informativeness Rank (Period 2)
-0.7942 (-0.8589) -0.9743 (-1.03) 0.0846 (1.13) -0.2713*** (-3.73) 0.3182*** (2.81) 0.0386* (1.79) 0.1525 (0.3802) 0.3449*** (2.72)

Trade Informativeness Rank (Period 3)
-1.91*** (-2.59) -1.35* (-1.81) -0.0173 (-0.2122) -0.0627 (-1.07) -0.1793** (-2.20) 0.0583*** (3.75) 0.1129 (0.3551) -0.0820 (-1.09)

Ln(trade frequency)

Ln(price)

Range (%)

NYSE Executions (%)

Ln(depth)
***Significant at the 0.01 level. **Significant at the 0.05 level. *Significant at the 0.10 level.

III. Conclusion
This paper presents evidence suggesting that the quarterly variation of the informativeness of trades for NYSE-listed stocks between 1998 and 2004 was related to institutional buying, uninformed trading, and informed trading. The results indicate a positive relation between institutional buying and trade informativeness before Regulation Fair Disclosure and decimal pricing. After these events, the evidence is consistent with both a rise in uninformed trading and a fall in informed trading. Similar evidence is found for the period following the enactment of the Sarbanes-Oxley Act (SOA). The decrease in informed trading may be a continuation of the decimalization effect. However, there is evidence pointing to a relation between uninformed trading and institutional buying in the previous quarter, a phenomenon that does not seem to be present before SOA.

RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS

61

References
Al-Suhaibani, M. and L. Kryzanowski, 2000, “The Information Content of Orders on the Saudi Stock Market,” Journal of Financial Research 23 (No.2), 145-156. Amihud, Y. and K. Li, 2006, “The Declining Information Content of Dividend Announcements and the Effects of Institutional Holdings,” Journal of Financial and Quantitative Analysis 41(No.3), 637-660. Bessembinder, H., 2003, “Trade Execution Costs and Market Quality after Decimalization,” Journal of Financial and Quantitative Analysis 38(No .4), 747-777. Brooks, R. M., 1996, “Changes in Asymmetric Information at Earnings and Dividend Announcements,” Journal of Business Finance and Accounting 23 (No. 3), 359-378. Chakravarty, S., B.F. Van Ness, and R.A. Van Ness, 2005, “The Effect of Decimalization on Trade Size and Adverse Selection Costs,” Journal of Business Finance and Accounting 32 (No. 5, 6), 1063-1081. Chiyachantana, C.N., P.K. Jain, C. X.Jiang, and R. A. Wood, 2004, “International Evidence on Institutional Trading Behavior and Price Impact,” Journal of Finance 59 (No. 2), 869-898. Collver, C. D., 2007, “Is There Less Informed Trading After Regulation Fair Disclosure?” Journal of Corporate Finance 13 (No. 2), 270-281. Conrad, J.. K. M. Johnson and S. Wahal, 2002, “The Trading of Institutional Investors: Theory and Evidence,” Journal of Applied Finance 12 (No. 1), 7-14. Coughenour, J. F. and D. N. Deli, 2002, “Liquidity Provision and the Organizational Form of NYSE Specialist Firms,” Journal of Finance 57 (No. 2), 841-869. Garrett, I. and R. Priestley, 2000, “Dividend Behavior and Dividend Signaling,” Journal of Financial and Quantitative Analysis 35 (No. 2), 173-189. Golbe, D. L. and M. S. Shranz, 1994, “Bidder Incentives for Informed Trading Before Hostile Tender Offer Announcements,” Financial Management 23 (No. 4), 57-68. Hasbrouck, J., 1991, “The Summary Informativeness of Stock Trades: an Econometric Analysis,” Review of Financial Studies 4 (No. 3), 571 605. Jain, P.K. and Z. Rezaee, 2006, “The Sarbanes-Oxley Act of 2002 and Capital-Market Behavior: Early Evidence,” Contemporary Accounting Research 23 (No. 3), 629-654. Jain, P.K., J.C. Kim, and Z. Rezaee, 2008, “The Sarbanes-Oxley Act of 2002 and Market Liquidity,” Financial Review 43 (No. 3), 361-382. Jorgensen, R D. and J.R. Wingender, Jr., 2004, “A Survey on the Dissemination of Earnings Information by Large Firms,” Journal of Applied Finance 14 (No. 1), 77-84. Karpoff, J.M. and D. Lee, 1991, “Insider Trading Before New Issue Announcements,” Financial Management 20 (No. 1), 18-26. Kyle, A.S., 1985, “Continuous Auctions and Insider Trading,” Econometrica 53 (No. 6), 1315-1335. Lee, C. M.C. and M.J. Ready, 1991, “Inferring Trade Direction from Intradaily Data,” Journal of Finance 46 (No. 2), 733-746. Nissim, D. and A. Ziv, 2001, “Dividend Changes and Future Profitability,” Journal of Finance 56 (No. 6), 2111-2133. Nofsinger, J.R. and R.W. Sias, 1999, “Herding and Feedback Trading by Institutional and Individual Investors,” Journal of Finance 54 (No. 6), 2263-2295. Sias, R.W., L.T. Starks, and S. Titman, 2006, “Changes in Institutional Ownership and Stock Returns: Assessment and Methodology,” Journal of Business 79 (No. 6), 2869-2910. Whitcomb, D.K., 2003, “Applied Market Microstructure,” Journal of Applied Finance 13 (No. 2), 77-80. Zhao, X. and K.H. Chung, 2006, “Decimal Pricing and InformationBased Trading: Tick Size and Informational Efficiency of Asset Price,” Journal of Business Finance and Accounting 33 (No. 5,6), 753-766.

62

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Shareholder Theory – How Opponents and Proponents Both Get It Wrong
Morris G. Danielson, Jean L. Heck, and David R. Shaffer

Shareholder wealth maximization is accepted by most financial economists as the appropriate objective for financial decision-making. Recently, wealth maximization has been criticized by a growing array of opponents for condoning the exploitation of employees, customers, and other stakeholders, and encouraging short-term managerial thinking. Although these critics are misguided, proponents of shareholder theory have helped to create this confusion by exhorting managers to maximize the firm’s current stock price. Because a firm’s stock price can be manipulated in the shortterm, incentives to increase a firm’s current stock price can distort operating and investment decisions. When wealth maximization is properly defined as a long-term goal, it is not as narrowly focused as critics believe. The main prescription of shareholder theory—invest in all positive net present value projects—benefits not only shareholders, but also key stakeholders including employees and customers.

Shareholder theory defines the primary duty of a firm’s managers as the maximization of shareholder wealth (Berle and Means, 1932; Friedman, 1962). The theory enjoys widespread support in the academic finance community and
Morris G. Danielson is an Associate Professor of Finance at Saint Joseph’s University in Philadelphia, PA. Jean L. Heck is an Associate Professor of Finance at Saint Joseph’s University in Philadelphia, PA. David R. Shaffer is an Associate Professor of Finance at Villanova University in Villanova, PA. Danielson and Heck gratefully acknowledge financial support from the Pedro Arupe Center for Business Ethics at Saint Joseph’s University.

is a fundamental building block of corporate financial theory. However, the shareholder model has been criticized for encouraging short-term managerial thinking and condoning unethical behavior. Smith (2003) notes that critics believe shareholder theory is “. . . geared toward short-term profit maximization at the expense of the long run.”1 Freeman, Wicks, and Parmar (2004) assert that shareholder theory “. . . involves using the prima facie rights claims of one group— shareholders—to excuse violating the rights of others.” This paper explains why such critiques of shareholder theory are misguided yet understandable. They are misguided because wealth maximization is inherently a long term goal— the firm must maximize the value of all future cash flows— and does not condone the exploitation of other stakeholders (Jensen, 2002; Sundaram and Inkpen, 2004a). The criticisms are understandable because many proponents of shareholder theory, in a stylized version of the model, exhort managers to maximize the firm’s current stock price (Keown, Martin, and Petty, 2008; Lasher 2008; Ross, Westerfield, and Jordan, 2008; Brealey, Myers, and Marcus, 2007; Melicher and Norton, 2007). This notion underlies the formal (e.g., stock options) and informal (e.g., pressure from the investment community and corporate boards) incentives that reward managers if a firm’s stock price continually increases.2 By
For example, Freeman, Wicks, and Parmar (2004) criticize managers for pursuing policies designed to continually increase a firm’s stock price. Fuller and Jensen (2002) criticize mangers for focusing undue attention on whether a firm meets analyst earnings forecasts each quarter, to avoid stock price declines.
1

Although incentive stock options typically vest over several years and can have long maturities, the presence of stock options also encourages managers to pursue policies designed to increase the stock price in the short-term (especially as the expiration date approaches). Danielson and Press (2006) argue that these incentives can create agency costs whenever the stock price falls below the option exercise price.
2

62

DANIELSON, HECK, AND SHAFFER — SHAREHOLDER THEORY

63
Although shareholder theory directs managers to maximize shareholder wealth, managers face formal and informal incentives to increase the firm’s current stock price. For example, incentive stock options will provide a positive payoff to managers only if the firm’s stock price increases from the grant date level. In addition, some managers face pressure from corporate boards and the investment community to continually increase firm value (Jensen, 2005). However, maximizing and increasing shareholder wealth are two very different objectives. If the business conditions facing a firm change unfavorably (through perhaps no fault of management), a firm’s maximum possible value can decrease. This is not an unusual or unlikely occurrence; Jensen (2005) notes that future events could reveal that the stock prices of perhaps 50% of all firms are too high (because a stock price is a function of a distribution of possible outcomes). As the business conditions facing a firm change, a firm’s stock price can diverge from its intrinsic value because information is not instantaneously and continuously communicated to the market. If business conditions change unfavorably, P 0 will exceed V 0 and the stock will be (temporarily) overvalued.4 To implement the shareholder model correctly, the firm should continue to invest in all positive NPV projects (which are now less valuable than the market originally expected), and the stock price will eventually decrease to the new intrinsic value. However, if managers (who will typically know that business conditions have changed before the rest of the market) are incentivized to increase the stock price, Jensen (2005) and Danielson and Press (2006) argue that efforts to further inflate (or to maintain) the stock price may destroy long-term value. These actions could include delaying new investments (even if the NPV is positive), reducing discretionary spending (e.g., advertising, R&D, maintenance, quality control, etc.), accounting manipulation, or adopting fraudulent business practices. Jensen (2005) uses Enron to illustrate the agency costs of overvalued equity. At Enron’s peak market value of $70 billion, Jensen estimates the company was only worth $30 billion. He notes that Enron’s managers tried to justify the excess valuation of $40 billion by “. . . trying to fool the markets through accounting manipulations, hiding debt through off-balance sheet partnerships, and over hyped new ventures such as their broadband futures effort.” Clearly, these efforts were not designed with the long-term interests of the firm in mind, and they did not pay off for Enron’s shareholders. Thus, the case of Enron does not provide evidence against shareholder theory. But this experience does
Although deviations between P0 and V0 can arise in the short-term even in efficient markets, evidence in Summers (1986) and Cornell (2001) suggest that such deviations can persist for prolonged periods.

focusing on the current stock price, which can be manipulated in the short-term by unscrupulous managers, proponents of shareholder theory open up the model to criticism. Opponents of shareholder theory often recommend that firms balance the interests of shareholders against those of employees, customers, and other stakeholders when making business decisions (Freeman, 1984). However, unless the interests of future stakeholders are explicitly considered, the stakeholder model can lead to the same type of short-term thinking that shareholder theory has been accused of encouraging. Indeed, the shareholder model, when viewed from a long-term perspective, provides a better framework than stakeholder theory in which to protect the interests of both current and future stakeholders. Thus, stakeholder theory is not superior to shareholder theory from an ethical perspective.

I. Should Firms Maximize the Current Stock Price?
In the shareholder model, the goal of the firm is to maximize the present value of future cash flows. If the cash flow a firm is expected to pay shareholders (in the form of dividends or stock repurchases) in year n is CFn, and the required return on equity is r, the intrinsic (per share) value of the firm’s equity today (V0) is defined by Equation (1).

V0 = ∑

∞

n =1

(1 + r )n

CFn

.

(1)

To maximize the value of Equation (1), managers should invest in all positive net present value (NPV) projects (Brealey and Myers, 2003). The right-hand side of Equation (1) highlights the long-term nature of this goal: shareholder wealth depends on the firm’s cash flows in all future years.3 The shareholder model is difficult to implement because the estimated cash flow stream on the right-hand side of Equation (1) cannot be observed. Thus, proponents of shareholder theory often assert that a firm’s current stock price (P0) equals its intrinsic value (V0) and instruct managers to maximize the firm’s current stock price. This is the stylized form of the shareholder model.

A large portion of shareholder wealth is often tied to cash flows to be received in the distant future. For example, if the firm is expected to pay a $1 dividend next year, and the dividend is expected to grow at a 4% rate per year (forever), the stock price today is $25 if the required return is 8% (= $1/(0.08 – 0.04)). In this example, dividends during the next 10 years only account for 31.4% of the stock price (= $7.86/25), leaving 68.6% of the value to be realized in years 11 through infinity. Clearly, shareholder wealth maximization is not a short-term goal.
3

4

64
show that efforts to increase a firm’s current stock price can be harmful if these policies are detached from strategies designed to maximize the firm’s long term cash flows.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

example, 10%. However, this policy would reduce both the funds available to invest in research and development and the incentive for firms to do so. Thus, future customers would not benefit from potential life-saving products that might otherwise have been developed. II. Does Stakeholder Theory Promote a The following example illustrates the potential problem. Long-Term Focus? Assume that a firm operates in a simple one-period world. The entrepreneur invested $100 in the firm at t = 0, and the Because of the perceived firm produces a cash flow of deficiencies of shareholder $160 at t = 1. If the required If a firm is forced to allocate a portion theory, stakeholder theory return is 10%, the economic has gained popularity in surplus of the firm is $50 (= of its economic surplus to employees recent years and is now used $160 – $100(1.10)). Because (by paying wages in excess of the to guide the business the firm has a realized decisions of a wide range of investment return of 60%, employees’ marginal productivity) or firms (Donaldson and stakeholder advocates might Preston, 1995; Jorg, to customers (by reducing prices), argue that the shareholders’ Loderer, and Roth, 2004; profits are excessive. From these stakeholders will benefit in the and Kaler, 2006). One of the their perspective, an equitable goals of stakeholder theory distribution of the economic short-term. However, these policies is to promote “an surplus might be to increase could stifle future innovation, hurting enhancement of distributive wages or decrease prices, justice within the confines of reducing the investment shareholders, stakeholders, and a basically capitalist return toward the required structure . . . .” (Kaler, 2006). society in the long-run. return of 10%. But, this Along these lines, the 1988 outcome would not be fair to Sloan Colloquy in its “Consensus Statement on Stakeholder the entrepreneur unless the policy were known before the Model of the Corporation” recommends that firms “attempt investment decision was made. to distribute the benefits of their activities as equitably as Most investments offer risky outcomes; it is likely that the possible among stakeholders, in light of their respective entrepreneur did not know with certainty that the project’s t contributions, costs, and risks.”5 To do this, Blair and Stout = 1 payoff would be $160 when the initial $100 investment (1999) argue that the board of directors should split a firm’s was made at t =0. Assume that at time t = 0, the investment economic surplus (i.e., investment returns in excess of the had an equal 50% probability of paying either $160 or $60 at risk-adjusted cost of capital) between shareholders, t = 1. If so, the expected payoff at t = 1 was $110, and the employees, customers, and other stakeholders. project had a net present value of 0 (= 110/1.10 – 100). On If a firm is forced to allocate a portion of its economic an ex-ante basis the project was acceptable, but it did not surplus to employees (by paying wages in excess of the create an economic surplus. employees’ marginal productivity) or to customers (by Once the future outcome is revealed, it would not be ethical reducing prices), these stakeholders will benefit in the short- to change the rules of the game and split the excess return term. However, these policies could stifle future innovation, ($160 – $110) between shareholders and other stakeholders. hurting shareholders, stakeholders, and society in the long- If the entrepreneur had known at t = 0 that the project would run. For example, US employees in the steel industry, the only yield, for example, $150 in the good outcome, the auto industry, and the airline industry benefited in the short- entrepreneur would not have made the $100 investment. term from lucrative union contracts negotiated in the latter Thus, proposals to split the realized economic surplus among half of the twentieth century. But these contracts ultimately various stakeholder constituencies have the potential for contributed to financial difficulties at the firms, reducing job reducing future investment, harming society (and potential security and compensation for today’s employees. Similarly, future stakeholders) in the long run. the current customers of pharmaceutical companies would Stakeholder theory, of course, does not advocate that firms benefit greatly if patent laws were revoked, and all drugs be managed in the interests of current stakeholders at the were then sold at a price equal to production costs plus, for expense of future ones. Instead, Freeman (1994) recommends that a corporation “. . . be managed as if it can continue to serve the interests of stakeholders through time.” Similarly, 5 This statement is reprinted in the appendix to Marcoux (2000).

DANIELSON, HECK, AND SHAFFER — SHAREHOLDER THEORY

65
Parmar (2004), “It is hard to imagine how anyone can look at the recent wave of business scandals, all of which are oriented toward ever increasing shareholder value at the expense of other stakeholders, and argue that this philosophy is a good idea.” However, proponents of shareholder theory point out that policies adopted by Enron, Worldcom, and Global Crossing clearly did not benefit the firms’ shareholders in the long-run, and thus are not evidence against shareholder theory (Sundaram and Inkpen, 2004b). Before dismissing critics of shareholder theory outright, it is important to recognize that supporters of shareholder theory often emphasize the model’s short-term implications when defining the theory. Indeed, many leading finance texts equate shareholder theory with the maximization of a firm’s current stock price, and executive compensation (e.g., incentive stock options) frequently rewards managers for increasing the stock price. Thus, it should not be surprising that some critics of the shareholder theory might (incorrectly) view it as being a short-term goal. We disagree, however, with those who would use the deficiencies of the stylized model as a reason to abandon shareholder theory in favor of stakeholder theory. Despite its current popularity, stakeholder theory provides little guidance about how to balance the often competing interests of various stakeholder groups (Marcoux, 2000; Jensen, 2002). In addition, stakeholder theory can encourage managers to adopt a short-term focus (much like the stylized version of the shareholder model) to the detriment of a firm’s long-term health. The shareholder model—when viewed from a long term perspective—still provides the best framework in which to balance the competing interests of various stakeholders (including both current and future stakeholders) when making business decisions. However, proponents of shareholder theory must recognize that it matters how the theory is defined and implemented. In particular, the goal of financial managers should be to invest in all positive net present value projects, regardless of whether these decisions will cause an immediate increase in the firm’s stock price. To focus managerial attention on this goal, corporate incentive structures should reward managers for maximizing a firm’s value in the long run rather than increasing its stock price in the short term.

DesJardins and McCall (2005) argue that a corporation should be managed as a social institution, providing benefits to stakeholders both now and in the future. However, the question of how a manager might balance the interests of current and future stakeholders has received very little attention in the stakeholder literature. One notable exception is Mitchell, Agle, and Wood (1997), who argue that managers should consider the urgency of various stakeholder claims when making decisions. But this approach would encourage managers to adopt a short-term focus when implementing stakeholder theory: the needs and requirements of current stakeholders will always be more “urgent” than those of future stakeholders.

III. The Shareholder Model and Long-Term Stakeholder Interests
One drawback of stakeholder theory is that the identity of the individual stakeholders is constantly changing. Thus, the customer or employee who extracts excess benefits from a firm during the current period is not the same person who loses future benefits. The identity of shareholders will also change over time, but there is a key difference. A large portion of any investor’s return (even a short-term trader) will depend on the firm’s stock price on the date of sale. Because an investor must find a person who believes the firm will produce sufficient cash flows to justify the prevailing market price, shareholder wealth maximization (when defined properly as a function of all future cash flows) is inherently a long-term goal. And, because a firm must continue creating value for employees and customers to generate future cash flows, the maximization of a firm’s long-term cash flow stream should not harm the firm’s stakeholders. Indeed, the interests of future stakeholders can only be satisfied if the firm remains financially strong.

IV. Conclusion
In the aftermath of financial scandals at Enron, Worldcom, and Global Crossing, shareholder theory faces increased scrutiny and criticism. As stated by Freeman, Wicks, and

66 References
Berle, A.A. and G.C. Means, 1932, The Modern Corporation and Private Property, Macmillan, New York. Blair, M.M. and L.A. Stout, 1999, “A Team Production Theory of Corporate Law,” Virginia Law Review 85 (No.4), 247-328. Brealey, R.A. and S.C. Myers, 2003, Principles of Corporate Finance, McGraw-Hill/Irwin, New York.. Brealey, R.A., S.C. Myers, and A.J. Marcus, 2007, Fundamentals of Corporate Finance, McGraw-Hill/Irwin, New York. Cornell, B., 2001, “Is the Response of Analysts to Information Consistent with Fundamental Valuation? The Case of Intel,” Financial Management 30 (No.1), 113-136. DesJardins, J.R. and J.J. McCall, 2005, “The Corporation as a Social Institution,” In Contemporary Issues in Business Ethics, 5th edition, Edited by J.R. DesJardins and J.J. McCall, Wadworth, Belmont, CA. Danielson, M. and E. Press, 2006, “Do Stock Options Always Align Manager and Shareholders’ Interests? An Alternative Perspective,” Advances in Financial Education 4 (No.2), 1-16. Donaldson, T. and L.E. Preston, 1995, “The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications,” Academy of Management Review 20 (No.1), 65–91. Freeman, R.E., 1984, Strategic Management: A Stakeholder Approach, Pitman, Boston. Freeman, R.E., 1994, “The Politics of Stakeholder Theory: Some Future Directions,” Business Ethics Quarterly 4 (No.4), 409–421. Freeman, R. E., A. C. Wicks, and B. Parmar, 2004, “Stakeholder Theory and ‘The Corporate Objective Revisited’,” Organization Science 15 (No.3), 364–369. Friedman, M., 1962, Capitalism and Freedom, University of Chicago Press, Chicago. Fuller, J. and M. Jensen, 2002, “Just Say No to Wall Street: Putting A Stop to the Earnings Game,” Journal of Applied Corporate Finance 14 (No.4), 41-46. Jensen, M.C., 2002, “Value Maximization, Stakeholder Theory, and the Corporate Objective Function,” Business Ethics Quarterly 12 (No.2), 235-256.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Jensen, M.C., 2005, “Agency Costs of Overvalued Equity,” Financial Management 34 (No.1), 5-19. Jorg, P., C. Loderer, and L. Roth, 2004, “Shareholder Value Maximization: What Managers Say and What they Do,” DBW Die Betriebswirtschaft 64 (No. 3), 357-378. Kaler, J., 2006, “Evaluating Stakeholder Theory,” Journal of Business Ethics 69 (No.2), 249–268. Keown, A.J., J.D. Martin, and J.W. Petty, 2008, Foundations of Finance, Pearson Prentice Hall, Upper Saddle River. Lasher, W.R., 2008, Practical Financial Management, Thomson South-Western, Mason. Marcoux, A.M., 2000, “Balancing Act,” In Contemporary Issues in Business Ethics, 4 th edition, edited by J.R. DesJardins and J.J. McCall, Wadworth, Belmont, CA. Melicher, R.W. and E.A. Norton, 2007, Introduction to Finance, John Wiley and Sons, Hoboken. Mitchell, R.K., B.R. Agle, and D.J. Wood, 1997, “Toward a Theory of Stakeholder Identification and Influence: Defining the Principle of Who and What Really Counts,” Academy of Management Review 22 (No.4), 853-886. Ross, S.A., R.W. Westerfield, and B.D. Jordan, 2008, Fundamentals of Corporate Finance, McGraw-Hill/Irwin, New York. Smith, H.J., 2003, “The Shareholders vs. Stakeholders Debate,” MIT Sloan Management Review 44 (No. 4), 85– 90. Summers, L.H., 1986, “Does the Stock Market Rationally Reflect Fundamental Values?” Journal of Finance 41 (No. 3), 591-601. Sundaram, A. and A. Inkpen, 2004a., “The Corporate Objective Revisited,” Organization Science 15 (No.3), 350–363. Sundaram, A. and A. Inkpen, 2004b., “Stakeholder Theory and ‘The Corporate Objective Revisited’: A Reply,” Organization Science 15 (No.3), 370–371.

Student Managed Investment Funds: An International Perspective
Edward C. Lawrence

The most comprehensive survey ever conducted on student managed investment funds shows there are now 314 universities worldwide that offer students the chance to learn about portfolio management by investing real money. In aggregate, students are directly managing more than $407 million in assets in 2007. Most of these programs supplement the more traditional investment courses, which are offered by every institution with a business college. Over the last two decades, student-managed investment funds have grown in both size and complexity as universities have tried to mirror real world experiences. The career success of students coming out of these programs demonstrates the benefits of providing students with as much hands-on experience as possible. This paper should be of interest to faculty, students, employers, and practitioners in the financial community who desire basic knowledge about state-of the-art teaching investments and portfolio management.

In the early 1970s, there was a strong movement in Western countries for universities to start providing students with both academic knowledge and the ability to apply new skills on the job. Employers were often critical of new graduates who had difficulty stepping into employment without first receiving extensive on-the-job training. To
Edward C. Lawrence is a Professor of Finance and Department Chair at the University of Missouri - St. Louis in St. Louis, MO. The author would like to acknowledge the assistance of Kerry Sallee, Ken Locke, Karen Wagster, Anthony Lerro, Brian Bruce, Larry Belcher, and all of the university faculty participants who generously gave their time to complete this survey.

overcome this obstacle, many business colleges began partnerships with major companies to offer students co-op and internship programs while the students were still pursuing their degrees. Deans also started encouraging their faculty to invite more guest speakers from government and industry to address classes on issues of the day. It also became common for professors to take classes on field trips to local employers to gain greater insights as to what it was like to work in a particular field. Finally, with the development of computers, interactive software became more prevalent, allowing students to simulate starting a new company, managing a bank, or investing in the stock market with play money.1 Although all of these approaches were a significant improvement over what educational institutions had done historically, there was still a need to offer students even greater realism and more practical experience.2 In the field of finance, student managed investment funds (SMIFs) were created to take investment education to the next level. These funds allow students to invest real money in the stock and bond markets. The vast majority of SMIFs have close faculty involvement to provide oversight and structure to student activities. Nevertheless, students are generally responsible for making all investment decisions and

Most basic investment courses today use simulations or play money to allow students some practice with security analysis, stock selection, portfolio composition and market timing. However, it is well recognized that play money often leads to excessive risk taking as students try to outperform the market without realistic penalties.
1 2

Even now, Pfeffer (2007) still argues that business schools are still not doing enough to ensure students can “translate business knowledge into applicable business skills” in real world situations. However, he does not specifically address the effectiveness of many of the initiatives mentioned above.

67

68
managing the portfolio. Some funds have outside professionals that serve in an advisory capacity to enrich the experience for students and showcase their programs. While SMIFs were first started in the US and have been around since 1950, only 12 colleges had them by 1969. Unfortunately, those programs were not widely known outside of those campuses. Today, there are 314 funds worldwide ranging in size from $2,000 to $62 million. In aggregate, students are directly managing $407 million in assets. The purpose of this paper is to discuss the evolution of SMIFs and the impact they are having on teaching investments around the globe. Surprisingly, there has not been a single paper presenting data on funds from outside the United States. There is much to be gained within academia and the investment community through the sharing of ideas and information on financial education in a variety of cultures and environments. Not only do new and existing SMIFs learn from the innovations of other successful programs, faculty and administrators benefit from considering a broader array of approaches to solving specific constraints faced by a particular school. Students desiring to embark on investment careers will be able to more fully evaluate the different programs being offered and select the one that most closely fits their learning style. Finally, employers and practitioners also need to become more knowledgeable of the various types of SMIFs. Besides the opportunity to hire highly trained students coming out of these programs, professionals should become educational partners by providing guest speakers, serving on boards, providing funding, sharing technical resources, etc.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

impossible to fully understand the scope of this movement in investment education without a broader database. Furthermore, many senior university officials were still reluctant to commit their scarce resources in such funds without convincing hard data showing the clear benefits and costs of such programs. By the early 1990s, with so many leading business schools embracing the basic concept, it became an “easy sell” for finance faculty and alumni in North America. This led to an explosion of programs, which have spread to other continents including Asia and Europe. In 1994, Lawrence expanded his study to include 34 programs in order to better describe their operations and funding sources. Johnson, Alexander and Allen (1996) investigated alternative decision making environments in student managed funds. By 2003, Neely and Cooley (2004) reported 134 funds had been established in the US alone. Ammermann and Runyon (2003) investigated risk aversion and group dynamics among students making portfolio decisions at California State University in Long Beach. All of these papers served as a major catalyst for the rapid growth in the number and size of SMIFs worldwide, especially in North America.

II. The Survey
From June 2007 to April 2008, a written survey was electronically sent to all universities in the US and abroad with both known and possible SMIFs.4 For US participants, survey participants were also asked to share their knowledge of other programs in their states. In the case of foreign countries, participants were solicited for information on existing and potential funds in their own country or neighboring countries. It was assumed that faculty involved with current programs would most likely know of other SMIFs from their professional contacts at conferences. Since locating foreign programs would be more challenging, the meeting roster of attendees at the 2007 Financial Management Association meeting in Florida was used to contact a large number of faculty from South America, Europe, Australia and Asia. This database was supplemented by the author who attended major academic conferences in both Europe and Asia during this time period in order to make personal contact with other finance faculty who may have had knowledge of foreign funds. Finally, a significant number of contacts were made with finance department chairs and deans at major

I. Previous Studies
One of the primary reasons student investment funds were slow to be established in the 1970s was the lack of organizational information and data on the benefits and costs of these programs. Until 1990, there was not even a list of universities in the US that had such funds. In fact, many of the faculty who were closely involved in SMIFs prior to this time had limited knowledge of other programs and almost no communication with their colleagues. As a result, it was very difficult for finance faculty to start new funds given the lack of operational data and instructional inexperience with such programs. Recognizing this problem, Lawrence (1990) conducted the first survey to profile and discuss the characteristics of almost two dozen established programs. Until then, it was common for paper authors to only describe a single fund.3 While these efforts were insightful, it was
Some of the earliest case studies included Belt (1975), Hirt (1977), Bear and Boyd (1984), Markese (1984), Kester (1986), Tatar (1987), Block and French (1991), Bhattacharya and McClung (1994) and Kahl (1997).
3

Anthony Lerro of the Association of Student Managed Investment Funds (ASMIF) graciously provided an initial list of American universities that were members of the association. While the list provided a good starting point for the US, it also contained a fair number of schools that did not yet have a program and was missing a large number of other universities that did.
4

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

69

business schools (not included in any of the previously than in the US. In other parts of the world, it could easily mentioned screens) in Europe and Asia to make sure there take another 20 years to catch up to North America. was as much international exposure as possible, given the The size of SMIFs today has expanded beyond what many obvious language barriers. people would have thought possible only a decade ago. There This protocol resulted in locating 314 SMIFs from around are 78 universities worldwide with more than $1 million under the world. Appendices A and B provide a complete list of all management by students. As reported in Table II, the largest US and non-US programs, respectively. Each program was fund is at the University of Wisconsin - Madison which has contacted up to 7 times by $62 million being invested in email and/or phone to some form by students.6 There In the field of finance, student encourage their participation are 8 SMIFs with more than in the survey. With faculty $10 million including 2 funds, managed investment funds (SMIFs) advisors frequently rotating Ohio State and the University were created to take investment in and out of the programs, it of Minnesota, with $25 million was sometimes very difficult each in assets. 7 The largest education to the next level. These to find the appropriate person non-US fund is Canada’s with sufficient knowledge to Simon Fraser University with funds allow students to invest real complete the 48 question $10 million. However, few money in the stock and bond survey. In addition, a small faculty members would argue number of the funds operate that a multi-million dollar markets. more like investment clubs portfolio is necessary to have with little or no faculty involvement. Yet, these funds give a successful program. The incredible expansion in fund size the students much of the same experience of investing real is even more impressive when one considers that most of the money, but as an extracurricular activity. Of the 314 funds, funds started with only $100,000 or $200,000 in initial seed 224 programs returned the completed survey for a 71% capital. Of course, almost all of the SMIFs continued to participation rate. For the remaining 90 SMIFs who declined receive additional investment capital from various sources to participate, summary information about their programs was as they demonstrated an ability to manage the money obtained from external sources including the university’s web successfully. site and media sources. One of the more interesting growth patterns for SMIFs is how widely they are being used in a broad range of educational environments. While 99% of all current SMIFs are housed III. The Growth and Size of Programs within business schools, there are a few exceptions. For After 40 years of very slow growth, the number of SMIFs example, Tufts University, without a business college, has a in the US exploded in the 1990s and 2000s as real money $1 million student fund that focuses on investing in funds began to supplement the more traditional methods of biomedical companies. Besides the traditional university teaching investments. As reported in Table I, the 1990s was undergraduate and graduate business students, there are the turning point. To remain competitive in the marketplace, several high schools that broke new ground by adapting the most business schools had to offer students the opportunity same learning8principles with students less prepared in finance to invest in the stock market with real money.5 The current and business. This is part of a broader trend where subjects decade has experienced the highest number of new programs created despite the data only including seven years. With 1,680 business schools in the US according to the AACSB, the country may be a long way from reaching a saturation point. However, with the exception of Canada, SMIFs are just in the early stages of development in the rest of the world. The first non-US fund was established at the University of British Columbia in 1987. Canada actually has a higher concentration of funds within institutions of higher learning
The University of Wisconsin - Madison actually has five distinct funds. Only one portfolio is invested in equity securities. The other four portfolios are fixed income portfolios, which provide a different set of learning experiences depending on the objectives of the fund. The largest portion of the fund is private money managed for clients based on set investment criteria.
6

The average size for US funds was $1.4 million and $1.2 million for nonUS programs.
7

A university having a student investment fund has become the gold standard for investment programs at all levels. As one faculty member stated, “One cannot have a top 10 MBA program today without it.”
5

It has been reported that the following American schools at one time had active funds: Dominican High (WI), Groton High (MA), Wisconsin Lutheran (WI), Jenks High (OK), Ariel Community Academy (IL), West Allis Central (WI), Gaithersburg High (MD), and Burnsville High (MN). These scaled down programs closely mirror those established at the university level in the way they are structured and operate.
8

70
Table I. Growth in New Funds

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

160 140 120

Number of New Funds

100 80 60 40 20 0 1950s 1960s 1970s 1980s 1990s 2000s

Decade

Table II. The Largest Funds (in US Dollars)
Panel A. US Universities (2007)

Rank
1 2 3 4 5 6 7 8 9 & 10 Tie

Institution
University of Wisconsin Ohio State University University of Minnesota University of Utah University of Texas Cornell University University of Arkansas University of Houston Baylor University Southern Methodist Panel B. Non- US Universities

Country
US US US US US US US US US US

Total Assets
$62.0 Million $25.8 Million $25.0 Million $18.2 Million $17.0 Million $13.5 Million $12.0 Million $9.2 Million $6.5 Million $6.5 Million

Rank
1 2 3 4 5 6 7

Institution
Simon Fraser University HEC Montreal Univ. of British Columbia Queens University Univ. Of New Brunswick Concordia University University of Alberta

Country
Canada Canada Canada Canada Canada Canada Canada

Total Assets
$10.0 Million $3.8 Million $3.5 Million $3.0 Million $2.2 Million $1.4 Million $1.3 Million

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

71
unlimited number of reasons that could not have been accurately forecasted a year or more in advance. However, unlike practitioners in real life, students lack a strong incentive system of monetary rewards for beating benchmarks or penalties for poor performance (being fired). Student portfolios often have constraints that most professionals do not have. For example, most SMIFs are structured to rely on group or committee decisions rather than those of a single portfolio manager. Sometimes there are more than 30 students involved with various levels of skill and knowledge. All have an equal vote in the ultimate decision. Depending on the quality of the student research and group presentation skills, decisions are not always based entirely on objective analysis. In addition, the majority of funds offer a one or two semester class that may encourage students to use short-term planning horizons since they may not be around to witness the final outcome of any particular investment. Obviously, it takes several economic cycles to really evaluate the success of any investment strategy. The fund at the University of Missouri - St. Louis is one of the exceptions with a credit program structured to allow students to participate for several years (see Table III for a summary). Professional managers can react almost instantaneously to rapidly changing market conditions without the need to assemble the group for a vote. This factor alone should favor practitioner performance, provided they are not trading on unfounded rumors and there really are fundamental market changes taking place. Professionals can also trade on margin or use derivatives to enhance returns which are not widely available techniques for the majority of SMIFs. Offsetting some of these advantages, professionally managed funds must absorb all of their own operating expenses, whereas most SMIFs get subsidized resources (e.g. facilities, computers, faculty salary) from the universities and rarely pay all expenses related to the fund.

that used to be taught only at the college level are now being introduced in high schools. According to McInerny (2003), some of the stimulus is being provided by Paul O’Neill, a former US Treasury Secretary. O’Neill has been aggressively promoting greater financial skills at the primary and secondary education level. It seems clear to faculty participating in these programs that real money portfolios have dramatically increased student interest in majoring in finance or business in college. A surprising finding of this survey is that at least six institutions in the US have allowed their SMIFs to become inactive over the past few years.9 Given how difficult it is for many schools to establish SMIFs and their popularity with students and employers, this was unexpected. In personal discussions with faculty and administrators at these schools, the most common reason the program became inactive was the loss of the key faculty member (often due to retirement) who advised the students. Other finance faculty were unwilling to take over the fund, partially due to the greater amount of time it takes to stay abreast of the financial markets on a daily basis. With many schools emphasizing research productivity for promotion and raises, it is most onerous for tenure track faculty to lead the fund activities for more than a few years. Of course, many schools have dealt with this issue by hiring nontenure track faculty or adjuncts to run the SMIFs.

IV. SMIFs versus Professionally Managed Funds
The central goal of SMIFs is to create a realistic learning environment for training the next generation of portfolio managers. Unlike professionally managed funds, which are solely focused on generating the highest risk-adjusted rates of return possible, SMIF returns are secondary in nature to the educational mission. 10 Faculty advisors generally recognize that some of the best learning experiences come from failures, not successes per se. As any experienced investor knows, there is always an element of luck and incomplete data behind any decision. Thus, a very carefully analyzed opportunity with great potential can fail for an almost

V. Funding Sources and Organizational Structure
The majority of older SMIFs received ear-marked money from alumni and other private donors to establish the funds. Twenty-eight percent of the funds got all of the money from the university’s own endowment. Another 23% of schools had only a single large donor. The balance of programs was a combination of capital sources, including many small donors and corporate donations. For universities wishing to establish a new fund, the average program in the US during the 2000s was started with approximately $414,000 in initial capital. The most common form of organizational structure is having the SMIF be part of the university endowment. About 62% of all funds are structured this way. Another 14% are set

These universities include the University of Central Florida, Southern Illinois University - Edwardsville, the University of Florida, the University of Missouri - Kansas City, Winthrop University, and the University of Louisiana.
9 10 Although there has been no systematic data collected on SMIF performance, the limited anecdotal evidence suggests students generally do as well and sometimes better than investment professionals or the market as a whole. For example, the Tennessee Valley Authority reported that over a three year period, the 19 universities participating in its program in 2002 outperformed the S&P 500 benchmark by 5.3% (Mansfield, 2002).

72
Table III. Profile of the University of Missouri - St. Louis Fund
Panel A. Fund Characteristics Date Established Size in June 2007 Annual Student Participation Fund structure Funding source Faculty Member Credit hours per semester Max credit hours Student level Application Decision process Investment style Investment types allowed Equity strategy Diversification required? Income Distributions 1988 $125,000 45 Part of endowment Small private donations Full-time regular 1 credit hour per semester 3 hours (may continue without credit) Undergraduate None Majority vote of students Growth and value Equities, fixed income and options Bottom-up approach Yes Scholarships

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Company sponsoring 20 schools and Stern Agee Group, Inc. supporting 5 universities. The basic model at these programs is for the company to provide all funding ($400,000 each for the TVA) with the company and universities sharing the profits. In case of a falling stock market, the company absorbs all losses and fully replenishes the money the following year. About 58% of universities have an advisory board associated with their programs. All of these boards have outside investment professionals and alumni serving as a valuable resource in a counseling capacity. This allows students to interact with professionals and showcases the program to the local community. In many cases, students make formal presentations to the boards to sharpen their presentation and analytical skills.

VI. Student Participation
Just over 5,000 students participate in SMIFs in the US each year, with another 500 students being trained at foreign universities. Approximately 71% of the programs in the US (45% of foreign programs) are structured as part of a formal class.12 The number of credit hours a student can earn ranges from 1 to 12. Of those providing credit, 44% allow students to earn 6 or more semester credit hours over 2 or more semesters. Another 39% of schools limit students to a maximum of 3 credit hours. Only 22% of programs limit the student learning experience to a single semester. The SMIFs that are not part of a formal class allow students to participate as an extra curricular activity. This less structured format permits greater inclusion since almost any student enrolled in the university, regardless of major field or prior course work, can join the group. In contrast, formal classes often restrict the quality of students usually through an application process (59% of schools have a formal application process to screen students). Unlike many other university programs, most SMIFs carefully control the level of student participation. Although a few schools allowed more than 100 students to manage the portfolios each year, the average fund in the US had only 29 student managers per year (23 students for foreign funds). For approximately 90% of SMIFs, students were responsible for making all investment decisions. In the other 10% of programs, advisory boards or a faculty member also shared in the decision making. At the London Business School, students performed all of the usual research on securities but they had to make formal presentations to a professional board, which actually made the final investment selections. While

Panel B. Actual 5 Year Historical Annual Returns (Including Dividends) Year 2003 2004 2005 2006 2007 Actual fund performance 36.78% 17.67% 5.76% 6.93% 12.25% S&P 500 28.68% 10.88% 4.91% 15.80% 5.49%

up as a separate entity, like a nonprofit foundation or trust to provide more autonomy from the university. It is also becoming more popular for programs to establish profit making companies (e.g. LLCs or partnerships) where students are managing the portfolio for private companies or other investors. At least ten of the largest funds, including the University of Wisconsin, University of Minnesota, Pennsylvania State University, University of Houston, and University of Texas are all managing some private investor money.11 It should be noted that this is a more complex structure in the US, which requires government reporting (e.g. partnership tax returns with K-1 forms) due to the taxable nature of the investments. Several innovative companies in the US have long provided money to support financial education at institutions in the markets they serve. The largest is the Tennessee Valley Authority (TVA), a large electric utility company, which sponsors 25 universities in its service area. Two brokerage firms followed the TVA’s lead with D. A. Davidson &
11 At the time of this survey, Penn State had 68 private investors and was planning to expand the number to 99. The faculty advisor also reports the additional burden from accounting and tax costs for the LLC run between $25,000 and $50,000 per year.

In France, universities have legal barriers for incorporating SMIFs into the curriculum. Thus, only a few informal clubs not sponsored by the school exist, which are advised by outside professionals.
12

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

73

the major goal of these programs is to strengthen student Despite a majority of participating faculty believing SMIF decision making through actual investment experience, 64% courses take significantly more time, 63% of the schools paid of the funds have guidelines that allow a faculty member or the same level of compensation as for a regular course. The an advisory board to veto student recommendations if an balance of the universities provided additional compensation investment is deemed inappropriate for the portfolio. in the form of research funding, supplemental pay, or a However, this is a power that is rarely used. In the previous reduced service load. This may partially explain why some five years for those programs with a veto power, the higher programs have become inactive because key faculty members authority vetoed less than 4% of all student decisions. This feel the compensation levels are not commensurate with the result indicates that students time involved. This would take their fiduciary roles as suggest college deans need to The size of SMIFs today has portfolio managers seriously take a closer look at the cost/ and act prudently. benefit ratio of SMIFs and expanded beyond what many To facilitate investment make a conscious effort to decisions, 65% of the programs adequately reward faculty people would have thought assign students to groups to involvement in these high possible only a decade ago. There manage the portfolios. profile programs. It would be Depending on the specific a shame if much of the progress are 78 universities worldwide with portfolio and its goals, these in financial education of the more than $1 million under groups are often based on types past three decades would be of securities being traded, allowed to erode based solely management by students. industries, etc. For 70% of the on short-term economic SMIFs, investment decisions savings. One possible solution are determined by a simple majority vote by the students. would be to allow the fund itself to provide additional Individual portfolio managers make the decisions 6% of the compensation to the participating faculty members. This time with the remaining funds using a combination of students, could be in the form of added salary, research grants, a course faculty advisors and/or boards to reach a consensus. Of the release, a graduate assistant, etc. classes, 42% of programs allow only undergraduate students, Given the nature of SMIFs, many of the programs have 10% permit only graduate students and 48% allow both levels local investment professionals closely involved. The most of students. direct role for outside professionals is to serve as an adjunct faculty member and run the actual program. A small but VII. Faculty and Professional Involvement growing number of schools take this approach, drawing on the general finance community to provide the course instructor. In most cases, the adjunct faculty member is retired With a small number of exceptions, faculty are closely and thus has time available to staff day sections of the class. involved with SMIFs at all levels.13 Because so many of the It works less effectively for active professionals who may programs are relatively new, many of the faculty involved have difficulty finding the free time during normal work hours today with SMIFs worked hard to obtain the original funding. when the financial markets are open. Even when the program Of the universities requiring students to participate through is being taught by a full-time faculty member, it is a formal class, 58% of these professors believe these classes commonplace to have professionals serve on advisory boards take substantially more faculty time than a regular course. and be guest speakers. There is no question that professional The average assessment was that the instructional load was involvement enriches the experience of students, faculty and 50% higher than a traditional class, or the equivalent to professionals. Frequent contact between the various parties teaching a 4.5 credit hour course rather than a 3 hour course. ensures that current practice is quickly incorporated into the Another 33% of faculty felt they spent about the same amount classroom and students leave better prepared to apply their of time as any other course. Only 9% of faculty thought it knowledge and skills. actually took less time than their normal instructional duties.

VIII. Investment Activity
A number of universities have real money funds that operate more as an extracurricular activity with no direct faculty involvement. In some cases, the students invest their own money. Schools that operate this way include the University of Edinburgh, Harvard University, Auburn University, the California Institute of Technology, Princeton University, Dartmouth University, Vanderbilt University and the Georgia Institute of Technology.
13

It is interesting that 28% of schools with a SMIF have more than one fund. Many of these funds have different investment objectives and are specifically designed to give students a

74
broader investment experience than a single fund could provide. For schools with more than one fund, the most common number was to have three distinct portfolios. It also makes sense to have more than one SMIF where there are multiple bodies of students including undergraduate/graduate, day/evening classes, etc. With the growing size of the average portfolio, it is not surprising to find 92% of universities have formal written investment guidelines. Diversification is a principle widely emphasized by most SMIFs. Counting a mutual fund as a single holding, the average portfolio contained about 30 individual securities with three SMIFs exceeding 75 different issues. Approximately 80% of the programs have clear guidelines that specifically require the funds be diversified. It is interesting that 19% of US SMIFs are not prohibited from becoming hedge funds to increase returns by taking on more risk. Only 10% of non-US SMIFs had this same freedom. In the first reported case of an actual hedge fund on a university campus, Cornell University changed its SMIF’s investment strategy from a indexed styled fund to a “marketneutral” hedge fund in 2002.14 The stated goal was to produce positive returns regardless of which direction the market was moving. The $3 million fund uses investors’ money (alumni and friends) which allow students to manage the money for the experience without any fees. As to types of security investments, some schools, such as the University of Toledo, are very restrictive and require that all investments must be in domestic markets. On the other extreme, Roger Williams University limits domestic investments to a maximum of 20% of the portfolio with the other 80% being comprised of international securities. Most SMIFs focused on traditional securities with common stock dominating portfolios, regardless of where the companies were domiciled. Corporate and Treasury bonds were also widely used to balance the portfolios. For alternative investments, real estate investment trusts (REITs) were the most popular followed at a distance by limited partnerships. For actual trading activities, full service brokerage firms were most often used by SMIFs, followed closely by discount brokerage firms. Bank trust companies were used only about half as much as either type of brokerage firm. The funds are large enough to negotiate some very favorable rates with full service brokerage firms. In addition, since the majority of the programs are charitable and tax-free by design, some of the brokerage firms are donating their services to the universities. For SMIFs operating a single fund, only 10% of the programs characterized their investment style as focusing on growth stocks. Another 23% considered themselves to be

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

more value investors. But the vast majority of the funds characterized their investment style to be more of a blend. The most employed equity strategy was the bottom-up approach with 37% of SMIFs primarily using this method. A close second was the top-down approach used by 27% of funds, followed at a distance by the buy-and-hold strategy reported by 11% of respondents. The balance of the SMIFs used a combination of these strategies and others (e.g. price momentum and contrarian) in making stock selections. Twice as many programs thought asset allocation was a very important consideration compared to those who deemed it not very important (38% v. 18%). Individual security selection was rated very important by 58% of SMIFs. In contrast, 65% of funds believed market timing was not important, most likely reflecting their long-term investment horizons.

IX. Comparisons of US and Foreign Funds
With the widely divergent political and economic climates abroad, it was not unexpected to find that SMIFs outside of North America evolved slower and in a somewhat different direction than those in the US. For the most part, Canadian and American universities share a similar educational environment. Thus the Canadians adopted SMIF organizational structures and operating procedures modeled after those already successfully employed in the US for several decades. Outside of North America, however, European and Asian schools were far more likely (by a ratio of 2 to 1) to have extra-curricular programs rather than formal classes. One result of a less formal structure is that fewer foreign universities (55%) have anyone with veto power over student investment decisions compared to US schools (64%). Of course, the funds outside of North America are much younger and smaller, averaging only $142,000. With maturity and more money at risk, stricter university controls may eventually develop. Unlike the US, none of the European or Asian programs had a taxable structure similar to a LLC or partnership. When it comes to investing, most American SMIFs focus on investing in common stock with an average portfolio in 2007 containing 82% of their money invested in these securities. Foreign programs generally leaned more toward balanced portfolios with greater allocations of fixed income securities. Canadian SMIFs averaged 70% of their money invested in common stock, while European and Asian funds were much lower at 59%. On the lower end of the scale, Hebrew University had the most balance with 25% in common stock, 20% in preferred stock, 15% in corporate bonds, 25% in Treasury bonds, and the remainder in other securities. In terms of investment policies, almost the same percentage of foreign SMIFs required diversification in their written policies as in the US (79% v. 81%). As to investment styles, only

14

Myers (2004).

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

75
hundred thousand dollars in cash flow while providing students with a valuable learning experience. Very few university programs have such a high benefit/cost ratio, especially since almost all universities have endowment funds that must be managed by someone. Historically, the limited evidence shows SMIFs have performed as well and sometimes better than funds managed by professional investment advisors (Mansfield, 2002). Of course, since SMIFs do not normally charge management fees, this saving alone favors student-managed funds even without the educational benefits.

10% of American funds with a single fund considered themselves “growth” investors in 2007 compared to 20% of the foreign funds. Regardless of where the fund was domiciled, the predominant investment style was a “blend” rather than a single focus. The equity strategies most used in the US and abroad was a bottom-up approach, closely followed by the top-down approach, or a combination of the two.

X. Benefits to the University Community
It has long been recognized that students learn more by hands-on experience than simply reading about a topic in a textbook. Besides learning the intricacies of portfolio management and trading, students also benefit in many programs by going on field trips to Wall Street and other financial markets. Schools like Virginia Tech and Gannon University have a long tradition of taking students annually to Wall Street to view the financial markets first hand. Roger Williams University has taken it to the next level by taking students abroad to the London and Frankfurt stock exchanges, which would not have been possible without a SMIF to generate the student interest and fund the activity. Eighty-one percent of all program directors cited better trained students as a major benefit of having a SMIF. Almost a third of faculty believed having a real money fund provided synergy and significantly improved the quality of the overall finance program. Conversations with faculty also indicated greater job opportunities for students participating in SMIFs. Many employers, including private equity and hedge funds, bank trust companies, and mutual funds have been aggressively recruiting students who have these experiences to draw on. These are highly competitive jobs that can be difficult for a new college graduate to obtain without sufficient experience. As anyone involved in a SMIF can attest, having a real money portfolio generates a substantial amount of media attention. This activity not only showcases the students and the finance discipline but also the business school and the university. Alumni in particular are highly supportive of SMIFs, which creates new opportunities for guest speakers, field trips, internships, student recruitment, etc. Finally, the programs provide badly needed financial support for student scholarships, visits to financial markets, operating trading rooms, and other university programs. For 2006, sixty-six of the American funds made cash distributions totaling more than $1.9 million to support academic programs, or an average of $29,381 per school. However, an even larger number of SMIFs reported making no cash distributions in the previous year. Many of these were still relatively new and therefore were still in the capital building years. Nevertheless, it was not all that unusual for the larger SMIFs to spin-off several

XI. Recent Developments
A. Trading Rooms
A growing contingent of programs are operating trading rooms to add even more realism to student learning. Many of the universities with SMIFs have invested up to $1 million to fully furnish and equip trading rooms. The expanded programs include Pennsylvania State University, Iowa State University, Rice University, Michigan State University, Stetson University, Texas Christian University, University of Michigan, and the University of Missouri - Columbia. These schools believe this development has raised the bar in attracting top students and community financial support. Of course, there are other universities with trading rooms that do not have SMIFs and simply simulate trading activities.

B. Social Responsibility Funds
Social responsibility funds are becoming more popular in academia and with investors in general. Bluffton University in Ohio has had an investment policy since 1956 of avoiding “sin stocks”, which include tobacco, alcohol, and defense companies. Villanova University follows a similar investment guideline with two of its funds. The University of California at Berkeley started a new social responsibility fund in February of 2008 with $1.2 million as part of its MBA program. Students will hold long positions in firms that are socially responsible and take short positions in firms with poor social records. The director of its program maintains one does not have to sacrifice financial returns for a good record of social responsibility (Alsop, 2007). Establishing socially responsible funds can be highly controversial in academia. Some opponents argue that it is pushing a political agenda. In 1997, Stanford University rejected a student proposal for such a fund, noting their endowment already had substantial stock investments in socially responsible companies and industries and thus it was not needed.

76
C. Using Investors’ Money

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

faculty, or the general community at large. Doing so may speed technology transfer from universities and fulfill one Several of the largest funds (including the universities of mission of higher education. All of these innovative programs Texas, Minnesota, Houston, Wisconsin, and Pennsylvania expand the practical training offered to finance students by State) manage investor money in one or more of their funds. conventional SMIFs in new dimensions. They encourage The University of Texas was the first large for-profit fund students to take a more entrepreneurial approach to raising when it raised $1.6 million of private investor money in capital in the private equity market (often with partner or December 1994 for their MBA students to manage. By 2007, investor money). Students benefit by evaluating business they had three distinct funds with different investment plans and performing due diligence before actually making the investment decision on objectives totaling $17 companies with little or no million from 60 investors. One of the most exciting financial performance record. These funds are usually developments over the last decade This focus provides a nice structured as a Limited complement to a regular SMIF Liability Company (LLC) has been the emergence of venture where the focus is on where the income is taxable established, publicly-traded capital funds managed by students. to investors similar to a investment opportunities. 16 partnership distribution. This Given the success of SMIFs, it is only Finally, the venture capital structure limits the number funds offer an excellent vehicle of investors, so they must natural that the programs would for the College of Business to make large contributions. In evolve in new directions. provide value added support to exchange for managing the other units within the money, the students and university sometimes get a management fee of between .5% university community. For example, the Colleges of Medicine, Science and Engineering produce a continuing stream of and 1.5% of the assets. innovative research and technology but have great difficulty proving the commercial viability of their inventions and D. Venture Capital Funds patents. Along with business schools, law schools can assist One of the most exciting developments over the last decade new startups with legal issues to further reach another segment has been the emergence of venture capital funds managed by of the community. A student managed venture capital fund students. Given the success of SMIFs, it is only natural that offers the best opportunity in years to capitalize and profit the programs would evolve in new directions. The University from the research strengths of universities while enhancing of Michigan created the first student-managed venture capital the teaching mission. fund in the US in 1997 with about $3 million in capital. Yale University, the University of North Dakota, the University E. Micro Finance Funds of Utah, Cornell University, the University of Wyoming, and There is an amazing amount of creativity surrounding Miami University of Ohio followed in Michigan’s footsteps with venture capital funds of their own dedicated to investing SMIFs in the way the programs are being re-engineered to in emerging companies.15 The University of Utah sponsored accomplish more than simply teaching students the basics of the largest venture capital fund with $18 million, which also investing. Several universities, including Columbia permits students from more than 15 other universities to University, are starting micro finance funds to make small participate in the activities. Although not a student-managed entrepreneurial investments in third world countries. An fund, the University of Maryland in 2003 worked with organization called PlaNet Finance (a microfinance investors to establish the New Markets Growth Fund with organization based in Paris) is working with Columbia and $20 million in capital run by professional managers but several European universities to sponsor these programs. If assisted by students and faculty. Others, including the Columbia’s program is a success, it will provide another venue University of Queensland and the University of Melbourne (both in Australia), have similar funds run by professionals. Many of these programs are designed to provide seed 16 The 2008 state budget for the Commonwealth of Massachusetts contained capital for businesses started by students, recent graduates, an amendment for establishing a student investment fund to encourage
For further information, see Rombel (2007), Yale Bulletin (2001), Daily Herald (2006) and the Business Wire (2006).
15

student entrepreneurship. There were to be three students on the governing board and it was specifically designed to fund new student businesses created within the Commonwealth. (See Section XX, Chapter 23A of the budget)

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

77
business schools are becoming even more relevant by addressing important issues in both the financial markets and society. The benefits of providing students with greater practical experience and technical skills in finance are widely recognized in the job market. Students graduating today from universities with SMIFs already have at least 1 or 2 semesters of actual trading and research experience. Although the university experience is not as intense as in a professional job, it still provides a solid foundation for the knowledge needed in portfolio management. The skills and techniques learned here can be further refined in the workplace over a much shorter period of time than what would have been possible in the 1960s or 1970s.

for students to learn about business while benefitting social welfare initiatives around the world.

XII. Conclusion
Over the past 50 years, student-managed investment funds have revolutionized the way in which investment education is taught in universities. These programs have expanded to 314 worldwide today from only a few funds in the 1950s. In the process, SMIFs are evolving in exciting new directions. These include managing money for private clients, establishing hedge funds, or venture capital funds and micro lending initiatives. While it is much too early to evaluate the success of these new programs, it does seem clear that Appendix A. Listing of All US Funds

University Name
Abilene Christian University * Adelphi University Alabama A&M University Alaska Pacific University * Alfred University * American University Anderson University Appalachian State University Arizona State University * Ashland University Auburn University Austin College * Austin Peay State University * Babson College Baldwin-Wallace College * Ball State University Bates College * Baylor University Belmont University * Bentley College * Binghamton University - SUNY * Bluffton University Boise State University Boston College * Boston University Bowling Green State University Brandeis University * Brigham Young University Bryant University Bryn Mawr College * Bucknell University Butler University California Institute of Technology California Polytechnic State Univ. California State University Fresno California State University - Long Beach

City
Abilene Garden City Normal Anchorage Alfred Washington Anderson Boone Tempe Ashland Auburn Sherman Clarksville Babson Park Berea Muncie Lewiston Waco Nashville Waltham Binghamton Bluffton Boise Boston Boston Bowling Green Waltham Provo Smithfield Bryn Mawr Lewisburg Indianapolis Pasadena San Luis Obispo Fresno Long Beach

State
TX NY AL AK NY DC IN NC AZ OH AL TX TN MA OH IN ME TX TN MA NY OH ID MA MA OH MA UT RI PA PA IN CA CA CA CA

Year Started
n/a 2007 1998 2000 1995 2002 2007 2000 1996 2000 1999 2007 1998 1997 2006 2005 2004 2001 2003 1997 2003 1956 1995 1983 2001 2006 1998 1984 2005 1975 2000 2007 1978 1992 1999 1995

Funds 2007 $000
319 100 330 185 200 100 10 116 515 375 50 1,000 400 1,300 175 577 100 6,500 400 555 130 174 149 360 25 265 13 1,866 425 100 750 1,000 490 453 90 100

78
Appendix A. Listing of All US Funds (Continued)
University Name
California State University - Northridge California State University - Los Angeles* Cameron University Canisius College * Carnegie Mellon University * Carroll College * Cedar Crest College Cedarville University* Centenary College of Louisiana * Central Michigan University Christian Brothers College Christian Brothers University Claremont Graduate School * Clemson University * Cleveland State University * College of New Jersey College of William & Mary College of Wooster Colorado College Colorado State University Connecticut College * Cornell University * Creighton University Culver Stockton College * Dartmouth DePaul University * Drake University * Drexel University Duke University * East Tennessee State University Eastern Illinois University Eastern Washington University * Elizabethtown College Emory University Fairfield University Florida Gulf Coast University Franklin and Marshall College Gannon University Gardner Webb University * George Washington University Georgetown University * Georgia Institute of Technology Georgia State University Gonzaga University Grinnell College * Gustavus Adolphus College * Harvard University * Henderson State University Humboldt State University Idaho State University Illinois College Illinois State University Illinois Wesleyan University Indiana State University * Indiana University * Indiana University of Pennsylvania Iowa State University

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

City
Northridge Los Angeles Lawton Buffalo Pittsburgh Helena Allentown Cedarville Shreveport Mt. Pleasant Memphis Memphis Claremont Clemson Cleveland Ewing Williamsburg Wooster Colorado Springs Fort Collins New London Ithaca Omaha Canton Hanover Chicago Des Moines Philadelphia Durham Johnson City Charleston Cheney Elizabethtown Atlanta Fairfield Fort Myers Lancaster Erie Boiling Springs Washington Washington Atlanta Atlanta Spokane Grinnell ST. Peter Cambridge Arkadelphia Arcata Pocatello Jacksonville Normal Bloomington Terre Haute Bloomington Indiana Ames

State
CA CA OK NY PA MT PA OH LA MI TN TN CA SC OH NJ VA OH CO CO CT NY NE MO NH IL IA PA NC TN IL WA PA GA CT FL PA PA NC DC DC GA GA WA IA MN MA AR CA ID IL IL IL IN IN PA IA

Year Started
1994 2001 1988 2003 2006 2004 1997 2008 2003 1997 2003 2003 2001 2004 2007 2000 1999 1955 2004 1998 2002 1998 1993 1996 1996 1982 1999 2007 1952 2000 1994 2004 2007 2006 2006 2005 1999 1952 2000 2005 1999 1986 2005 2000 2000 1998 na 2001 2006 2005 1995 1982 1993 2000 1986 2005 1999

Funds 2007 $000
2,000 100 800 300 64 50 52 75 120 469 400 400 381 300 100 170 590 1,300 24 190 77 13,500 2,500 55 400 341 239 250 162 370 136 50 130 1,200 300 220 204 126 25 1,500 200 810 368 200 122 123 na 343 7 59 458 383 740 437 500 223 195

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

79

Appendix A. Listing of All US Funds (Continued)
University Name
Ithaca College * Jacksonville University James Madison University * John Carroll University Kansas State University Kennesaw State University Kutztown University of Pennsylvania Lafayette College Lehigh University Lipscomb University Longwood University Loras College Louisiana State University * Loyola College * Marquette University Marywood University Massachusetts Institute of Technology * McNeese State University Miami University Michigan State University Michigan Technological University Middle Tennessee State University Middlebury College Millsaps College Mississippi State University * Mississippi University for Women Montana State University - Bozeman* Montana State University - Billings * Moravian College Murray State University * Nebraska Wesleyan University New Mexico State University New York University North Carolina State University North Dakota State University Northeastern University Northern Arizona University Northern Illinois University Northern Michigan University Northwest Nazarene University Northwestern University Oberlin College* Ohio Northern University Ohio State University Ohio University Oregon State University Ouachita Baptist University Pace University Pacific Lutheran University * Pennsylvania State University Portland State University Princeton University * Purdue University Radford University Rice University Roanoke College * Roger Williams University

City
Ithaca Jacksonville Harrisonburg Cleveland Manhattan Kennesaw Kutztown Easton Bethlehem Nashville Farmville Dubuque Baton Rouge Baltimore Milwaukee Scranton Cambridge Lake Charles Oxford East Lansing Houghton Murfreesboro Middlebury Jackson Mississippi State Columbus Bozeman Billings Bethlehem Murray Lincoln Las Cruces New York City Raleigh Fargo Boston Flagstaff DeKalb Marquette Nampa Evanston Oberlin Ada Columbus Athens Corvallis Arkadelphia Pleasantville Tacoma University Park Portland Princeton West Lafayette Radford Houston Salem Bristol

State
NY FL VA OH KS GA PA PA PA TN VA IA LA MD WI PA MA LA OH MI MI TN VT MS MS MS MT MT PA KY NE NM NY NC ND MA AZ IL MI ID IL OH OH OH OH OR AR NY WA PA OR NJ IN VA TX VA RI

Year Started
2005 1987 1999 1996 2002 2006 2005 1950 1962 2003 2002 1998 2005 2006 2005 2006 1964 2007 1996 2003 1998 1998 1987 1989 1998 1999 1985 1985 1962 1998 2005 2007 2000 2004 2007 2007 2000 2000 2006 2003 1964 2004 1989 1990 2001 2005 2000 2002 1982 2005 1997 2006 2000 2003 1996 2004 2004

Funds 2007 $000
24 454 146 170 1,100 100 190 455 360 450 430 172 1,000 500 1,200 75 27 21 375 4,200 1,300 325 275 200 400 385 50 50 1,442 440 250 5,013 2,001 135 110 50 997 230 210 70 2,375 281 128 25,810 2,000 60 20 280 92 5,000 251 10 400 495 900 500 122

80
Appendix A. Listing of All US Funds (Continued)
University Name
Rollins College Rutgers University * Saint Bonaventure University Saint Cloud State University Saint John's University Saint Joseph's University Saint Louis University Saint Mary's University Salisbury University Samford University San Diego State University * Santa Clara University Scripps College * Seattle University * Shippensburg University Southeast Missouri State University Southern Illinois University Southern Methodist University Southern New Hampshire University Southwestern University Spring Arbor University Stanford University * State University of New York - Geneseo Stetson University Syracuse University * Tennessee State University * Tennessee Tech University Texas A & M University * Texas Christian University Texas Tech University Texas Wesleyan University Trevecca Nazarene University Trinity University Tufts University * Tulane University * Union University * University of Akron University of Alabama - Huntsville University of Alabama - Birmingham * University of Alabama - Tuscaloosa* University of Alaska University of Arizona University of Arkansas-Fayetteville University of California - Los Angeles University of California - Berkeley* University of Chicago * University of Cincinnati * University of Colorado - Boulder University of Colorado - Colorado Springs University of Connecticut * University of Dayton * University of Delaware * University of Denver * University of Georgia University of Houston University of Idaho * University of Illinois University of Iowa

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

City
Winter Park New Brunswick St. Bonaventure St. Cloud New York Philadelphia St. Louis San Antonio Salisbury Birmingham San Diego Santa Clara Claremont Seattle Shippensburg Cape Girardeau Carbondale Dallas Manchester Georgetown Spring Arbor Stanford Geneseo DeLand Syracuse Nashville Cookeville College Station Ft. Worth Lubbock Ft. Worth Nashville San Antonio Medford New Orleans Jackson Akron Huntsville Birmingham Tuscaloosa Fairbanks Tucson Fayetteville Los Angeles Berkeley Chicago Cincinnati Boulder Colorado Springs Storrs Dayton Newark Denver Athens Houston Moscow Champaign Iowa City

State
FL NJ NY MN NY PA MO TX MD AL CA CA CA WA PA MO IL TX NH TX MI CA NY FL NY TN TN TX TX TX TX TN TX MA LA TN OH AL AL AL AK AZ AR CA CA IL OH CO CO CT OH DE CO GA TX ID IL IA

Year Started
1999 2000 2003 1999 2001 2004 2002 2007 2000 2008 1992 2000 na 2004 1994 1990 2000 1980 2004 1999 2005 1978 2007 1980 2006 1998 2000 2000 1973 1997 1998 2003 1998 na 1999 2003 1996 1998 2007 1998 1995 2000 1971 1987 1999 2005 2000 2002 2004 2000 1994 1996 1999 2007 2002 1989 1999 1994

Funds 2007 $000
750 1,605 45 115 2,700 117 916 1,000 388 500 100 350 200 50 81 835 360 6,500 59 349 12 180 18 3,100 1,100 400 500 250 1,500 2,200 776 405 1,340 1,059 2,419 400 100 428 385 50 550 930 12,000 2,000 120 1,000 350 300 58 2,300 6,300 800 550 101 9,177 400 390 536

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

81

Appendix A. Listing of All US Funds (Continued)
University Name
University of Kansas University of Kentucky University of Louisville University of Maine University of Maryland University of Memphis University of Michigan University of Minnesota - Minneapolis University of Minnesota - Duluth University of Mississippi University of Missouri-Columbia University of Missouri-St. Louis University of Montana * University of Nebraska - Lincoln University of Nebraska - Omaha* University of Nevada * University of New Hampshire * University of New Mexico University of North Alabama * University of North Carolina - Chapel Hill University of North Carolina - Wilmington * University of North Carolina - Charlotte * University of North Dakota University of North Florida * University of North Texas * University of Northern Colorado * University of Northern Illinois * University of Northern Iowa * University of Notre Dame University of Oklahoma University of Oregon University of Pennsylvania * University of Pittsburgh University of Portland University of Rhode Island University of Richmond University of Rochester University of South Dakota University of Southern California * University of Southern Mississippi University of St. Thomas University of Tampa University of Tennessee - Martin University of Tennessee- Knoxville University of Tennessee - Chattanooga University of Texas University of the Pacific University of Toledo * University of Tulsa University of Utah University of Virginia - McIntire School University of Virginia - Darden Graduate * University of Washington * University of Wisconsin-Eau Claire University of Wisconsin-Madison University of Wisconsin-Whitewater University of Wisconsin- Oshkosh University of Wisconsin - Platteville* University of Wyoming Utah State University *

City
Lawrence Lexington Louisville Orono College Park Memphis Ann Arbor Minneapolis Duluth University Columbia St. Louis Missoula Lincoln Omaha Reno Durham Albuquerque Florence Chapel Hill Wilmington Charlotte Grand Forks Jacksonville Denton Greeley DeKalb Cedar Falls Notre Dame Norman Eugene Philadelphia Pittsburgh Portland Kingston Richmond Rochester Vermillion Los Angeles Hattiesburg St. Paul Tampa Martin Knoxville Chattanooga Austin Stockton Toledo Tulsa Salt Lake City Charlottesville Charlottesville Seattle Eau Claire Madison Whitewater Oshkosh Platteville Laramie Logan

State
KS KY KY ME MD TN MI MN MN MS MO MO MT NE NE NV NH NM AL NC NC NC ND FL TX CO IL IA IN OK OR PA PA OR RI VA NY SD CA MS MN FL TN TN TN TX CA OH OK UT VA VA WA WI WI WI WI WI WY UT

Year Started
1994 1999 2004 1993 1993 1999 2000 1998 2003 2001 1967 1988 1985 1981 2000 2004 1995 2006 2003 1952 2007 1997 2005 1999 2003 1992 1999 1999 1998 1996 1999 1996 1999 2003 2001 1993 1995 2001 1986 2002 1999 2003 1998 1998 1998 1994 2007 2005 1998 1998 1994 1990 na 2003 1970 1999 2000 2001 2005 1985

Funds 2007 $000
1,523 400 50 1,253 1,350 475 3,700 25,000 440 335 1,354 125 50 1,300 1,400 107 50 2,400 400 1,424 1,000 235 676 772 277 1,100 200 115 5,000 505 900 700 351 65 151 325 200 520 2,600 308 3,000 155 460 1,000 510 17,000 1,100 1,000 1,577 18,173 500 6,200 50 250 62,000 85 135 190 1,700 50

* Did not respond to survey. Information collected from media and institution’s web site.

82
Appendix B. Listing of All Non-US Funds
University Name Birla Institute of Tech. & Sciences Bishop's University Bond University Brock University Concordia University Hebrew University of Jerusalem HEC Montreal London Business School Maastricht University Massey University McGill University * Punjab College of Technical Ed Queens University * Simon Fraser University St. Francis Xavier University St. Mary's University University of Alberta University of British Columbia * University of Calgary University of Edinburgh University of Guam University of Manitoba University of New Brunswick University of Toronto * Wilfrid Laurier University City Pilani Sherbrooke Gold Coast St. Catharines Montreal Jerusalem Montreal London Maastricht Auckland Montreal Ludhiana Kingston Vancouver Antigonish Halifax Edmonton Vancouver Calgary Edinburgh Mangilao Winnipeg Fredericton Toronto Waterloo Province Rajasthan Quebec Queensland Ontario Quebec Quebec Limburg

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Country India Canada Australia Canada Canada Israel Canada United Kingdom The Netherlands New Zealand Canada na Canada Canada Canada Canada Canada Canada Canada United Kingdom US Territory Canada Canada Canada Canada

Year Started Funds 2007 $000 2007 1996 na 1995 1999 1999 1999 2003 1994 1995 na 3 2001 2003 2000 2005 1998 1987 1996 1997 2006 1997 1998 2007 2001 3 485 28 17 1,378 580 3,810 300 70 15 10 3,000 9,983 2 184 1,292 3,514 361 na 53 11 2,200 17 340

Quebec India Ontario British Columbia Nova Scotia Nova Scotia Alberta British Columbia Alberta Guam Manitoba New Brunswick Ontario Ontario

* Did not respond to survey. Information collected from media and institution’s web site.

References
1997, “Students Press for Socially Responsible Endowment Fund,” Stanford University News Service (February 5). 2001, “Yale SOM Launches Student-Managed Venture Capital Fund,” Yale Bulletin 29, 1 (February 16). 2006, “Local News,” Daily Herald (April 12). 2006, “Record $18 million Closing for Largest Student-Run Venture Capital Fund,” Business Wire (June 19), 1. Alsop, R., 2007, “MBA Students to Run Socially Responsible Fund,” The Wall Street Journal Online (September 14). Alsop, R., 2007, “Talking b-School: Haas Takes New Tack on Investing,” The Wall Street Journal (September 18, 2007), B8. Ammermann, P. A. and L. R. Runyon, 2003, “Risk Aversion and Group Dynamics in the Management of Student Managed Investment Fund”, Journal of the Academy of Business and Economics 1 (No. 1). Ary, E. J. and R. L. Webster, 2000, “A Survey of University Student Investment Funds,” Midwest Review of Finance and Insurance 14 (No. 1), 9-18. Bear, T. and G. M. Boyd, 1984, “An Applied Course in Investment Analysis and Portfolio Management,” Journal of Financial Education 13 (Fall), 68-71. Belt, B., 1975, “A Securities Portfolio Managed by Graduate Students,” Journal of Financial Education 4 (Fall), 7781.

LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS

83
Mansfield, D., 2002, “TVA’s Student Investors Outperform Market, TVA Managers,” The Florida Times Union (April 5). Markese, J. D., 1984, “Applied Security Analysis and Portfolio Management,” Journal of Financial Education 13 (Fall), 65-67. McInerny, P. M., 2003, “The Student-Managed Investment Fund at the High School Level,” The Clearing House 76 (No. 5), 252-254. Myers, L., 2004, “CU Student-Managed Investment Fund Thrives After Change in Strategy,” Cornell Chronicle 35 (No. 24), 5 Neely, W. P., 2002, “A Survey of Student-Managed Funds: The Way They Work,” presented at the Financial Management Association meeting, (October 17). Neely, W. P. and P. L. Cooley, 2004, “A Survey of Student Managed Funds,” Advances in Financial Education 2 (Spring), 1-9. Pfeffer, J., 2007, “What’s Right and Still Wrong with Business Schools,” Biz Ed 6 (No. 1), 42-48. Rombel, A., 2007, “Student-Run Venture Capital Firm at Cornell Invests in Local Tech Firm,” Business Journal Central New York 21 (No. 12), 8. Tatar, D. D., 1987, “Teaching Securities Analysis with Real Funds,” Journal of Financial Education 16 (Fall), 40-45.

Bhattacharya, T. K. and J. J. McClung, 1994, “Cameron University’s Unique Student-Managed Investment Portfolios,” Financial Practice and Education 4 (No. 1), 55-59. Block, S. B. and D. W. French, 1991, “The Student-Managed Investment Fund: A Special Opportunity in Learning,” Financial Practice and Education 1 (No. 1), 55-60. Dolan, R. C. and J. L. Stevens, 2006, “Business Conditions and Economic Analysis: An Experiential Learning Program for Economic Students,” Journal of Economic Education 37 (No. 4), 395-405. Hirt, G. A., 1977, “Real Dollar Portfolios Managed by Students - An Evaluation,” Journal of Financial Education 6 (Fall), 57-61. Johnson, D. W., J. F. Alexander and G. H. Allen, 1996, “Student-Managed Investment Funds: A Comparison of Alternative Decision-Making Environments,” Financial Practice and Education 6 (No. 1), 97-101. Kahl, D. R., 1997, “The Challenges and Opportunities of Student-Managed Investment Funds at Metropolitan Universities,” Financial Services Review 6 (No. 3), 197200. Kester, G. W., 1986, “Extending Investments Beyond the Classroom Through Investment Clubs,” Financial Management Collection 1 (No. 3), 9-10. Lawrence, E. C., 1990, “Learning Portfolio Management by Experience: University Student Investment Funds,” Financial Review 25 (No. 1), 165-173. Lawrence, E. C., 1994, “Financial Innovation: The Case of Student Investment Funds at United States Universities,” Financial Practice and Education 4 (No. 1), 47-53.

84

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Behavioral Basis of the Financial Crisis
Joseph V. Rizzi

Financial institutions suffered large losses following the collapse of the credit markets despite making huge risk management investments. Major risks are frequently ignored due to behavioral biases resulting in incorrect decisions. These biases are reinforced by organizational obstacles, such as misaligned compensation systems. This article outlines a supplemental behavioral risk framework, and applies it to the structure finance market. Behavioral finance can improve how risk decisions are made. You ignore behavioral risk at your peril.

Major strides were made in quantitative risk management during the 1990s. Yet despite these advances, financial institutions suffered large losses following the collapse of the subprime and structured products markets. How this could have occurred given sophisticated tools and massive risk system investments is a concern. A further concern is the likelihood of repeating this experience during the next cycle. Although we know how risk decisions should be made, less is known on how these decisions are actually made. Risk management should encourage profitable risk taking while discouraging unprofitable and catastrophic risk. In most institutions, however, political power and capital flows to successful individuals. Unfortunately, it is difficult to determine whether they are truly successful or just lucky. Our existing risk measures account for perhaps 95% of what

occurs. The major catastrophic risks lurk in the fat tails of the remaining 5%. We tend to underestimate these improbable risks due to behavioral biases. Institutions and regulators are changing their risk systems and personnel to address this issue. The problem, however, is not only with the systems or the quality of the personnel but within the individuals themselves. Most individuals have a model of how the world works. When challenged by events, we try to explain away the events. Behavioral economics provides insight into risk-assessment errors and possible remedies. This article outlines a behavioral risk framework to address judgment bias and develop appropriate responses. Behavioral finance recognizes that decision processes influence perception and shape our behavior. The framework supplements current quantitative risk management by improving responses to risk changes over time. The framework will then be applied to the structured finance crisis.

I. Behavioral Finance Framework
Risk can be classified along two dimensions. The first concerns high-frequency events with relatively clear causeeffect relationships. Other risks occur infrequently. Consequently, the cause-effect relationship is unclear. The second dimension is impact severity. No matter how remote, high-impact events cannot be ignored because they can threaten an institution’s existence as was demonstrated in the current market crisis. The dimensions are reflected in the risk map in Figure 1. Quadrant A events include retail credit products including credit cards. Many small defaults are expected. Screening helps identify groups with higher default probabilities. These groups are charged higher rates to offset the risk. Quadrant B represents many internal operational risks such as check processing errors. The costs are absorbed and the focus is

J.V. Rizzi is a Senior Investment Strategist at CapGen Financial in New York, NY. The views expressed represent those of the author, and not CapGen Financial.

84

RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS

85
on how risk managers should act. These models build on expected utility theory (EUT), which views individuals as expected utility maximizers.2 Empirical support of EUT is mixed with numerous reported anomalies.3 Examples of anomalies include holding losers, selling winners, excess trading, and herding. An alternative, prospect theory,4 can explain these facts. Instead of being expected utility (E(U)) maximizers, investors are viewed as expected regret (E(r)) minimizers focusing more on losses than gains. This is reflected in Figure 2. EUT focuses on wealth changes. The value function in prospect theory is based on gains or losses relative to a reference point, usually par or the original purchase price. Behavioral finance examines how risk managers gather, interpret, and process information. Specifically, it concentrates on perception and cognitive bias. It recognizes models can influence behavior and shape decisions. These biases can corrupt the decision process, leading to suboptimal results as emotions override self-control. Market signals are complex. They include both information and noise. Information concerns facts affecting fundamental values. Noise is a random blip erroneously interpreted as a signal.5 Risk managers have developed shortcuts, rules of thumb, or heuristics to process market signals. These beliefbased heuristics incorporate biases or cognitive constraints, which will now be investigated.

Figure 1. Risk Map

Frequency

A B I

C D

Impact A: B: C: D: High frequency/low impact events: reflected in risk pricing. Low frequency/low impact events: treated as a cost of business. High frequency/high impact events: managed through control. Low frequency/high impact events: frequently ignored.

on mitigation and prevention through improved processing and training. Type C events include concentrated exposures to high risk borrowers. These well known risks are managed by constant management monitoring and control. Type D events are frequently ignored due to a low frequency. Examples include many of the structured finance products which represented short positions in an option. They offered long period of steady income punctuated with occasional large losses. Cyclical risks are low-frequency-high-impact events characterized by their negative skew and “fat-tailed” loss distributions. Investors incurring such risk can expect mainly small positive events but are subject to a few cases of extreme loss. These risks are difficult to understand. The difficulty stems from two factors. First, there is insufficient data to determine meaningful probability distributions. In this case, the statistics are descriptive, not predictive. Consequently, no amount of mathematics can tease out certainty from uncertainty. 1 Second, and perhaps more important, infrequency clouds hazard perception. Risk estimates become anchored on recent events. Overemphasis on recent events can also produce disaster myopia during a bull market, as instruments are priced without regard to the possibility of a crash. These facts lead to risk mispricing and the procyclical nature of risk appetite. Quantitative risk-management models are based on portfolio and option pricing theory and provide a framework
1 This is the Knightian distinction between risk, randomness with knowable probabilities and uncertainty, randomness with unknowable probabilities. See F. Knight 1921, Risk, Uncertainty and Profit, Houghton Mifflin, 1921.

A. Regret
Risk is forward looking. Regret, however, is backward looking. It focuses on responsibility for what we could have done but did not do. Regret underlies several biases. We try to minimize regret by seeking confirming data, suppressing disconfirming information, and taking comfort that others made the same decision. Consequently, regret can inhibit learning from past experiences. Sunk costs are the first regret bias considered. Sunk-cost bias involves avoiding recognizing a loss despite evidence the loss has already occurred and a further loss is likely.

M. Friedman and L. Savage, 1948, “The Utility Analysis of Choices Involving Risk,” Journal of Political Economy.
2

D. Ellsberg, 1961,“Risk, Ambiguity and the Savage Axioms,” Quarterly Journal of Economics 643.
3

A. Tversky and D. Kahmeman, 1992, “Advances in Prospect Theory: Cumulative Representation of Uncertainty,” Journal of Risk and Uncertainty, which builds on their earlier work. Prospect theory is a key component of Behavioral Economics. Behavioral finance is a subset of Behavioral Economics, applying its concepts to asset pricing. This article uses the terms interchangeably.
4 5

See E. Black 1986, “Noise,” Journal of Finance July.

86
Figure 2. Investors Minimize Expected Regret

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Value function value + . Convex slope indicates pain of loss (regret exceeds value of gain . The conflict between E(u) maximizing and E(r) minimizing underlies many anomalies . Investment decisions involve 3 Rs: return, risk and regret Utility

Losses

Gain

Reference point -

Examples include the reluctance to sell impaired assets at reduced prices. Usually this is defended as the market prices being too low. Most institutions, however, reject the logical alternative of acquiring additional exposure at the market price to exploit the alleged under pricing; thus, illustrating in this instance, price is of secondary importance relative to regret. Panic conditions are also based on a combination of regret and herding. In a crisis, the reference is pessimism, and we actively seek bad news to confirm our belief. Thus, to minimize regret, we follow the herd not to be left behind and engage in panic selling. This further depresses prices leading to continued forced selling and the creation of a negative feedback loop as occurred in the fourth quarter 2008. Another regret-related bias is the house money effect. Risk managers will assume greater risks when they are up in a bull market and lower risk in a bear market. Regret is perceived to be less when risk of winnings is involved, than risk of initial capital. This procyclical phenomenon leads to “buy high and sell low” behavior, reflected in Figure 3. This illustrates the George Soros reflexivity or feedback principle, whereby markets affect psychology and psychology affects markets. Positive feedback is self amplifying, while negative feedback is self corrective. For example, collateral values rise during a bull market. This increases their access to lower priced funding and liquidity, which fuels further gains. Finally, regret leads to confusing risk with wealth. Larger, better-capitalized financial institutions can absorb more risk than smaller institutions. Their greater risk tolerance lessens their downside sensitivity, especially during a bull market

when income levels are high. Thus, risk appetite increases with wealth. Risk and return are, however, scale invariant. Larger institutions confuse the ability to absorb risk provided by capital with the desirability of the risk position. Therefore, they acquire underpriced, higher-yielding, higher-risk assets in bull markets.6

B. Overconfidence
Overconfidence occurs when we exaggerate our predictive skills and ignore the impact of chance or outside circumstances. It results in an underestimation of outcome variability.7 Overconfidence is reinforced by self-attribution and hindsight. Self-attribution involves internalizing success while externalizing failure. Structured finance bankers and quantitative risk managers took credit for results during the boom, failing to consider the impact of randomness and mean reversion creating an illusion of control.8 Hindsight involves selective recall of confirming information to overestimate their ability to predict the correct outcome, which inhibits

This is consistent with the H. Minsky financial instability hypothesis. Investors increase their risk exposures driving bull markets until they have taken on too much. See H. Minsky, 2008, Stabilizing an Unstable Economy McGraw-Hill.
6

This is magnified by the naïve use of market-based risk-management tools.
7

Studies indicated the underestimate at 15 %-25 %. The direction of the overconfidence is usually positive reflecting a related optimism bias.
8

RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS

87

Figure 3. Risk/Market Appetite, Structured Finance
Risk Appetite Low High

Housing bubble 2004-1H07 Bull

High

Market State

H
Contrarian 2H07 - ?

Risk Level

Bear

Low

Low Financial Institution Profitability

High

learning. Disappointment and surprise are characteristics of processes subject to overconfidence. Industry and product experts are especially prone to overconfidence based on knowledge and control illusions. Knowledge is frequently confused with familiarity. This is reflected in the number of industry experts including most famously the former Federal Chairman who missed the collapse of the housing and structured credit bottom.9 This is due, in part, to misguided overreliance on quantitative credit scoring models without understanding their limitations. Key model limitations include the following:

sequence and is not an independent observation. History becomes less relevant as markets and underwriting practices change. This was especially true for mortgage default models. They ignored the impact of securitization of mortgage originator underwriting practices.10 • Uncertainty: Decisions involve both risk, known unknowns, and uncertainty, unknown unknowns, elements. Financial models adequately contemplate the former but inadequately deal with the later. Managing uncertainty requires judgment, not calculation. Control reflects the unfounded belief of our ability to influence or structure around risk. Risk is accepted because we believe we can escape its consequences due to our ability to control it. Examples include the perceived ability to distribute or hedge risk, independent of the likelihood of being better or faster at identifying risk than the market. This reflects an optimistic underestimate of costs while overestimating gains. Optimism is heightened by anchoring when disportionate weight is given to the first information received. This is usually based on the original plan, which tends to support the transaction.

• Homogenous populations: Statistical models require large
homogenous populations with a long history of observations. The new structured finance credit portfolios were small, heterogeneous, and concentrated with limited histories. • Statistical Loss Distribution: Loss distributions for credit are skewed, with unexpected event losses hidden in the distribution’s fat tails. Models tend to be blinded by the mean and underestimate extreme events. • Historical basis: History is a guide, not the answer. The past represents but one possible outcome from an event

Inappropriately designed incentive compensation reinforces overconfidence.
9

U. Rajan, A. Seru, V. Vig, “The Failure of Models that Predict Models: Distance, Incentives and Defaults,” Working Paper, September, 2008.
10

88

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Time-delayed consequences magnify overconfidence as purchased to hedge the position. Since this is not priced, it individuals weigh short-term performance at a higher level creates incorrect capital allocation incentives.14 Thus, the than longer-term consequences. These occur whenever short- “lucky fool” is rewarded and encouraged with bonuses and term benefits clash with long-term effects. Although we know increased capital until luck turns and losses are incurred. of the potential negative long-term effects, we believe that Examples include the numerous apparently lucky real estate they will not happen to us, at least during the current experts at institutions like Bear Stearns and Lehman. accounting period. An example is dropping credit Eventually, all lucky streaks come to an end as this one did underwriting standards to increase short-term income, market during the summer of 2007. share, or league table status Another statistical error as occurred during the height prevalent during a boom is Behavioral finance examines how of the boom. extrapolation bias. This occurs when current events or risk managers gather, interpret, C. Statistical trends are assumed to and process information. It continue into the foreseeable Statistical bias involves future, independent of recognizes that models can confusing beliefs for historical experience, sample probability and skill for size or mean reversion. influence behavior and shape chance by selecting evidence Undoubtedly, this resulted in decisions. in accordance with our many of the projections expectations.11 Economics is underlying structured credit a social science based on human behavior. Prices are not proposals. The major error focused on the belief that housing determined by random number machines.12 Rather, they come prices would not decline nationwide in the US from trades by real people. Feedback loops, prices, trades Perhaps the most dangerous statistical bias is disaster and people complicate statistical modeling, and invalidate myopia. This occurs whenever low-frequency but highthe use of normal distributions as used in the physical sciences. impact events are underestimated. Since the subjective Institutions find it difficult to accept chance and are probability of an event depends on recent experience, frequently fooled by randomness. A manifestation is the expectations of low-frequency events, like a market or firm representative bias, whereby we see patterns in random collapse, are very small. These types of events are ignored events. We interpret short-term success as “hot hands” by a or deemed impossible, particularly when recent occurrences skilled banker. Risk-adjusted return on capital and other are lacking. This causes a false sense of security as risk is measures are unable to distinguish results based on luck versus underestimated, or assumed away, and capital is misallocated. skill. Unlikely events are neither impossible or remote. In fact, Statistically based risk management practices are inherently unlikely events are likely to occur because there are so many limited. They are unable to reflect the hidden risk that the unlikely events that can occur.15 Thus, the longer the time state of the world may change rendering current state data period, the higher the likelihood of a “Black Swan” event obsolete. For example, switching from a boom to a bust cycle occurring.16 impacts correlations. Formerly diversified positions begin moving together, triggering unexpected losses. They are D. Herding unexpected because such movements are unfamiliar. We tend to view the unfamiliar as improbable, and the improbable is The previous discussion concerned individual frequently ignored. psychological aspects of risk decision making. There are Actions and outcomes can be unrelated. Consequently, it also social aspects to decision making when individuals are becomes important to examine the decision process and not influenced by the decisions of others as reflected in herding just the outcome.13 As Scholes notes, to value risk or price and ‘group think’. reserves you must reflect the values of the options not
M. Scholes, “Crisis and Risk Management,” AEA Papers and Proceedings, May, 2000
14

See P. Bernstein, 1996, “The New Religion of Risk Management,” Harvard Business Review (March-April).
11 12

15

P. Bak, 1996, How Nature Works Springer-Verlay.

W. Sharpe, 2007, Investors and Markets, Princeton University Press 11. P. Rosenweig, The Halo Effect, Free Press, 2007.

13

Black swans are high impact unexpected rare events. The term was popularized by N. Taleb in The Black Swan: The Impact of the Highly Improbable (Random House, 2007).
16

RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS

89
Consequently, mutually reinforcing individual biases and unrealistic views are validated.20 Experts are prone to group think. They tend to limit information from all but other expert sources. Thus, they repeat statements until they become accepted dogma regardless of their validity, due to a lack of critical thinking. The recent subprime collapse illustrates this fact. The industry participants used the same consultants and models for their projections. The consultants based their reports and recommendations on the surveys of industry participants. Once the perception of a bull market took hold, it was reinforced and accepted uncritically. When the crash occurred, the experts were taken by surprise by a supposed perfect storm. This is illustrated in the 2006 Business Week cover story in which risk officers at numerous institutions, including Bear Stearns and Lehman, are surveyed.21 They believed that despite the risks taken they were safer than ever. This belief was based on complex risk models and market diversification. The faith in risk management encouraged institutions to increase their risk exposures, believing they were under control.

Herding occurs when a group of individuals mimic the decisions of others. Through herding, individuals avoid falling behind and looking bad if they pursue an alternative action. It is based on the social pressure to conform, and reflects safety by hiding in the crowd.17 In so doing, you can blame any failing on the collective action and maintain your reputation and job. Even though you recognize market risk, it pays to follow the crowd. Managers learn to manage career risk by clinging to an index. Essentially, principal loss is converted into benchmark risk. Herding reduces regret by rationalizing that you did no worse than your peers. It constrains both envy during an upswing and panic in a down market. This is critical in banking when performance contracts are based on relative performance measures tied to peer groups. 18 Herding underlies why banking experts’ forecasting abilities are poor. The experts tend to play it safe by staying close to the crowd and extrapolating past performance.19 A related effect is an informational cascade. A cascade is a series of self-reinforcing signals obtained from the direct observation of others. Individuals perceive these signals as information even though they may be reacting to noise. This is referred to as a positive feedback loop or momentum investing, which can produce short-term self-fulfilling prophecies. Herding amplifies credit cycle effects, as decisions become more uniform. The cycle begins with a credit expansion leading to an asset price increase. Investors rush in to avoid being left behind using rising asset values to support even more credit. This explains why bankers continued risk practices even though they feared this was unsustainable and leading to a crisis. Eventually, an event occurs, such as a move by the central bank, which triggers an asset price decline. This causes losses, a decline in credit, and an exit of investors, which strains market liquidity.

F. Sentiment Risk
The aggregate investor error based on biases is sentiment risk. It can be either optimistic or pessimistic and is time varying as reflected in Figure 4. Sentiment risk is zero in an efficient market. Paul Samuelson has noted markets in the short-term can be micro efficient concerning individual instruments, but macro inefficient regarding the market as a whole. Additionally, during the short-term the direction of the inefficiency is likely to widen due to momentum and herding.22 Most risk models ignored sentiment risk. This causes losses when sentiment changes leading to closed markets and mark to market losses which has threatened the basis of originate to distribute model. Investor responses based on the interplay of sentiment and market valuation is reflected in Figure 5. During a late stage boom with high sentiment levels, A, behavioral risk factors will dominate and quantitative risk measures will be unreliable. This is reflected in the famous comment “As long as the music is playing, you have to get up and dance”. This is characterized as irrational exuberance where prices are driven principally by momentum and herding
See J. Chevalier and G. Ellison, 1997, “Risk Taking by Mutual Funds as A Response to Incentives,” Journal of Political Economy.
20

E. Group Think
Group think, or organizational pressure, enhances cognitive biases. It occurs when individuals identify with the organization and uncritically accept its actions. Once the commitment is made, inconsistent information is suppressed.

17

This is reflected in Keynes” famous statement that it is better for a banker ’s reputation to fail conventionally, than to succeed unconventionally.

The industry expert impact is significant, as most large financial institutions adopted best practices based on similar experts.
18

Relative performance measures are a form of sophisticated “me-too” metrics. Rather than focus on absolute value creation, they focus on arbitrary market silos that may be in a downturn.
19

E. Thornton, D. Henry and A. Carter, 2006, “Inside Wall Street’s Culture of Risk,” Business Week (June 12).
21 22

H. Shefrin, 2008, Ending the Management Illusion, McGraw-Hill.

90
Figure 4. Senitment Risk

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

B A EA: EB: C: D: fundamental value financial Price credit bubble: optimism liquidity trap: pessimism

E C

D

Time

Figure 5.
Sentiment High Low

Market Valuation

Boom

A

B

Bust

D

C

reflected in high liquidity levels. When sentiment is low, fundamentals will rule as in B and C. Price may diverge from fundamentals 23 , but are quickly eliminated by arbitrageurs in B and C. D represents irrational despondency found in market bottoms reflected in tight liquidity.

II. Remedies
Behavioral finance demonstrates how biases influence risk perception leading to underestimation of improbable events. We base our actions on experience of what has happened. This ignores beyond the data exposures leading to future blindness. Consequently, we misjudge actual risk leading to surprise losses. Recognizing and dealing with biases is complicated by three factors. First, bias can be amplified within organizations due to incentive misalignment and group think. Next, we diminish information inconsistent with our existing views, while searching for conforming information. Finally, this leads to

See T. Debels, Behavioral Finance (Garant Uitgevers, 2006) 183 for a discussion of various forms of behavioral finances that can occur in markets.
23

RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS

91 • Obtain a second opinion: Anticipation of review by an

a false sense of security and reduced vigilance. Potential responses to reduce biases will now be explored.

unrelated third party encourages greater care. • Beware of experts: Seek diversity to avoid myopic focus A. Principles on issues within an expert’s area of interest. Experts frequently ignore the benefits of alternatives. Historically, risk management has been primarily an • List a wide variety of possible scenarios: Focus on exposure accounting and control system. While this unpopular and unlikely controllership reporting possibilities. View the future function is important, No matter how remote, highas a collection of eventualities protection is needed against rather than as a single it becoming a regulatory and impact events cannot be ignored prediction. The preferred quantitative ritual. The decision is one that works because they can threaten an emphasis should be on across several possible forward-looking dynamic institution’s existence, as was eventualities, and not just the management, which involves current market state. demonstrated in the current three components: • Avoid herding: Develop 1. Risk appetite involves market crisis. independent analysis. This determining how much an requires encouraging institution is prepared to contrarian views supported by compensation programs. lose, and is frequently defined by earnings at risk or a potential • Postmortem: Review both successful and unsuccessful ratings downgrade. decisions. The focus should be on whether the results were 2. The actual institutional risk profile is monitored through scenario analysis and stress testing applied against actual luck or skill-based. The key is to avoid rationalizing and hindsight bias and to learn from the experience. portfolio movements. • Directly engage the environment: Independent 3. Corrective action is taken when a mismatch between the organization’s risk profile and appetite occurs. The corrective investigation is needed to verify and to avoid filtered action can be at the transaction, when involving major information. • Heterogeneous risk team: Construct a diverse exposures, or portfolio level. independent risk team. Rotations can be used to maintain Investors have difficulty in processing market signals. This diversity. comes from a failure to distinguish noise, price movements • Lengthen risk horizon: Given the long-tail nature of credit without meaningful changes in economic prospects, from true risk, increase the evaluation horizon beyond the traditional information. This difficulty places decisions at risk (DAR) accounting-based yearly horizon. as reflected in Figure 6. Uncertainty, as opposed to risk, is difficult to manage due to several biases. Chief among these are optimism and overconfidence based on an illusion of control based on flawed models. These biases are amplified in certain organizations by compensation and governance problems. Bureaucracy and opaqueness inhibits responses until it is too late, leading to massive losses. No process is foolproof. A backup procedure is needed to remove the temptation to accept unintended catastrophic risk. This is provided by portfolio control and enforced by the board of directors. Excess concentrations must be reduced or covered by additional capital. Also, strategies are needed for each position, allowing adaptation to random changes in market states. Institutions should invest in portfolio strategies, not in illiquid excess concentrations, which have two components. The first concerns risk acceptance or transactional approval based on the institution’s risk underwriting criteria. Second is risk reduction through diversification. Diversification, however, does not prevent losses. Rather, it prevents losing everything at one time. The focus is on position size to avoid over betting. Institutions must guard against unproductive naïve diversification, which emphasizes the number of portfolio assets instead of their asset-class diversity. This is critical because correlations are scenario specific and

B. Alternatives
Practical alternatives to biased based bounded rationality exist. Some options include:

• New markets and product limits: Behavioral bias is strongest in areas where inexperience reigns, such as in the structured finance area with its new technology and limited history. Thus, strict limits to control exposure in these areas are needed.

92
Figure 6. Decisions at Risk (DAR)

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Uncertainty Beyond the data events Experiences Exposures Black Swans Rare Large Impact

Bias Optimistic Overconfident Illusion of control

Amplifiers Incentives Bureaucracy Opaqueness

approach one during a crisis. Thus, the key is the overall asset-class allocation and not necessarily the number of assets in an asset class.

yield curve reduced the attractiveness of the carry trade, putting pressure on institutional accrual and trading budgets. In the search for yield, institutions adopted a procyclical asset heavy 5Ls strategy:

III. Current Crisis
A. Setting
A herding among financial institutions occurred during the last several years. Consequently, they invested too much at the same time in the same areas. This was done under guise of the “originate to distribute” model. This model allowed institutions to rationalize poorly structured, underpriced products by selling them to others. They failed, however, to consider the impact of markets closing down, leaving them with large high risk exposures. Some institutions like Wells Fargo and Pittsburgh National escaped the herding. It is difficult to determine if they were lucky or smart. The pressure to herd is illustrated by Morgan Stanley. Under its previous management, Morgan Stanley refused to participate in structured products. Its performance suffered relative to its peers. Consequently, in 2005 it was replaced by a new team. They vowed to regain market share by matching its peers, which it achieved in 2008 by recording record losses. A declining economy and falling markets triggered aggressive Federal Reserve interest rate cutting and liquidity injections in 2001 to 2002. Liquidity-driven technicals improved, resulting in falling risk premiums increasing credit asset prices. Institutions responded by adopting an assetintensive carry trade strategy, which involves borrowing shortterm to invest in longer-term risk assets. A credit bubble formed as liquidity-driven technicals surpassed fundamentals. This was reflected in historically low credit-risk spreads in the real estate, leveraged buyout and structured credit markets. Spread narrowing and a flattening

• Longer duration (e.g., mortgage backed securities) • Long tailed option type risk found in the AAA tranches
of structured securities • Large positions (e.g., multibillion dollar mortgage warehouse facilities) • Leverage levels approaching 30:1 • Less liquid assets (e.g., collateralized debt obligations) • The 5Ls strategy involves going long on higher-risk assets for the institution’s own account instead of distribution. The strategy is reflected in principal finance, merchant banking, bridge loans and warehousing activities. These activities represented up to 75% of revenues at some institutions. The risk inherent in the 5Ls strategy were obscured by judgment biases in the following areas:

• Unproven business models were justified based on optimistic plans while down-playing the negative possibilities due to group pressures. • Institutional overconfidence in risk management models based on the illusion of control. This causes an observation of safety, which creates as illustrated by the Peraud Paradox, risk.24 • Peer pressure that was not based on independent economic reasoning. Nonetheless, as the Morgan Stanley example illustrates, ignoring peer pressure can be hazardous to your career.

Persaud Paradox is the observation of safety created by using the same models as your peers, which creates model risk.
24

RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS

93
It is difficult to price rationally when risk seems remote and hard to measure and conditions seem favorable.25 The last market correction had occurred more than three years ago and was largely forgotten by the first half of 2007. Thus, risk sensitivity had diminished. This recognition problem is rooted in the complex nature of cyclical risk.

This caused a major credit boom. During the boom, prices can exceed underlying fundamental economic values as illustrated in Figure 4. Such cycles, while predictable, are difficult to manage for several reasons. First, financial institution compensation is tied to peer group comparisons. Thus, firms and individuals not following their peers suffer. Next, organizations frequently discourage pessimism. Therefore, conservative risk managers and bankers are pressured to become optimistic or leave. Finally, institutions risk losing bankers if their risk activities are curtailed. Frequently, positive short-term results mask long-term risks. Seemingly high returns can reflect the subjective probability of an event that has not occurred in the time period studied. Investing in such instruments is profitable most of the time. Eventually, a beyond the data event occurs. The housing event occurred in mid-2007 and has continued for more than eighteen months, costing billions in provisions. Individuals and institutions succumbed to a bias of “assuming the absence of evidence implied evidence of risk absence”.

C. Regime Changes
Procyclical risk appetite and feedback loops underlie credit cycles.26 As risk appetite increases, credit extension expands. Investors use the increased debt capacity to bid asset prices higher. The higher asset prices increase collateral values, which supports additional credit expansion creating a virtuous credit cycle with increasing liquidity. A tipping point or event can, however, prompt investors to adjust simultaneously their positions triggering a decline in asset prices.27 Figure 7 shows that the decline can trigger a vicious cycle leading to reduced collateral values, curtailed credit, declining investor demand, falling asset prices and reduced liquidity. The tipping point represents a change in investor sentiment based on an awareness of the risk that investors have assumed. Once the tipping point is reached, feedback overwhelms fundamentals and the trend dominates. Tipping points represents triggers, not causes of the change in investor actions. Overvalued assets, which are vulnerable to bad news, are prone to volatile investor sentiments. Thus, tipping points are unexpected and occur during the height of an over-valued bull market.

B. Concerns
The appropriateness of the 5Ls portfolio strategy depends on several factors. First, it works best early in the cycle before the opportunities are exploited by the competition and spreads narrow. Next, the strategy involves incurring increased systematic or beta risk exposure versus value-adding alpha returns. Structured products are less liquid than market investments. Consequently, the return on structured products reflects compensation for liquidity risk. This risk was poorly reflected in risk management models. The liquidity premium was mischaracterized as alpha. Thus, liquidity risk was under reported. This was subsequently discovered during the crisis. Finally, pricing and trading discipline is needed to ensure an adequate risk premium is earned. Maintaining discipline becomes increasingly difficult as the cycle continues. Warning signs began to form during the first six months of 2007:

D. Reinforcement
The complexity of low-frequency/high-impact cyclical risk is compounded by institutional factors such as budgets and compensation systems that reinforce the behavioral bias effect. These systems favor “consistent” earnings and misread low –frequency/high-impact risk “profitability.” Such risk is similar to underwriting out-of-the-money put options. The premiums appear profitable until the put event occurs. Risk models also contribute to the problem by presenting the illusion of safety and control, leading to over optimism.
25 President of New York Federal Reserve T. Geither, as quoted in the Financial Times, May 12, 2005.

• Continued Federal Reserve tightening • Rating agency downgrades • Flattening yield curve • Increased mortgage defaults
Unfortunately, apparent success breeds an inability to imagine the possibility of failure, and the warnings were ignored. Firms continued to underestimate the likelihood and impact of unlikely events. Widespread credit risk under pricing existed due to an emphasis on nominal returns. This suggests a correction when investor emphasis shifts from return on capital to return of capital.

Procyclical risk appetite is aggravated by ratings-based regulatory capital requirements. The regulatory rules reduce capital requirements during a bull market as ratings increase, thereby encouraging credit expansion. Conversely, they increase capital requirements during a bear market as ratings are pressured, leading to a credit contraction.
26

See D. Sornette, 2003, Why Markets Crash, Princeton, NJ, Princeton University Press. This is similar to physical events such as forest fires and earthquakes arising from “criticality.”
27

94
Figure 7. Credit Cycle

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Virtuous Cycle (Disaster Myopia) 2004-1H07

Vicious Cycle (Disaster Magnification) 2H07 - ?

Asset Prices

Tipping Point

Asset Prices

Investors

Collateral Values

Investors

Collateral Values

Credit

Credit

Backward-looking risk models confuse history and science. Unfortunately, financial markets are not actuarial tables. Some important model issues include the following:

• Inadequate consideration of the cyclical effect on, and correlation among, probability of default, loss given default and exposure at default • Poor understanding of the interaction between liquidity and credit risk as bull markets create their own liquidity, which can evaporate in a downturn • Feedback impact of models on markets is ignored. The observation of safety created by using the same models as your peers creates model risk.28 • Difficulty reflecting out-of-sample, beyond-the-data possible effects.
Consequently, models underestimated low frequency/highimpact cyclical risk. The underlying exposure builds during a bull market as apparent risk declines, while the losses materialize in the bear market cycle. This anomaly is due to social and psychological biases resulting in bounded rationality. Ignoring these facts substitutes an inaccurate normative model for the real world. The objective is to supplement existing quantitative risk management with developments taken from the evolving field of behavioral finance. In so doing, it can reduce future losses
This is the “Persaud Paradox,” where the observation of safety creates risk. See A. Persaud, 2003, Market-Liquidity and Risk management, Liquidity Black Holes, London: Risk Books, Ch. 11.
28

during the credit cycle as risk management evolves to a more balanced system, incorporating human behavior. This requires taking low-probability-worst-case scenarios seriously, and developing appropriate responses. The process is similar to earthquake engineering, which does not attempt to predict a shock. Rather, the focus is on constructing a structure to withstand a certain shock level. Currently, counter cyclical capital charges decrease during bull markets as ratings improve as demonstrated in Figure 3. This fact underlies the procyclical bias in portfolio strategies as lower bull market capital requirements increase returns, encouraging an inappropriate, asset-heavy, 5L portfolio strategy. Supplementing currently determined capital charges with a requirement tied to asset prices would encourage a shift to a counter cyclical portfolio strategy. Capital levels would relate to changes in asset prices. The higher capital allocation serves as a risk-taking budget constraint during bull markets by dampening compensation-related returns. 29

E. Lessons Learned
Risk decisions are at a risk from behavioral bias. This is especially true when dealing with high impact low probability risks. Governance mechanisms represent possible control over the bias by introducing outside viewpoints. The specific action taken depends on the source of the bias. Figure 8
See C. Goodhart, 2005, “Financial Regulation, Credit Risk, and Financial Stability,” National Institute of Economic Review ( Apr.il), 118.
29

RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS

95

Figure 8. Bias Response Choices

Managemen Biased Rational

Rational Investor Markets Biased

A A

B

D

C

reflects the possible combination of market and managerial bias. Biased managers operating in an efficient market, A, need to be protected by their boards of directors and regulators from overreacting to market noise through tight controls. Classical financial theory is represented in B with efficient markets and rational managers who require limited oversight. Rational managers operating in a biased market, C, will exploit market inefficiencies by selling over priced claims. When both managers and markets are biased, D, which characterized the late stage of the boom, the situation becomes problematic. Boards and regulators are likely to fall prey to the same behavioral biases as affecting mangers and controls are likely to fail.

IV. Conclusion
Presently, it is difficult to consider the end to the bear market. Defaults are increasing and liquidity remains fragile. This difficulty is compounded by behavioral bias reinforced

by institutional factors. While no two cycles are identical, we must resist the temptation to say, “This time is different.” The deeper we are into illiquid credits, products and structures, the more difficult it becomes to manage risk. The key is to identify potential adverse scenarios, stress-test to determine their impact, compare the test results to our risk appetite and take appropriate portfolio decisions. This entails adopting counter cyclical portfolio strategies despite negative short-term revenue implications. This also requires adopting difficult infrastructure changes. Organizational obstacles inhibit appropriate responses to high-impact low-probability risks. Chief among the obstacles are short-term compensation systems which reinforce behavioral biases. This leads to a potentially fatal neglect of the longer-term build of risk. As Robert Merton noted “The amount of risk we take personally, individually, or collectively is not a physical given constant. We chose it.”30 Behavioral finance offers a means to choose wisely, as it affects both individual decision making and market efficiency. You ignore behavioral risk at your own peril.

30 R.C. Merton, interviewed by N. Nickerson, 2008, “On Markets and Complexity,” Technology Review (April 2).

96 References
Bak, P., 1999, How Nature Works: The Science of Selforganized Criticality, Springer Verlag, New York. Bernstein, P., 1996, “The New Religion of Risk Management,” Harvard Business Review 72 (No. 2), 4751. Black, F., 1986, “Noise,” Journal of Finance 41 (No. 3), 529-543. Chevalier, J. and G. Ellison, 1997, “Risk Taking by Mutual Funds as A Response to Incentives,” Journal of Political Economy 105 (No. 6), 1167-1200. Debels, T., 2006, Behavioral Finance, Garant Uitgevers, Antwerpen. Ellsberg, D., 1961, “Risk, Ambiguity and the Savage Axioms,” Quarterly Journal of Economics 75 (No. 4), 643669. Friedman, M. and L. Savage, 1948, “The Utility Analysis of Choices Involving Risk,” Journal of Political Economy 56 (No. 4), 279-304. Goodhart, C., 2007, “Financial Regulation, Credit Risk and Financial Stability,” National Insitiute of Economic Review 192 (No. 1), 118-127. Merton, R.C., 2008, “On Markets and Complexity,” Interview by Nate Nickerson, Technology Review (April 2). Minsky, H.P., 2008, Stabilizing an Unstable Economy, McGraw Hill Professional, Columbus, OH.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Persaud, A., 2003, Liquidity Black Holes: Understanding, Quantifying and Managing Financial Liquidity Risk, Risk Books, London. Rajan, U., A. Seru, and V. Vig, 2008, “The Failure of Models that Predict Failure: Distance, Incentives and Defaults,” Chicago GSB Research Paper No. 08-19, Ross School of Business Paper No. 1122. Rosenzweig, P.M., 2007, The Halo Effect—and the Eight Other Business Delusions that Deceive Managers, Simon and Schuster, New York. Scholes, M., 2000, “Crisis and Risk Management,” American Economic Review 90 (No. 2), 17-21. Sharpe, W. F., 2006, Investors and Markets: Portfolio Choices, Assets Prices, and Investment Advice, Princeton University Press, Princeton, NJ. Shefrin, H., 2008, Ending the Management Illusion, McGraw-Hill, Columbus, OH. Sornette, D., 2003, Why Markets Crash, Princeton University Press, Princeton, NJ. Taleb, N., 2007, The Black Swan: The Impact of the Highly Improbable, Random House, New York. Thornton, E., D. Henry, and A. Carter, 2006, “Inside Wall Street’s Culture of Risk,” Business Week (June 12). Tversky, A. and D. Kahmeman, 1992, “Advances in Prospect Theory: Cumulative Representation of Uncertainty,” Journal of Risk and Uncertainty 5 (No. 4), 297-323.

University of Rochester Roundtable on Bankruptcy and Bailouts: The Case of the US Auto Industry
The GeVa Theatre Rochester, NY February 2, 2009
Panelists: Thomas Jackson, Charles Hughes, James Brickley, Joel Tabas, and Clifford Smith Moderator: Mark Zupan

 Mark Zupan: Good evening, and welcome to this discussion of a very topical and pressing issue: today’s problems with the US auto industry, and the potential role of bankruptcy in dealing with them. I’m Mark Zupan, Dean of the University of Rochester’s Simon School of Business, and I will be serving as moderator. I’m not going say much about the topic itself—I’ll leave that to our panelists, who are the experts. What I will tell you is that bankruptcy, like business school applications, is a “negative beta” activity. In other words, when the market’s up, both business school applications and bankruptcy cases tend to go down. But when the market’s down, our applications go in the reverse direction, and so does the amount of attention and effort devoted to bankruptcy. We have five panelists tonight. Three of them—Tom Jackson, Cliff Smith, and Jim Brickley—are distinguished academics from the Simon School faculty. The other two— Charlie Hughes and Joel Tabas—are both Simon School alums who have gone on to become accomplished “practitioners” in their own fields, Charlie as an auto company executive and Joel as a bankruptcy lawyer. I’ve asked each of our five panelists to provide a brief statement of their thoughts on the problems of the US auto industry, and possible solutions, including Chapter 11, to those problems. After we hear from each of them, we’ll open up the discussion to questions from the audience. Our first speaker will be Tom Jackson. Tom—along with

his former student, Douglas Baird, former dean of the University of Chicago Law School—is widely regarded as one of the world’s top two authorities on US bankruptcy law. We feel very privileged to have him at the University of Rochester. From 1995-2005, Tom served as President of the University. Since stepping down from that position, he has held joint appointments at both the Simon School and in the University’s political science and economics departments. Before coming to Rochester in ’95, he was the provost and dean of the University of Virginia Law School. So, Tom, would you please start things off for us?

I. The Social Function of Bankruptcy: Uses and Limitations
Tom Jackson: Thanks, Mark, for the kind words. Let me start by saying how much I appreciate this Depression-era stage set that GeVa has provided as the backdrop for our discussion tonight—it seems very appropriate for the topic. I want to begin this discussion by providing a broad economic framework for this issue of bankruptcy vs. bailouts because I suspect we haven’t seen the last of businesses—or industries—facing such choices. My field, as Mark told you, is bankruptcy—and bankruptcy is a process for reorganizing troubled companies that is rooted in the economic goal of increasing efficiency. Bailouts, by contrast, are a means of

97

98
rescuing troubled companies where, for good or ill, politics tend to mix with and override fundamental economic considerations. So I’d like to talk about what bankruptcy can do, and perhaps what it can’t—and I’ll do so in the context of the recent controversy over about what the Detroit automotive manufacturers should have done. Chapter 11 is designed to do one thing well—and, for the most part, I think it does so. And that is to rearrange the capital structure of companies with more debt than assets to allow those that should survive to survive—and allow those that should fail to fail. The criteria for survival in such cases are economic ones: can the troubled company, if properly reorganized and recapitalized, be made profitable enough for its new investors to earn a fair rate of return on their money? If the answer is yes—in which case, presumably, the new capital will be provided—the company gets reorganized under Chapter 11. But if the answer is no, the best outcome for the original investors is to shut down the business and sell the assets piecemeal to the highest bidders, either in Chapter 11 or after converting to Chapter 7. Whether bankruptcy or bailouts, however, it’s important to recognize that there is a difference between financial failures and business failures. Financial failures are cases where the assets, although valuable when kept together as part of a going concern, are worth less than the liabilities— and these companies, as a general rule, get reorganized in and come out of Chapter 11. Business failures, by contrast, are cases where the assets themselves are worth less when continued as part of a firm—even if the firm were to be recapitalized or given new money—than sold off piecemeal to new owners. In practice, of course, we often see elements of financial and business failure mixed together. But Chapter 11 is premised on the idea of separating these two ideas in such a way that companies facing a financial but not a business failure will be reorganized and continued—and business failures will be sold off in parts. To see this distinction, consider the case of Johns Manville in the 1970s, a company that appears to have been a very efficient manufacturer of building supplies. The company became hopelessly insolvent not because of any problems with its then-current business line, but because of the tort liability associated with its manufacturing of asbestos 20, 30, and 40 years earlier. Keeping the company going—which required writing down the claims against it and converting many of them to equity interests—was the right outcome since Manville’s was a financial and not a business failure. And, again, Chapter 11 is designed to do just that. Conversely, one can have a business failure without a financial failure. My family had a business in Kalamazoo, Michigan that made gas lights at the turn of the century—a business that was not a growth industry in a world of electric light bulbs. Now, because it made very little use of debt, the

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

business was able to survive and be converted, over the course of 50 years, into one that makes pneumatic air cylinders— which it continues to do to this day. But if that business had instead been financed with debt, it would almost certainly have filed for bankruptcy. Unless new owners and investors could be convinced that the existing management could effectively make the transition to a new business, the assets would have been sold off in a Chapter 7-type proceeding— and, sooner or later, someone else would have entered the business of pneumatic air cylinders. But as I suggested earlier, most corporate failures—even those in very large companies—tend to result from a mix of financial and business failure. Part of the blame in such cases can be laid to having the wrong business model, and the current management team may not be quite up to the task. But much of the current problem can also be attributed to past business mistakes in combination with accumulated debts and liabilities that the current management may or may not be responsible for. And before one can discuss Detroit—and bankruptcy— one needs to figure out which model it fits: Is it mainly a financial failure, a problem that can be addressed largely by rewriting claims and contracts and providing new capital? Is it really at bottom a business failure? Or does it have elements of each that need to be addressed? And, I hasten to add, the same questions need to be asked when designing government “bailouts” as well. It makes no sense to bail out a failed restaurant that was operated by mom and pop. Mom and pop will leave the scene, and someone else will take their place. Any intervention by government will only make things worse. Detroit has a 40-year—perhaps longer—history of decisions and actions that, in retrospect, have turned out to be wrong. Some, though by no means all, can be blamed on past management. As a result, one or more of the manufacturers in Detroit are almost certainly insolvent in the classic sense: that is, their liabilities exceed their assets. Any solution to Detroit’s problems has to figure out how to get these things back in line. I suppose giving them money from the government is one way to do it. But is it the best way? And when it comes to addressing the question of business failure, one or more of the manufacturers in Detroit are probably also not “efficient” producers any more. But, again, that’s not necessarily because its current management is incompetent, but because the accretion of mistakes over the past 40 years has produced manufacturing operations that are not as efficient as its competitors’. But other than noting the consequences for operations today, the real need here isn’t to explain the past. The most important, and often overlooked, question is how to deal with the future. How do we identify and save those parts of the US auto industry that are worth saving? And how do we

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

99

ensure that whatever companies emerge from the current mess sounds: by trying to prop up less efficient enterprises, you are profitable enough to stand on their own, and so avoid impose large costs on the rest of the economy—on US creating permanent corporate dependents? consumers, who end up paying higher prices for cars; on US And I think it’s important here to begin by identifying the taxpayers, who foot the bill for today’s (as well as tomorrow’s) fundamental issue, one that often seems to be ignored in the subsidies. You also impose costs on other, more efficient current debate: Is there too much manufacturing capacity competitors who, although they may be foreign companies, going forward in the US employ lots of US auto industry? I’d say workers—and these “yes, without question.” companies will get We need to pull huge capacity out of the Rather than a baseline of dragged down by the system. We can take it out of one or more 16 million cars, we need excess capacity to contemplate a baseline preserved by any bailout. of the Detroit manufacturers, or we can of 12-13 million cars. What basis do I have take it across the board—but either way, Auto manufacturing, to for my claim that Detroit be sure, has always been is less efficient? There the capacity needs to come out. We need a cyclical business— are many ways to count again, I know first-hand, it, but let me name just a to deal with the consequences of doing having grown up in few. Let’s start with the that. It won’t be pretty. It’s going to Michigan. Cyclical number of different businesses will fluctuate. kinds of vehicles. GM, mean shutting down plants, car dealers, But there are a lot of which has well over a and suppliers—and putting people out of reasons—cars that last half-dozen major longer, perhaps a shift in “brands” of cars in the work. Once you start with this premise, cars as a “status symbol,” US alone—not counting and the reality that, even distinct brands such as you then have to ask which method, in the early years of this Holden in Australia or bailout or bankruptcy, is likely to decade, demand seemed Opel in Europe—is the to be kept artificially high only manufacturer in the accomplish this downsizing in the most through a number of world I can think of with cost-effective way. devices such as “rebates” more than three lines in and fleet sales—to think one country. Along with -Tom Jackson that the fluctuations are too many models, GM likely to be around a also has far too many median level that is two or three million vehicles smaller than dealers. Both of these are the consequence of early- to midit had become over the past decade. 20th-century mergers and an earlier strategy that is reflected Now, if these estimates are correct, then that is the gorilla in the company’s name—General Motors. With a business in the corner. It means that we need to pull huge capacity out model that seemed to work in the 1950s, GM encouraged of the system. We can take it out of one or more of the Detroit new buyers to start by buying Chevys and, as they worked manufacturers, or we can take it across the board—but either their way up the economic ladder, to move to Pontiacs, then way, the capacity needs to come out. We need to deal with Oldsmobiles, then Buicks, and finally Cadillacs. But that the consequences of doing that. It won’t be pretty. It’s going model made less and less sense as we entered the latter part to mean shutting down plants, car dealers, and suppliers— of the 20th century and the first part of the 21st century. and putting people out of work. Once you start with this Changing strategies, under the best of circumstances, would premise, you then have to ask which method, bailout or have been difficult—although that doesn’t explain why GM bankruptcy, is likely to accomplish this downsizing in the continued to add brands, such as Saturn and Hummer. And most cost-effective way. change was made much more difficult by a franchise system Now, it’s probably true that if you decide to take capacity for dealers that, with the help of state politicians and law, out of the automobile industry as a whole rather than just was effectively frozen in place—and ensured the continued Detroit, you will “save” jobs. But that is true precisely existence of too many brands. because Detroit is less efficient than the rest of the industry; Besides too many models and dealers that cannot be any time you take jobs out of companies that are more dropped without major expense, another cause of Detroit’s efficient, you probably save jobs. But this is as perverse as it current problems was their successful efforts to persuade

100
lawmakers to limit foreign competition in the 1960s and 1970s. Part of the justification for allowing Detroit to be protected by such barriers to entry were many of the same arguments that we hear today for a bailout, including the desirability of protecting Detroit’s ways of doing business and the high wages that came with them. But such wages of course translated into the high labor costs that plague the industry today, as well as its continuing reputation—fair or not—for producing a lower-quality product. In other words, by succeeding in its efforts to limit foreign imports, Detroit not only preserved its high-cost wage structure but effectively guaranteed its own failure to respond effectively to product innovations by its foreign competitors. After all, why change unless you’re forced to? Although the difference between Detroit’s and other carmakers’ US labor costs has been exaggerated—the oft-cited $70 an hour versus $45 an hour mistakenly includes retiree pensions as a wage rather than a fixed cost—the reality is something like $55 an hour versus $45 an hour, or a 20% difference, which isn’t small potatoes. So, with industry excess capacity and Detroit’s inefficiencies as the problem that should be addressed by any intervention—bankruptcy or bailout—the question I’d like to focus on is: What can bankruptcy do to fix the problem? In the case of the automotive industry, bankruptcy— Chapter 11 in particular—does several things extraordinarily well. But it also faces a couple of serious hurdles. Let’s start with how bankruptcy can “help” Detroit. First, bankruptcy law allows the rejection of what lawyers call “executory contracts”—things such as leases, franchise agreements, supply contracts, and labor contracts. That ability would allow Detroit to convert many obligations to franchisees that are imposed by state law into unsecured claims against the company. To give you some idea of the cost of eliminating those franchise agreements outside of bankruptcy, when GM shut down Oldsmobile it reportedly paid as much as $2 billion to Olds dealers pursuant to these state laws. So that’s Plus 1 for bankruptcy. Bankruptcy would also probably allow the industry to turn its unfunded pension obligations to retirees into unsecured claims. Unlike current wages, which represent marginal costs, pension obligations to retired workers are fixed costs that have contributed to one or more of Detroit’s manufacturers being insolvent. Bankruptcy’s ability to deal with accrued pension obligations is Plus 2 for bankruptcy. Now, it’s true that the net effect would be to shift those liabilities to the Pension Benefit Guaranty Corporation, and thus to us the taxpayers—and so the end result would be a government subsidy no matter what Congress does. But as I will suggest later, shifting these kinds of one-time “social costs” from the private sector to the government is a better use of subsidies than propping up businesses that need to shrink to survive. By removing the burden of their pension costs, we can get a

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

much clearer picture of what it will take to turn them into viable standalone enterprises. Bankruptcy will also allow a manufacturer to reject its current labor contracts, although the union might—and probably would—strike. Still, over all, a Plus 3 for bankruptcy. A fourth and final benefit of bankruptcy is that someone other than current shareholders and their representatives will be deciding on the appropriate size of these companies going forward. I think this is an important benefit that hasn’t received much attention. Once a company is insolvent, its management—put in place by the equity interests that are now under water—are effectively playing with other people’s money. Since the equity interests are already under water, they cannot be made any worse off, and so they have a natural tendency both to take greater risks and to drag out any “day of reckoning” in which they will be firmly shut out with nothing. Chapter 11 will transfer that equity ownership to new people, whose money—or financial recovery—will be at risk, and who are thus much more likely to make the best decisions about what to do with the assets going forward. Under Chapter 11, the current management could remain in place; but the decision to keep them there will be in the hands of the new owners—that is to say, the existing creditors whose interests are converted into equity in any reorganized company, as well as the investors that agree to provide funding for the new, slimmed-down companies. But having discussed the potential benefits of bankruptcy in this setting, what are its limitations—what do we need to worry about? The biggest question mark for bankruptcy has to do with whether Chapter 11 is a self-fulfilling prophecy in the sense that no one will buy cars from a GM or Ford or Chrysler in bankruptcy. Most of the time when we buy something, we pay little or no attention to the fact that the selling company is in Chapter 11. We don’t stop flying on United because it is reorganizing. We don’t stop shopping at Bloomingdale’s because it is reorganizing. (In fact, Bloomingdale’s reportedly achieved new levels of profitability and efficiency when operating in Chapter 11 under Allen Questrom in the ’90s.) But that’s because we care only about the immediate “thing” we are purchasing. For the most part, if the company ceases to exist after we buy or fly, we don’t care. But that’s not true for cars. We care about the warranty. It isn’t whether we’ll get parts or service—I have little doubt that businesses will spring up to provide that stuff. The question is whether we will get those parts and services “for free”—as our original deal provided—for a period of, say, five years. This right—the warranty—has a certain economic value to the buyer, one that, just to put a number on it, might be estimated at around $1,000. The problem here is that if you buy a car from GM after it files for Chapter 11, your

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

101

warranty claim, while having “administrative expense” priority in GM’s Chapter 11, would be only an unsecured claim in any subsequent liquidation of GM. So unless you are confident that GM will “make it” for the five years for which your warranty is good, you won’t value the warranty at its full $1,000. Someone needs to figure out how to deal with this problem. Government guarantees have been held up as a “solution,” but that has a major moral hazard problem—that is to say, if the government guarantees warranties, GM has an incentive to build lousy cars. Another possibility, which to me is more palatable, might be to raise the priority of the warranty claims above those of unsecured creditors in any subsequent liquidation. This solution is likely to be better because it would entrust the question of GM’s reorganizing—and optimal size—to those people whose money would be on the line in the Chapter 11 proceeding. But even if this issue is solved, the problem of warranties for people who bought GM cars before bankruptcy needs to be addressed as well. Those warranty claims would be unsecured claims in Chapter 11. The outcry over that would almost certainly require GM to “assume” those claims as an expense of Chapter 11 as well. Concerns have also been voiced about auto parts being made available—though I tend to think this problem is relatively minor since suppliers will continue, or will spring up, to provide the parts. At any rate, these are serious issues that require careful thought and responses—indeed, the kind of response that GM (and others) should have been working on in terms of a “prepackaged” bankruptcy instead of putting all their eggs in the bailout basket. (And, by the way, the statements made by GM’s management and board that they “never considered” bankruptcy as an option make sense only in one scenario—a world where Chapter 11 would spell the end of the current equity owners’ interests and where the political branch appeared to hold out the only hope of postponing, if not avoiding, any such day of reckoning.) And if I’m right about the overcapacity problem, Chapter 11 has a lot going for it, and perhaps a lot more than a government bailout. This isn’t an exercise of imagining a perfect world; it is an exercise of comparing bankruptcy to alternatives and, specifically, to a bailout. If nothing else, bankruptcy—by the “self-selective” nature of the companies that will be using it—is much more likely to focus the solution to the excess capacity problem on that part of the industry the excess capacity should come out of—namely, the less efficient producers that are more likely to become insolvent (in part, because such companies tend to find it more expensive to raise new equity). A bailout, on the other hand, which is far more likely to tolerate (or ignore) the excess capacity problem—because taking it seriously requires one to talk about and focus on shutting plants and putting people out of

work—is likely not only to extend the problem, but to make it far worse and even intractable. Even with conditions put around them, bailouts will continue the existence of those companies within an industry that are least deserving of continuation on almost any scale. If you think I’m exaggerating, consider that many of today’s bailout proponents view the proper role of government as returning the industry to its “normal” production of something like 16 million cars a year. This is a clear prescription for an industry that will face “permanent” overcapacity and a predictable series of future crises—and perhaps permanent government support. Of course, bankruptcy can’t do it all. There is no denying the seriousness of the dislocations and hardship that will be produced—not so much because of bankruptcy but because of the underlying need to pull capacity out of the system, one way or another. Dealing with such dislocations seems to me a useful role for governments—and one that isn’t talked about enough. The government, in my view, would be far better off figuring out a good way of providing relief to those harmed by the transition than propping up companies in industries with excess capacity. Doing so will only make the temporary support permanent. So, my suggestion is to let bankruptcy work, and deal with the issues of overcapacity through a thoughtful government response. This way, we avoid sliding into a “solution” that either ignores the underlying issue of overcapacity or responds to it by spreading the solution around and dragging down all manufacturers. And even if you are unpersuaded by my proposal, let me leave you with one last point: one can’t understand bailouts without understanding bankruptcy. Bankruptcy is an incredibly important and useful tool, one that plays an essential function in a healthy free-market economy—and I think we all understand that such an economy is the underlying source of our collective wealth. Even though it operates company-by-company, bankruptcy can be used to pull excess capacity out of entire industries. It has accomplished as much with the airlines and steel industries. We hardly give it a second thought any more when it is used to take out a Linens ‘n Things—because less efficient than Bed, Bath and Beyond—or a Circuit City—because less efficient than Best Buy. Of course, there has always been a lot of mystique surrounding automobiles—and “what’s good for General Motors is good for the US” But I wonder if the trend toward bailouts—and I do see it as a trend, not just a once-in-a lifetime response—is the reflection not only of politicians’ perceived demand for immediate government “action,” but also of the public’s and policymakers’ failure to understand the positive role of bankruptcy. Bankruptcy may not always produce the right result, but it most certainly cannot if it is not understood—and therefore not given the chance.

102
Thank you.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Mechanic” would be more apt. When I was going to graduate school here in Rochester in the late ’60s, I used some of my spare time to modify a 1963 Chevrolet Impala for drag racing II. The Case for Bailouts on the street. One of my crowning achievements was teaching my wife to beat all the high school kids in that car. Zupan: Thanks, Tom. One of the great pleasures of my I’ve been thinking of this evening in terms of three words: job is getting to see where a degree from the Simon School bailout, bankruptcy, or bust. When I say that, I’m thinking ends up taking people. not in terms of the Detroit Our next speaker, car companies, but rather Charlie Hughes, who’s in terms of our nation. If we are determined to push the social an alum from our class We are a mess. It’s not of 1970, is arguably one agendas of energy independence and just the banking industry, of the foremost the housing industry, the climate control, let’s make sure we do it branding experts in the car industry; it’s the entire automobile industry. In with street smarts and guts. Let’s raise country. We are at a a career that has crossroads. What kind of the federal gas tax. Let’s have one set of included stints as the a future do we want to CEO of both Mazda regulations for emissions and fuel economy have? And, yes, I North America and understand that services nationwide. (How shortsighted and Land Rover North are playing a growing role America, Charlie has in our economy relative arrogant is it for people in each state to managed or represented to manufacturing. But are 11 different brands, demand their own emissions standards?) we going to continue to domestic as well as be a nation of makers and And let’s rush—and I do mean rush—to international. While builders, or will we end running Land Rover, for harmonize those standards with Europe. up a nation of money example, he introduced changers? Think of the powerful platform we could their sport utility line— Like you, I have high built it from scratch in hopes for President achieve if we could get agreement on the late ’80s during a Obama. Yet one can’t period of a year and a standards for what’s basically 70% of the help but wonder if he will half, developing a be pragmatic and tough global car market worldwide—and we supplier and distributor enough. We are in network, and eventually could then take that agreement to India uncharted waters, and the growing sales of that stimulus package—at and China where the real pollution is line to 22,000 per year. least what I’ve seen of Charlie has also occurring and get them on board. it—is a troubling start. recently co-authored a But let’s look at how we -Charlie Hughes book called Branding got here. Iron, and appeared on We have a failed national news networks, Presidency behind us, a Congress with approval ratings that including Bloomberg, to discuss the auto industry bailout— would shame a child molester, a financial crisis born of and, as you might have guessed, he has views on the subject slipshod government oversight, and a widespread ethical that are going to differ from Tom Jackson’s. So, having heard meltdown in our financial industry. We have both states and from one of the world’s foremost authorities on bankruptcy, a federal government that are dominated by special interests. let’s now hear from someone who has spent his most of his I don’t know how many of you watched the Congressional career in the auto industry—someone who can share his hearings where the car companies were taken in hand and firsthand knowledge of not only the industry’s weaknesses taught a few lessons—some of which they deserved. But and vulnerabilities, but also its strengths and Nancy Pelosi, our Speaker of the House, couldn’t restrain accomplishments. herself from using that occasion to push her green agenda, Charlie Hughes: Thanks, Mark, and good evening. My even if it means sinking our domestic car industry. role tonight is to play “Joe the Plumber”—or maybe “Joe the As a nation, we are behaving like fourth-generation heirs;

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

103

we don’t understand how the business that made us wealthy really works. We apparently don’t understand how today’s world operates. On too many issues, we are appallingly ignorant. We are mad at Detroit. During the recent hearings in Washington, six out of ten constituents told their Congressmen to let them die. But these people are clearly unaware of some important realities. The US car companies are far better than you think—though admittedly not as good as they need to be. Let me cite a few facts to make my point: What company runs the most efficient plant in North America? The answer is the Chrysler Jeep plant in Toledo. I might also add that Chrysler, which is the company that’s in the most trouble of the Big Three, is viewed by industry experts as the equal of Toyota in running the most efficient plants throughout North America. What line of cars had the best JD Power rating for initial quality in mid-size cars, which is the largest and most competitive segment? What company had the most cars with IIHS highest rating for crash safety? The answer is Ford, with 16 cars. Number two was Honda, with 13 cars. Who builds a large SUV hybrid that gets better mileage than the Toyota Camry? The answer is General Motors, and the car is the Cadillac Escalade Hybrid. You can’t get much bigger than that—and the car gets 20 miles a gallon around town. And, finally, what is the real difference in pay for factory workers between Toyota and Ford? It’s $9 an hour if you do the calculation the conventional way. But if you factor in the typical bonus the Toyota workers have gotten during the good years—though not this year—the difference is less than $4 an hour. But here’s the irony I see in what’s going on today. What got GM in trouble were its hubris and quick-fix mentality. As our government tosses around trillion dollar fixes, do the words hubris and quick fix come to mind? Starting with the credit crisis, to the Wagner Labor Act, CAFÉ, and transplant factory tax subsidies, our government has played no small role in creating the problems of our auto industry in Detroit. So what’s to be done? Here is my short list of suggestions: First, do no harm. We will debate tonight whether bailout or bankruptcy is the better course. But this is not a lab experiment; and if we get it wrong we are in real trouble. Chapter 11 has never been tested on an industry that is so intertwined with our entire economy. Second, treat each of the Detroit car companies according to their degree of distress and specific circumstances. Ford, for example, is in reasonably good shape: they have a good plan, a solid cash base, and they haven’t taken any money

yet. If they end up needing it, we should support them. GM is a different story. It’s got too much debt, too many brands, and they’ve already borrowed money—and it needs to demonstrate its long-term viability to receive more. But with that said, I can’t imagine this country without them. And, finally, there’s the case of Chrysler, which I think we need to help find an international partner. Fiat has volunteered—and we should see whether that marriage can work. I think our government should continue to support Chrysler until we find out. Now to my third prescription: if we are determined to push the social agendas of energy independence and climate control, let’s make sure we do it with street smarts and guts. Let’s raise the federal gas tax. Let’s have one set of regulations for emissions and fuel economy nationwide. (How shortsighted and arrogant is it for people in each state to demand their own emissions standards?) And let’s rush— and I do mean rush—to harmonize those standards with Europe. Think of the powerful platform we could achieve if we could get agreement on standards for what’s basically 70% of the global car market worldwide—and we could then take that agreement to India and China where the real pollution is occurring and get them on board. Fourth, let’s make sure that when we think of our auto industry, we believe in fair trade, not one-way free trade. Since World War II, every economy that we would consider to be an economic powerhouse has cultivated a strong, homebased car industry. Germany, France, Japan, Korea, and now China all view their auto industries as springboards to economic growth. Not just for the jobs, or the exports, but because the foundation of technological development in these countries—and ours as well—is the auto industry. You may be surprised to know this, but during the Congressional hearings in December, Silicon Valley came out in support of Detroit saying that if one or two of the Detroit auto companies were to go out of business, at least two big names in technology would follow into Chapter 11. I’ve worked for eight different car companies, and six were importers—from Germany, Italy, Britain, and Japan—and I have consulted for the Koreans. All those countries fight fiercely for the success of their homegrown car companies, and in ways we don’t fully appreciate. Their car companies are vitally important to them, and they play the game as a team sport. Sad to say, we are a world champion athlete going to seed. We have gambled our money away and are left staring at our gambling debts. We are at a crossroads; do we want to be a nation of builders—or money changers? Thank you.

104 III. The Difference between Automakers and Banks
Zupan: Thanks, Charlie. Now let’s hear from Jim Brickley, who is the Gleason Professor of Business Administration at the Simon School. He’s an accomplished scholar in organizational economics, competitive policy, corporate governance, and compensation policy. He’s also, along with Cliff Smith, one of the co-authors of the leading textbook on organizational architecture. Jim is also a highly regarded teacher on our campus, having been a past recipient of our highest teaching award. He has published extensively on the topics of franchising and vertical organization, is widely regarded as an expert on distribution systems, and has done extensive consulting to law firms and a variety of corporations on topics like organizational design and governance issues as well as franchising and distribution systems. Jim Brickley: Thanks, Mark. Let me start by saying that the auto industry is clearly very important to the US economy. It employs roughly two million people in manufacturing and in sales and service jobs, and it helps to support many other jobs throughout the economy. It is thus an important contributor to our national GDP, and to our R&D effort as well. But the American auto companies also, of course, have problems, and they are problems that unfortunately run deeper than the current economic recession. Given the importance of this industry, we all hope that productive solutions to these problems can be found. The question we are discussing here tonight is whether these problems are best addressed though government bailouts or reorganizations using the Chapter 11 bankruptcy process. But before we get into the case of the auto companies, let’s talk briefly about the problems with US banks and financial institutions. People often ask why the government has been so quick to bail out banks and other troubled financial institutions, while at the same time being resistant to the idea of bailing out the auto industry. Aren’t both industries important to the economy, and weren’t there just as many management blunders in banking as in the auto companies? The answer is that the banking and auto industries have fundamentally different effects on our overall economy. While policy makers might view bankruptcy as a workable option for auto companies, the use of a similar process in the case of large banks—one that would put a freeze on all creditors’ claims—could have far more serious effects on the overall economy. The banking and financial system in an economy is like the circulatory system in a human being; just as people can’t survive if their hearts fail and blood doesn’t get to vital organs, economies can’t function with major disruptions in the flow of credit. Virtually every business of

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

any size in this country depends on financial institutions to finance its operations and investments. Consumers depend on banks to provide a relatively risk-free place to hold their savings—not to mention their mortgage and auto loans, insurance, and other financial services. Because of their importance to both businesses and individual savers—and their role in linking the two groups— the failure of major banks and financial institutions would send shockwaves throughout the economy, leading to widespread lack of confidence in the banking system and even financial panic. The bankruptcy of Lehman Brothers gave policymakers a frightening glimpse of the potential for a large domino effect when a big, well-known financial institution defaults on its agreements. That event, along with the near bankruptcy of AIG, resulted in a literal “run on the banks” that threatened Goldman Sachs, Morgan Stanley, and just about every major financial institution in the US As I already suggested, the financial panic triggered by the failure of leading financial institutions would have restricted the flow of funds to the rest of the economy—even more than it already has—as investors pulled their funds out of the banks, and the banks became increasingly reluctant to lend to consumers, to the business community, and even to one another. Now, to come back to where I started, the auto industry is very important. Failures in the industry will have harmful effects on many people—including people who work for other auto-related companies—and the overall economy. But having said that, allowing a large manufacturing company to file for bankruptcy, even one as large as GM, would not have the devastating system-wide effects that would occur if the government allowed large financial institutions like Chase or Bank of America to default on their obligations. As Tom Jackson was just suggesting, Chapter 11 could well help the auto industry address some of its most pressing problems. But let’s take a closer look at the challenges now facing the auto industry. Wall Street analysts, when discussing the problems of the Big Three auto companies, tend to focus on unions, and on their labor costs and debt. But another critical problem is the inefficiency stemming from their number of brands and models and from their distribution systems, or dealer networks. I think that these issues of corporate strategy and structure are likely to be addressed more effectively through bankruptcy than bailouts. As Tom told us earlier, the Big Three auto companies developed much of their product lines and dealer networks starting back in the 1950s and ’60s, when they dominated the US auto market. It is widely acknowledged that these companies now have far too many brands, models, and dealerships, given their current market shares, which are collectively less than 50% of the US market. The Big Three now market 112 different car and truck models in the US through 15 distinct brands. In contrast, their major

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

105

competitors—the top three Japanese companies—offer only own dealerships in many states and prohibit direct marketing about half the choices, with 58 models and seven brands. to consumers through other media such as the Internet. In GM by itself has eight brands and 70 models, and thus more fact, a number of attempts by the Big Three to introduce new brands and models than the Japanese companies combined. marketing channels have been blocked by dealer-initiated And as Tom also told you, the Big Three also have far too lawsuits or regulatory actions. many dealers. GM currently has some 6,700 dealers that I have studied the effects of franchise and dealer protection operate 14,000 laws across a broad range franchises for its eight of industries. My research While policy makers might view brands. Its closest indicates that such laws competitor, Toyota, has lead to less efficient bankruptcy as a workable option for auto only 1,200 dealers with distribution systems and just 1,600 franchises, the destruction of companies, the use of a similar process in and thus nearly 90% corporate values. the case of large banks—one that would fewer. Now, the auto Consistent with these companies have all findings, a study by the put a freeze on all creditors’ claims—could recognized the need to FTC has concluded that have far more serious effects on the overall reduce their brands, state laws preventing auto models, and manufacturers from economy. The banking and financial dealerships. But, as owning their own system in an economy is like the Tom said earlier, this is dealerships has cost US going to be difficult, consumers billions of circulatory system in a human being; just and very expensive, to dollars a year in the form accomplish outside of of higher auto prices. as people can’t survive if their hearts fail bankruptcy. Auto How do we address this and blood doesn’t get to vital organs, dealers are a wellproblem? It is unrealistic organized and powerful to expect 50 state economies can’t function with major political force in their legislatures to reform disruptions in the flow of credit. local communities. these laws in the face of Over time, they have opposition from the local -Jim Brickley secured protective car dealers. My legislation in almost all suggestion is that the US states that makes it very costly for the auto companies to federal government consider national legislation that would discontinue brands or close or combine dealerships. For supersede state laws and grant the auto companies more example, it reportedly cost GM over $1 billion to settle flexibility to design efficient distribution systems. disputes with dealers when they stopped making Oldsmobiles And let me leave you with one final thought: Inefficient a few years ago. franchise laws are but one example of how political Now, as Tom also said, in the case of bankruptcy, all of the considerations often trump economics in legislative or company’s dealer contracts become subject to cancellation regulatory solutions. Restructuring and consolidating the and reworking. As a result, the auto companies would have automobile industry will require many tough choices—and much more flexibility to reconfigure their brands and there will be losers as well as winners. Bankruptcy dealership systems in a quick and efficient way. Of course, proceedings are much more likely to focus on economic some restructuring is going on as we speak. The number of considerations in making these tough choices than a bailout American car dealerships has been falling almost daily as process that involves politicians and politically-motivated these businesses fail. But relying on local business failures “car czars.” In the long run, the industry will be much stronger to reduce the number of dealers—thanks to all their legal if we allow economics rather than politics to drive the recourse to and demands on the Big Three for life support— outcome. is a very protracted and costly way of addressing the basic problem. What is needed instead are systematic and IV. A Bankruptcy Practitioner’s coordinated changes in these companies’ product lines and Perspective on Chapter 11 dealership systems. State laws not only make it expensive to alter dealership Zupan: Thanks, Jim. Now let’s hear from Joel Tabas, a contracts, they also prevent manufacturers from owning their

106
Simon alum from the class of 1980 and the managing partner of Tabas, Freedman, Soloff and Miller, a Miami-based law firm that specializes in reorganization and bankruptcy. As part of his practice, Joel has dealt with Ponzi schemes, real estate reorganizations, and healthcare workouts and bankruptcies. He has found himself operating airlines, retailers, and restaurants—and participated on creditors’ committees in complex reorganization cases involving such names as Planet Hollywood, Brothers Gourmet Coffee, and The Discovery Zone. Joel has graciously agreed to join us tonight in the midst of what are pretty busy times for his business. Joel Tabas: Thank you, Mark. And let me start by saying that it’s a great honor to be taking part in this discussion. Cliff Smith was my finance professor when I was in the MBA program in the late ’70s. Tom Jackson’s classic article on reform of the US bankruptcy system was required reading when I went to law school. And, like President Jackson, by the way, I too was struck by the stage backdrop behind us. In Miami, we’re dealing with an incredibly distressed real estate market—and this Depression-era stage set looks very familiar, makes me feel right at home. As Tom started out by saying, when evaluating any kind of distressed corporate situation and the range of possible solutions, it’s very important to understand what can be accomplished in Chapter 11. Most people have an instinctive aversion to the word “bankruptcy”; they think of it as a death sentence for companies. There’s good reason for this: History tell us that about 90% of all companies that enter into a Chapter 11 proceeding for reorganization do not emerge as going concerns; instead they are sold to outside investors or end up liquidating in a Chapter 7 or similar proceeding. That’s the bad news about bankruptcy—but there is some good news here as well. After all, 10% of the companies that file Chapter 11 do emerge as independent viable enterprises. One of the main distinguishing features of such successful reorganizations is planning and preparation. The companies that come out of Chapter 11 tend to be those that carefully explore the potential benefits of a bankruptcy before going into it—they don’t just passively react. I would argue that the 90% failure rate is in large part the result of inadequate pre-bankruptcy planning, of the tendency of many companies to wait until it is too late to rehabilitate the business. In this sense, the high rate of failure is not really attributable to the Chapter 11 process itself, but rather to the fact that so many patients arrive in bankruptcy almost “DOA”—in which case they tend to get put on artificial life support for a short period before going into liquidation. I have represented both debtors and creditors in the reorganization process. If you’re helping a debtor negotiate with creditors in a distressed situation, you have to understand—and to make sure that the creditors understand—

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

the likely outcome of a bankruptcy proceeding. Just the prospect of Chapter 11, with its “automatic stay” provision and the potential rejection of “executor” contracts, is very helpful in getting concessions from lenders and other major claimholders. As Tom mentioned earlier, such claims tend to be reduced significantly in Chapter 11, and are often converted to equity interests. In the case of the auto industry, as Tom also said, Chapter 11 could be very effective in getting concessions from not just creditors, but from the franchisees or dealers and the unions as well. Another important advantage of bankruptcy—one that could be especially helpful in the case of the US auto makers—is its role in centralizing and coordinating the reorganization process. When dealing with large numbers of creditors that are dispersed around the country and have the option of seeking different venues and courts, a private, outof-court workout process would be a nightmare—the legal fees and expenses would be astronomical. The beauty of the bankruptcy proceeding is that the debtor files a bankruptcy in one particular forum—and all of the disputes are focused for the most part in that forum. So, instead of General Motors facing litigation throughout the country on franchise disputes, in Chapter 11 it would be handling the litigation involving all of those franchisees in the one forum where the bankruptcy is filed. So, that is an extraordinary benefit that bankruptcy brings to a situation like this. It focuses the efforts and avoids the potential for inconsistent consequences. Avoiding this possibility is likely to mean some cost savings for the franchisees. One of the things that happens early on in many big bankruptcy cases is the formation of “committees” of creditors or other claimants with similar situations. That was how I got involved in the Planet Hollywood case that Dean Zupan mentioned. My client was a creditor, and we were invited to become part of the committee of unsecured creditors. The role of such committees in such cases is to act pretty much as the boards of directors of public companies are supposed to act. They have fiduciary obligations to their constituents—namely, all the similarly situated, unsecured creditors—that resemble the obligations of corporate directors to the company’s shareholders. In other words, they are not supposed to be using the platform for personal gain, or to benefit their clients at the expense of other claimants. They’re supposed to be trying to maximize the recovery of all the creditors. They have the right, and are given the resources, to hire professionals—accountants and other financial types as well as accountants—to help them make the managerial decisions that have to be made. Now, it’s true that maximizing the recovery of creditors is not necessarily the same thing as maximizing the health and future viability of the entire enterprise; there is some potential for conflict here, and for a premature liquidation of the

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

107

business. But even so, I would argue that the formation and functioning of such creditor committees is a critical feature of the bankruptcy process—one that does not exist at all outside of Chapter 11. I’m a believer in having people with the economic interests involved in the key decisions about the future of the business, especially if a big portion of their claims is going to be converted into equity. I’m convinced that such people are far better able to help fashion how the company will go forward than the typical regulator, who is beholden to all the various constituents of the enterprise. Moreover, in determining the company’s future, debtors are greatly aided by the automatic stay provision I mentioned earlier. By putting a halt to all the disputes and lawsuits, the automatic stay provides a breathing spell that enables all of the constituents—all of the parties to the process—to make important decisions: Can the company be reorganized and restructured in a way that will allow it to succeed? Or is it worth more dead than alive and a candidate for liquidation? Still another advantage of Chapter 11—and this one is very timely—is its ability to restrain excessive or unearned executive pay. Early on in bankruptcy proceedings, all of the top executives basically have to submit their compensation packages for approval by the court and vetting by the creditors. So, this brings all compensation arrangements out into the daylight. Earlier in this decade, we used to see people filing for compensation packages with golden parachutes. But that practice has now been largely ended by the courts. Now, let’s come back to this issue of franchises that everybody has identified as a big problem for the auto makers. As has already been noted, most states have passed laws that make it very difficult and expensive for the manufacturers to shut down their franchisees. We’ve been involved with a few Ford franchisees in the Miami area that have recently filed bankruptcy and shut down. I can tell you that they’re all struggling—and it’s going to be a widespread situation if the economy stays the way it is now, and there are likely to be significant damages to the manufacturers associated shutting down franchises. But, as Tom pointed out earlier, if a manufacturer files bankruptcy, it could deal with its franchisees’ claims in one forum—and everyone could be treated the same. There could even be a committee for the franchisees so that they too could have an economic voice about the firm’s future. In fact it’s more than likely that, at the end of any successful reorganization process, the franchisees will become significant equity holders in the auto makers—and if this happens, they’ll actually have a stake in the health of the underlying business. The same comment also holds, by the way, for the unions: Only after becoming major equity holders are they likely to act in ways designed to preserve the goingconcern value of the enterprise. Another valuable aspect of bankruptcy is its ability to

increase disclosure and transparency. As already mentioned, executive compensation is typically submitted to courts for approval. But professional fees also have to be submitted on a periodic basis for approval with the courts as well. While I’ve seen studies suggesting that the costs of a bankruptcy proceeding in terms of professional fees would be much higher than in a private workout, I think that there are certain aspects of private workouts that have not been incorporated into the analysis. My guess is that, especially in a case like GM or Chrysler, there would be significant cost savings not only on the debtor’s side, but for the creditors as well— because of their coordinated representation by the committees I mentioned. Another aspect of a bankruptcy proceeding that will facilitate information flow is the provision—specifically rule 2004—that gives any party “in interest”—be it a creditor, an equity holder, or the government in its role as The United States Trustee—the right to obtain financial information from the debtors, including information about their plans to restructure and rehabilitate the debtors. Bankruptcy effectively gives such parties the right to take depositions from the debtor—a right that would not be available outside of a bankruptcy in an out-of-court workout or a bailout situation. I would also argue that, thanks to years of litigation in high profile cases involving many of the complex issues now facing our auto makers, there is a very well established set of case law and dynamics and parameters that are used by the courts in arriving at the judgments they make about whether to reorganize companies or let them fail. In bankruptcy courts, you will be dealing with jurists who handle reorganizations and feasibility determinations on a regular basis. So you have a very well-developed area of the law that will not be available in an out-of-court situation—where you’re likely to see a race by all creditors to a state courthouse instead. And let me come back to the point about the creditors committees that I made earlier. The committees and other constituents with financial interests are going to determine through a process of negotiation the important features of the company that emerges from a bankruptcy—what products it will continue to make and sell, and how the company will be financed. My own experience suggests that Chapter 11 can provide a cost-effective process for restructuring the companies that are deemed by the court to be worth saving. For one thing, it provides a very effective way of eliminating obstacles to private workouts. One obstacle is holdouts among creditors to a negotiated solution—and the Chapter 11 can be used to “cram down” such a solution. Another obstacle is entrenched managers or owners. It’s always tough for someone to admit they’ve taken the wrong tack—that their management strategies haven’t worked and they should not be given another chance. While the process can sometimes

108

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

get a bit heated and hostile, I’ve found that the adversarial value of their companies, they could be facing a D&O suit. process that leads up to confirmation of a plan generally tends In sum, the auto manufacturers need to carefully consider to yield a good outcome—one that typically reflects the the possibility that Chapter 11 is the low-cost way of working concerns and interests of all the major constituencies. through their problems and preserving their companies as Before I close, let me mention one other important viable—though likely much smaller—going concerns. advantage of Chapter 11—a feature designed to help debtors Bankruptcy, for all its flaws and bad press, may have a lot to raise new capital. offer under these This feature is likely circumstances. Thank I’m a believer in having people with the to be most valuable, you. of course, in cases economic interests involved in the key where the capital V. The (Longdecisions about the future of the business, markets are otherwise Run) Costs of unwilling to provide especially if a big portion of their claims is Bailouts new capital. That appears to be the case going to be converted into equity. I’m for the US auto Zupan: Thanks, convinced that such people are far better makers, which is why Joel. Batting cleanup the government is tonight on our panel is able to help fashion how the company will contemplating an Cliff Smith, who is the go forward than the typical regulator, who expansion of the Louise and Henry bailout money already Epstein Professor of is beholden to all the various constituents of provided. The capital Finance at the Simon markets are not going School. Cliff is, first of the enterprise. to be giving money to all, an accomplished -Joel Tabas the Big Three— scholar. He has long they’re unable to raise been one of the main equity or debt—and so they’re going to the government. But editors of the Journal of Financial Economics, which is if one of the auto makers were instead to file for Chapter 11, headquartered at the Simon School and, along with the it could go to the court and say, “To raise new capital, I need Journal of Finance, is one of the top two journals in the field. to be able to issue super-priority debt financing—debt that is He’s published 16 books and some 90 articles. He won a going to come ahead of the other secured creditors in my major prize a year ago for his impact on the field of insurance. capital structure.” And to the extent they were successful in He is also a very dedicated and talented teacher. In a career raising private capital on those terms—which is hard to predict at the Simon School that is now in his 35th year, Cliff has under the current circumstances—the further bailout of the received our full-time MBA Teaching Award ten times and industry could effectively be financed by private investors. our Executive MBA Teaching Award an amazing 19 times! Cliff Smith: Thanks, Mark. It’s good to be here. As a If that fails, the other option would be to have the government long-time subscriber, I appreciate what GeVa has done for provide the super-priority financing. So, there are a number of features of the US bankruptcy the local arts community. I want to thank them for letting us code that, in my view, could be used to help US auto makers use this wonderful facility. It’s become an old saying that people who do not study to work their way out from under their current burdens. And, as President Jackson suggested, they should be weighing all history are doomed to repeat mistakes that have already been their options very carefully. One reason they should be made. I thought it might be useful to look at precedents to weighing those options—and its one that I’ve haven’t heard our current circumstances, and to try and glean lessons from mentioned tonight—is that if the officers and directors of the past. When you talk about bailouts in the auto industry, people these companies do not consider bankruptcy, and the companies end up in liquidation, the directors could be facing in the US tend to point to Chrysler as an example of a success director and officer suits, which is a fertile area of law right story. They will say, “Chrysler got their act together and now. What those suits are alleging is that is when a company things worked out wonderfully. Let’s just do it again?” Now, enters what is known as “the zone of insolvency,” directors if you say that fast enough, and don’t think about it very hard, have fiduciary duties that are supposed to shift from the it sounds good. But it’s important to remember that Chrysler shareholders to the creditors. If directors have failed to was not the only bailout that we lived through during the consider bankruptcy as a means of preserving the enterprise ’70s and ’80s.

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

109

Remember the US savings and loan industry and what happened to it? In the early ’80s, when interest rates on Treasury notes and bonds got into double digits, executives from many S&Ls went to Congress looking for help. Since most of these S&Ls were holding mainly long-term fixedrate mortgages with rates around 5-8%, they were effectively insolvent. My dad was a banker in Greensboro, Georgia in those days, and he liked to tell people, “You can’t write 8%, 30-year mortgages, fund them with CDs paying 12%, and expect to make it up on volume.” What happened next? Well, Congress effectively changed the bank accounting standards in such a way that the S&Ls could maintain at least the appearance of solvency and continue to stay in business. So, for the next few years, we had lots of “zombie” S&Ls—they were dead, economically speaking, but were still walking around underwriting risky mortgages and investing in risky commercial real estate. It was those transactions that ended up doing most of the damage. The net result of this regulatory “forbearance” was that, despite the best efforts of the Resolution Trust Corporation ten years later, US taxpayers ended up footing a bill that has been estimated at about $130 billion. My point here, then, is that although the S&L bailout is today widely viewed as having been a good thing, what seems to have largely vanished from the collective memory is any sense of the eventual cost of that initial act of forbearance. By failing to deal with the troubled S&Ls effectively in the early ’80s, our government turned what would have likely been relatively modest losses into much larger ones. So, the first lesson from history is that bailouts are a risky business—and not only is the outcome uncertain, but bailouts can have the effect of increasing risk within the system. If you go back and look at accounts in AutoWeek of Chrysler’s post-bailout success story, you will see articles in the late ’70s and early ’80s about Chrysler’s bold, new, innovative models. As a finance professor, when companies use words like “bold,” “new,” and “innovative,” what I hear is “risky,” “risky,” “risky.” And that leads to an interesting problem for regulators—and of course the rest of us as taxpayers. As the political process is unfolding and people are saying, “Well, the cost that we’re forecasting for this bailout is X dollars, and the US auto industry is clearly worth more than that,” I would recommend a fair amount of skepticism because those costs are regularly understated by what can turn into scary amounts. One of the big reasons these cost estimates turn out to be understated is that the behavior of the companies that are bailed out tends to change. They are being given the opportunity, in a sense, to play poker with someone else’s money. If you’re ever invited to a poker game and allowed to play with someone else’s money, I’ve got a piece of advice: increase your bets.

Bringing out risky new products is one way automakers can do it—but there are others. Before Chrysler got its bailout package in the ’70s, product warranties in the industry covered 12 months or 12,000 miles. After Chrysler’s debt was guaranteed by us, the taxpayers, Chrysler management decided to expand Chrysler warranties to five years or 50,000 miles. Now, as things turned out, those bold new products generally were well-received and well-produced. So the resulting warranty claims didn’t eat us out of house and home. But think about this from Chrysler’s perspective. “We’re going to try something that is bold, new, and innovative. If it works, we’re heroes. If it doesn’t work, we’re giving the company to the Treasury.” It is like flipping a coin where heads I win tails you lose. Thus, my second history lesson is that bailouts allow companies to play poker with the taxpayers’ money. That is what both Chrysler and the S&Ls did when the government gave them a second chance—and that is what I would expect US automakers to do this time around. We are going to see lots of outsized bets being funded not by private investors, but by taxpayer dollars—bets that are going to be initiated by corporate managers with little to lose and overseen by government officials with limited expertise, and perhaps even less to lose. My third point is that the forecasted duration of this bailout is something that can easily expand. Think about the history of US agriculture since World War II. During the War, most European wheat fields were turned into battlefields. In response, Roosevelt granted draft deferments to US farmers along with instructions to “crank up production and feed the Allies.” And they did a marvelous job. But what happened after V.E. Day? The swords were turned into plowshares, the European battlefields back into wheat fields, and there was a massive increase in the global supply of agriculture products. The resulting oversupply and plunge in crop prices meant that the US agricultural industry faced hard times. This huge increase in supply and crash in prices put the US at a political crossroads with respect to its agriculture industry. What was to be done? One option was to do nothing. If the government did nothing, agricultural prices would likely have remained low for two years, or maybe three—and US farmers would have had a tough row to hoe. You would have seen many leaving that industry. Who would have been most likely to leave? Well, the people with the most opportunities other places, those with the most flexibility. So you would have seen younger farmers leaving while older farmers stayed. People with college degrees and more opportunities in other industries would be more likely to go. But after a few years, the wrenching adjustments would have been behind us, and we would have been back in normal operation, though with

110

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

far fewer people working in the industry. (And the same, by has provided us with an example. In the 1970s, the U.K. the way, would likely have been true if the government had government engineered a bailout of British Leyland, the not bailed out Chrysler in the early ’80s. Had we made the maker of Austin, Morris, Mini, MG, Rover, and Jaguar. tough choice back then, we would not now be facing the Leyland had a weak balance sheet, contentious labor relations, magnitude of problems Detroit is forced to deal with— and inefficient manufacturing: moreover, it had suffered a because the industry, and the overcapacity problem, would substantial loss in market share. The U.K. government poured likely never have reached their current levels.) about $16.5 billion (in current dollars) into the company The other choice facing US policymakers back then was during the ’70s and ’80s. The bailout ended up lasting longer of course to bail out and costing more than the US agriculture had been forecast—and Here in the US, it’s always been a very large industry. And we all it ultimately failed to know how that one save the company: number of people putting their own intuition turned out. We British Leyland into their business models and strategies, and decided to pay our eventually went out of farmers not to business, with select putting their own capital on the line to back produce. That pieces being sold to happened in the ’40s foreign auto makers. their bets. What you wind up with when you and then in the ’50s— I think we all agree allow that kind of experimentation is a very and then again in ’60s, that we are discussing ’70s, ’80s, and ’90s. an incredibly important large portfolio of options. As any finance We’re still doing it set of problems for the professor will you, a portfolio of options is today. If you believe US auto industry. In that this bailout of the making our policy dramatically more valuable than an option auto industry is choices, we need to something that we’re think carefully about the on a single portfolio. going to do once and long-run consequences -Cliff Smith be done with, perhaps of whatever policy you need to think choices get made— again. The costs of the bailout is likely to turn out to be about whether and how these companies can be made to stand massively understated—and it could well turn into a kind of on their own, and how many of our taxpayer dollars we are perpetual annuity. Thus, the third history lesson is that willing to use to see if we can make it happen. bailouts can persist—sometimes for decades. Bailout advocates in Congress regularly announce, “We’re VI. Bank Bailouts and the Credit Crunch not planning on just handing suitcases full of money to General Motors, Ford and Chrysler. We’re going to put Zupan: Thanks, Cliff. I’ll now invite the other panelists constraints on what they can do. We’re going to put constraints to join us on the stage, and we will take some questions from on how they can pay people. Nancy Pelosi is talking about forcing them to start making “green” cars—and she’s not the audience. Here’s the first one: “Should Lehman Brothers talking about her favorite paint color. To me, this begins to have been forced to go bankrupt?” Tom, can you start us off sound like allowing the government to run the industry. on that one? Jackson: Any time you’re looking at a large financial Unfortunately, the government’s track record in running institution, there are many more linkages with the rest of the businesses is not the best. Think about Fannie Mae and Freddie Mac—not to mention the US Postal Service. Thus economy, and things are much more complicated. my fourth history lesson is that the government is unlikely to Commercial banks can’t use bankruptcy; they need to go through some other regulatory process. In the case of Lehman be especially good at running businesses. We were told earlier that we’ve never had a bankruptcy Brothers, what I’ve been told by people suggests that it’s tied applied to an industry that is as large and important to our in such an important way to the financial infrastructure that I economy as the US auto industry—and that this crisis is just think they probably should have rescued it instead of letting too big to be managed as an experiment. Yet this same logic it go. I think our regulators learned a lesson from that failure. should also rule out a bailout: we have never bailed out an My guess is that they were too quick to believe that this would industry that is this large and important either (unless we count be the last failure and that we could survive it—and when the agricultural industry). But if we look overseas, history they quickly saw there would be huge problems unwinding

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

111

all the claims, we went back to a model of stepping in. And I suspect it was probably the right thing to do under the circumstances. Zupan: Next question: “Banks have taken federal monies yet don’t appear to have increased their lending in a significant way? When do you expect the banks to start lending to other companies?” Cliff, can you take a shot at that one? Smith: Banks are making loans right now to companies with lots of tangible assets and established credit histories. Wall Street will take your debt to public markets if you’ve got a triple A credit rating. But what has happened, and what almost always happens during these kinds of financial dislocations, is that credit spreads have risen dramatically. So the curtailing of access to credit has been most pronounced for businesses with weaker credits. Now the real problem here is all the uncertainty about how long it is going to take before the economy recoveries and, as a result, about how much collateral lies behind the business, and how much debt it is really capable of supporting. Thus, if you’re a start-up company with little in the way of tangible assets and not much of a track record, you’re going to have trouble persuading a commercial bank to make you a loan, or an investment bank to help you raise debt capital. Hughes: I’d like to jump in here, since I think we’re avoiding the biggest issue with the banks—namely, their unwillingness to lend to each other because they don’t trust each others’ balance sheets. I think there are two main ways out of this problem: the Japan model and the Swedish model. The Japanese approach was to accommodate the banks, to allow them to continue to operate and make more loans while cleaning up their balance sheets very gradually. The Swedes said, “We’ve got to clean up the balance sheets right away and we’ll nationalize the banks—take temporary ownership and control of them—to accomplish that.” Sweden came back pretty quickly while Japan was in a recession for over ten years. So, while I think it was good that the government pumped in some cash and kept other institutions from collapsing, I think we’re avoiding the big issue. You can’t have a banking system where institutions can’t trust each others’ balance sheets. It’s like a game of liar’s poker. Tabas: I represent some local banks in Miami, and the amount of new loans—particularly real estate loans—are down as much 90% in some cases. One of my best friends, a well-known appraiser in Miami, is refusing to appraise residential real estate values because the prices on singlefamily homes have plummeted about 40% on average—and condominiums are down 50% or more. Because of this situation, banks are being forced to write down their assets. One local bank recently wrote down its real estate-based assets from about $6 billion to $4 billion—and because of their capital requirements, it’s very hard for them to make new

loans. And this is a kind of a self-perpetuating problem in the sense that the markdowns and capital requirements seem to be compounding the difficulties, creating a downward spiral. Our real estate market clearly overshot on the way up ’04 and ’05. Now I think it has overshot on the way down. But market participants tend to overreact—and in some cases perhaps bank regulators, too. The result is that right now people in Miami are not able to borrow money for real estate from banks. Jackson: I think that cleaning up the banks’ balance sheets is a necessary but not a sufficient step in dealing with our present problems. Even if you clean up their balance sheets, the banks have to make sure that the people who are trying to borrow the money are capable of repaying the loans—because if they’re not, then we’ve only added to the existing troubles. Things look awfully murky out there. As Joel said, they’re having a tough time getting people to step up and make appraisals on the properties. So it hasn’t been a big surprise to me that the bailouts have failed to produce an immediate increase in bank lending. That’s going to take time. So, this is a very complicated and multi-faceted problem— and cleaning up the balance sheets is, as I said, a necessary part of the process of getting credit flowing again. But other things have to happen too. Smith: Well, in thinking about this question, I think it’s important to start with an understanding of what banks have a comparative advantage in doing. If you are a fairly large business with a good track record of producing earnings and cash flow, your first choice will typically be to go to Wall Street and have them package your debt as a public issue. Banks, on the other hand, tend to finance smaller companies that, even if publicly traded, have substantially less information produced about them. For regional and community banks in particular, it’s these kinds of smaller, more opaque enterprises that have always been their bread and butter. Another way of saying this is that banks acquire a lot of what’s known as “specific knowledge” about their corporate clients—the kind that is not easily transferred from one lender to another. And that suggests that this idea of cleaning up bank balance sheets so they can start trusting each other has some important limits. Financial institutions—and particularly smaller banks—are by their nature somewhat opaque institutions that hold many assets that are difficult for outsiders to value. That’s why I’m frankly skeptical about the government’s plan to buy troubled assets. In cases where insiders have an advantage over outsiders in valuing bank-originated assets— and as I say, that’s especially been true of the smaller regional banks—I think it makes more sense to recapitalize those banks

112
with infusions of equity than to buy individual assets.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

VII. Global Competition and Jobs
Zupan: Next question: “Does reduction of capacity in US industries imply that American workers are supposed to relocate to foreign countries to work?” Hughes: I don’t think many people are aware of this, but before the credit crisis began to set in, both General Motors and Ford went through massive restructurings that took out almost half of their production capacity. They were forced to buy out thousands of workers at $140,000 a shot because of contracts with the UAW. I think they were pretty smart and decisive in doing that. Had they not done that, the companies would be in much more trouble than they are now. But, if I can be a little patriotic, I find it bizarre to say, “We’ve got three million units worth of excess capacity; let’s take it out of the US producers.” If we were to do this, we would be the only country in the world to take that approach. Jackson: I don’t think this question of domestic versus foreign production is nearly as simple as you make it out. Some of G.M.’s most efficient operations are manufacturers in other countries, such as Holden in Australia. Obviously a lot of the foreign companies have now built US plants that employ US workers. So distinguishing between US versus foreign production is not straightforward. The real question here is whether we are going to continue to have the capacity to produce 16 million cars when we don’t need it. I think that using taxpayer dollars to subsidize that overproduction is a terrible idea, and that we have to figure out some way to take capacity out of the system. I don’t believe that the jobs lost by Detroit are necessarily going overseas—they’re just going to be shifted to more efficient producers here in the US, most of them, I would guess, in the service sector rather than manufacturing. Brickley: To expand on Tom’s point, something like 6070% of the Toyotas that are sold in this country are also assembled in this country. Since there are lots of American investors who own shares in Toyota, it’s no longer even clear what it means to be a foreign company. As Tom said, Honda, Toyota, and the other Japanese companies employ lots of US workers here in the US And since GM now imports parts that are made all over the world, I’m not sure it even makes sense to talk about a US-produced car. Hughes: That’s all true. But we still import a huge number of cars. Again, I find it very odd that we would be having any conversation where people say, “We should be supporting cars that are built somewhere else over cars that are built here.” I’m not talking about putting tariffs on imports. My point is that, in the past few years, the Big Three have already made huge efforts to take out excess capacity; and although we may well have three million units of excess capacity in

the US, not all of that capacity is sitting in the United States. So if we are talking about supporting our US producers— and there now seems to be a national and political will to do that—then it seems to me that we should be willing to provide the capital needed to rehabilitate them. This way, and given some time, they can become the efficient producers that we want.

VIII. The Role of Greed
Zupan: Another question: “It seems that all the problems we’re currently dealing with can ultimately be traced to greed. When will we learn how to deal with this? Smith: I’ll tell you when. When the physicists figure out how to repeal the law of gravity, the economists will be right behind them repealing the law of demand and abolishing greed. All you can do is to recognize greed—or what we economists call “self interest”—and then try to set up our institutions so that self interest becomes mainly a force for good. That’s a matter of getting the incentives right inside organizations—something that I believe is incredibly important. Brickley: Greed is a pretty loaded term, I agree. When you hear it, it’s important to keep in mind what another guy named Smith—not an economics professor, but a Professor of Moral Philosophy—told us over 200 years ago. Adam Smith’s message was that self interest plays a very important role in creating lots of the good things that we all take for granted. It drives innovation, all the new products and services that are the real source of prosperity. Now, one question we are asking is whether people are any more self-interested now than they were, say, in the caveman era. But, as the environment becomes more complex, there are new and sometimes destructive ways to pursue self-interest—things like the off balance sheet partnerships that brought down Enron and some of the more speculative uses of derivatives by companies that we’ve seen in recent years. You couldn’t have done these things 20 years ago because the financial instruments just weren’t available. Smith: And to add to what Jim’s just said, I think it make sense to view our entire financial system as engaged in a kind of Darwinian process of trial and error. We keep trying different things, we make mistakes—and then we learn from our mistakes and make adjustments. One of the strengths of capitalism is that it tends to prevent people from persisting in error, making the same mistakes over and over again. We will no doubt make mistakes in the future. We will continue to have boom and bust cycles of the kind we’re now going through. Now, one important lesson underscored by recent experience is that problems are going to arise whenever individuals and companies are granted a lot of “free

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

113

options”—that is, whenever they can acquire assets or do deals without putting any of their own capital at risk. We saw that kind of behavior by people getting mortgages—and also by banks that originated the mortgages with the idea of securitizing and selling off as much as they could. That’s a clear prescription for too many mortgages and too many securitized deals. Jackson: True, but it wasn’t just opportunistic or greedy lenders and homeowners at work here; government policy clearly had a hand in producing the housing and mortgage bubble. It was government policy, pushed by Republicans and Democrats alike, that effectively encouraged lenders to drop standard downpayment requirements and come up with creative financing—all with the idea of realizing a bipartisan government notion that everybody should own a home. When people and institutions respond in predictable ways to those policy initiatives, I’m not sure we learn much from identifying the source of such behavior as “greed.” Hughes: That’s all true. On the other hand, I tend to think that behavior crosses the line from financial incentive to greed when you have a financial community that’s willing to sink a global economy. When you look at how the banks bundled these mortgages into securities—bringing in the best and the brightest from places like MIT to do the statistical analysis to put these packages together, and ending up with leverage ratios of 40 to one—you have to ask how that all came about. I don’t know many bankers that are comfortable with the idea of operating with that kind of leverage. I think that at that point you can say that the driving force was greed.

IX. Solving the Dealer Problem
Zupan: Ok, we have time for one more question, and here it is: “Instead of relying on bankruptcy, wouldn’t it be better to deal directly with the adverse effects of franchising and dealer protection laws just by changing state and Federal law?” Jim, you’re the expert on franchising, why don’t you take this one? Brickley: Well, I see two different issues here. One has to do with the states, almost all of which have these laws that make it difficult for the auto companies to operate efficiently. Now, the dealers have to worry about protecting their investments—and I think much if not all of this protection could be provided by private contracts with the manufacturer. I think it’s important for the government to back these contracts. But the way things are now, the automakers are prohibited by state laws from owning dealerships—and they are also prevented from selling cars directly to consumers over the Internet. I think both of these prohibitions are sources of inefficiency that increase the cost of automobiles—and, in my view, they should be overridden by federal legislation. The second issue raised by the dealers—by, say, General

Motors’ need to deal with 14,000 franchise contracts—is one that I don’t think can be addressed effectively by legislative action. To have a chance of becoming a competitive producer, GM must renegotiate these contracts. But, as Tom said before, this renegotiation is going to be very difficult outside of bankruptcy. If they try to accomplish this outside Chapter 11, people are going to be fighting over pieces of the pie instead of trying to preserve the overall operating value of the firm. So I think that the federal government can address some of the restrictions on the auto makers’ dealings with their dealers. There are more and very urgent problems that cannot be handled through legislation. Jackson: Like Jim, I think it would be great if we could remove some of these inefficiencies through legislation—and without resorting to Chapter 11. The history of the last 20 years of General Motors would probably look very different if the company hadn’t been forced to contend with the state franchise laws. But getting political action on this is likely to be difficult. As Jim mentioned, there’s no doubt that such changes would be blocked at the state level. But whether they could be accomplished at the federal level is also highly questionable. It’s this uncertainty about the political process that makes me think that bankruptcy is the right way to go. As I said earlier, the rejection of executor contracts in bankruptcy suggests that Chapter 11 is the ready-made solution to these franchise problems. So, I agree with the premise of the question that a legislated, across-the-board solution would be preferred if possible. But given the realities of the political process, I don’t think we can get it done. Smith: Let me add to Tom’s point. It’s a fairly wellestablished principle in political science that these kinds of “collective action” problems are generally likely to be intractable. You’re extraordinarily unlikely to get a political solution in this case simply because the people who benefit from these franchise laws represent a small number of wellorganized people with large concentrated benefits—namely, the profits from the dealerships. At the same time, the people hurt by these laws—namely, anybody who ever bought a car— are a widely dispersed group of individuals, each bearing a relatively small cost and having little interest in the issue. So this is the collective action problem at work. It’s hard to get millions of people excited about being mugged for a few hundred bucks each when that winds up transferring suitcases full of money to people who get big benefits and make big political contributions. That problem keeps a lot of politicians from forgetting about their commitment to the public good. Hughes: I want to jump in here. In talking about the dealers, I agree that we probably don’t have the will to make a lot of changes that we should. I agree that we should have

114
a federal franchise law. If a dealer goes out of business, or the company wants to stop doing business with a certain dealer, there should be a contract that says, “This is what we’re going to pay you.” We talked about example of GM’s shutting down Oldsmobile earlier. No one knows, or is willing to reveal, the actual costs of ending relationships with the dealers— but in that case it was reportedly over $1 billion, and maybe as high as $2 billion. That subject’s got Rick Wagoner so afraid he won’t even touch it anymore. There are now some 440 Saturn dealers that, although excellent dealers, are not making any money. They should be put to rest. Since they also own a lot of other franchises, you would not be putting them out of business. Now, if it was merely a matter of General Motors going out and saying, “We will buy back the parts and tools, and pay you all the money that we owe in accordance with our contracts,” then we wouldn’t be talking about anything like $1.2 billion to $2 billion. But the dealers are asking for a lot more than that—they want “Blue Sky.” The problem, however, is that there is no longer any Blue Sky in the Saturn franchise; it hasn’t made any money in the last dozen years. But the dealers are still asking for it—and that’s where the problem becomes intractable. So if we did pass a federal law—though I realize it’s unlikely to happen, like a lot of other things we talked about tonight—we could solve that problem. I don’t think bankruptcy, by the way, would be the solution to this problem—though when you’ve called on and negotiated with as many dealers as I have, it sure sounds sweet to be able to do that. But there are other issues that also need to be recognized and addressed. Let me mention one other interesting piece of auto industry history. There’s no question there are some hidden costs and inefficiencies in the system, but at one point in the past, the manufacturers once had the right to their own car dealerships. When the dealers were getting their way with state governments, they succeeded in passing legislation that prevented the automakers from owning dealerships. The interesting thing here is that, behind the scenes, it was people from the manufacturers who were working to get this provision passed—because their own dealers were losing so much money that they wanted a way out. Smith: You mean the manufacturers needed a law to protect them from themselves? Hughes: That’s basically right. There are few things more common than believing you can do something as well as somebody else. Brickley: Well, let me weigh in on this one. If you look at unregulated or less regulated distributor relationships in other industries, you almost never see a so-called “corner solution” where you have either all independent dealerships or 100%

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

company owned-stores. Hughes: Right. Brickley: Most companies use a mix of both arrangements—say, 80% dealerships and 20% dealer-owned stores—depending on variables such as location, and the probability of repeat business. But the state governments have taken that option away from the auto companies. In fact, the governments have even prevented the auto companies from writing their own contracts with the dealers in the sense that the provisions in state law effectively override the contractual agreements where they come into conflict. Hughes: That isn’t the real obstacle. People do buy cars over the Internet every day. Smith: From the manufacturer? Hughes: Not from the manufacturer. But I think you’re making the assumption that it would be more efficient for the consumer to buy directly from the manufacturer than from the dealer. I think that’s a mistake. Smith: I wasn’t making that assumption. I’m assuming that allowing people to experiment with a different model is something that has a lot of value. That by putting a regulatory stop sign at the intersection that says, “You can’t turn down that street,” you take away that opportunity to learn something you didn’t know. Hughes: Well, let me tell you a bit more about what the dealers actually do. One thing we know is that, when you buy a car, it’s probably not the last time you have to go into a dealership. Even Toyotas sometimes have to go back. So there’s a whole array of services in a car transaction that go beyond just buying a car. And, at the moment, the industry has a network of dealers that in most instances has been willing to give the cars away for almost nothing, but is there to service and trade them and help buyers sort out their finances in a way that manufacturer cannot do. It does seem to work. Brickley: Well, let me give you an example of something Ford tried and then got blocked by regulation. In Texas around the year 2000, Ford had a bunch of used cars that they wanted to be able to market directly to buyers over the Internet. The idea was that if they sold the cars, they would then have to contract with some of their dealers to deliver them to the buyers. But this experiment never got off the ground. The dealers who were not part of these arrangements went to the Texas courts and argued that such arrangements were a violation of Texas law. I agree with Cliff that, by tying your hands behind your back and saying you can’t try something, you will never know what might have worked best. Smith: For those of you here who are old enough to remember, this all reminds me of those discussions back in the ’80s about Japanese industrial policy. In those days, publications like Business Week and Fortune and the Harvard Business Review were all talking about how Japan, Inc. was

JACKSON, HUGHES, BRICKLEY, TABAS, AND SMITH — BANKRUPTCY AND BAILOUTS

115

competing the US right off the map. It was an Al Gore kind of national industrial policy in which the future development of the entire economy was orchestrated by the Japanese Ministry of Finance. What happened in this case is that a very small number of admittedly really smart people made huge coordinated bets with the Japanese manufacturing industry. When those bets turned out well, Japan’s productivity soared—and the country ended up moving from ground zero after World War II to being the world’s second largest economic power. But that approach seems to have lost its magic in the last two decades. And that’s not the way we do things in the US Here it’s always been a very large number of people putting their own intuition into their business models and strategies, and putting their own capital on the line to back their bets. What we wind up with is a tremendously robust and resilient economy in which literally millions and millions of these small bets are being made all the time. Some of these bets turn out wonderfully—take Google for example. But a lot of them crash and burn—and you rarely hear about them. Now, the problem with these dealer laws we’re talking about is that they absolutely prevent certain kinds of experimentation. You are legally prohibited from trying certain business models and practices. I just want to say that stopping that kind of experimentation is not without costs.

I’m not arguing that if Ford had been allowed to sell cars directly on the Internet, it would have been a multibillion dollar product line for them. In fact, it may well have blown up in their face. My point is more narrow: The problem here is that we will never know. I’d much rather have the American business community continue to make thousands of calculated bets, putting their money where their mouth is, than having somebody in Washington or Albany say, “As a regulatory matter, we’re not going to let you see if that would work or not.” What you wind up with when you allow that kind of experimentation is a very large portfolio of options. As any finance professor will you, a portfolio of options is dramatically more valuable than an option on a single portfolio. The value of the successes is almost sure to outweigh the losses from the failures for a pretty simple reason: options give you right to keep the upside, but cut your losses and move on when you’re failing. That’s something the US economy has been pretty good at—cutting its losses when necessary and moving on to something more promising. Zupan: Well, let’s leave it at that—and let me thank all of the panelists for taking part in an instructive and entertaining discussion.

116

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Pioneers of Finance

An Interview with Vernon L. Smith: 2002 Nobel Laureate in Economic Sciences and Father of Experimental Economics
Terrance Odean and Betty J. Simkins

“For having established laboratory experiments as a tool in empirical economic analysis, especially in the study of alternative market mechanisms” – 2002 Nobel

 On January 9th 2009, Terry Odean and Betty Simkins interviewed Vernon L. Smith for this issue of the Journal of Applied Finance.1 Vernon Smith is widely regarded as the “father of experimental economics” for his pathbreaking work in this area. After decades of research, the once novel field of ‘experimental economics’ has become a recognized strand of the literature that contributes to our understanding of market mechanisms and more broadly, to the field of behavioral financial economics. In 2002, Vernon Smith was a co-recipient for the Nobel Prize in Economics “for having established laboratory experiments as a tool in empirical economic analysis, especially in the study of alternative market mechanisms”.2 He has written or cowritten more than 200 articles and books on capital theory, finance, experimental economics, and natural resource economics. Dr. Smith remains very active in the economics profession, currently serving as Professor of Economics and Law at Chapman University School of Law.3
1

A video of the interview is available at the Journal of Applied Finance website.

2 Daniel Kahneman was the other co-recipient of the 2002 Nobel Prize “for having integrated insights from psychological research into economic science, especially concerning human judgment and decision-making under uncertainty”.

He has previously held faculty positions at George Mason University, the University of Arizona, Purdue University, Brown University, and the University of Massachusetts. He serves or has served on the board of editors of the American Economic Review; The Cato Journal; Journal of Economic Behavior and Organization; the Journal of Risk and Uncertainty, Science, Economic Theory, Economic Design, Games, and Economic Behavior; and the Journal of Economic Methodology. In addition, he is past president of the Public Choice Society, the Economic Science Association, the Western Economic Association and the Association for Private Enterprise Education. Professor Smith is a distinguished fellow of the American Economic Association, the 1995 recipient of the Adam Smith award, and an elected member of the National Academy of Sciences. Furthermore, he has been a Ford Foundation Fellow, Fellow of the Center for Advanced Study in the Behavioral Sciences, Sherman Fairchild Distinguished Scholar at the California Institute of Technology, Fellow of the Econometric Society, Fellow of the American Association for the Advancement of Science, and Fellow of the American Academy of Arts and Sciences.
3

116

ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. SMITH INTERVIEW

117

In this interview, Vernon Smith shares his insights on markets. Among the issues he addresses are: * the relation between experimental economics and behavioral economics, * the insights his research on speculative bubbles in experimental markets provides for understanding the recent bubble is US residential real estate, and * his view as to reasons for the dramatic rise and fall in oil prices last year. Terry Odean: You received the 2002 Nobel Prize in Economics for your work in experimental economics and you shared the prize with Daniel Kahneman for his contribution to behavioral economics. How are experimental economics and behavioral economics related?

Experimental and behavioral economics are actually complimentary.
Vernon Smith: Well, experimental and behavioral economics are actually complimentary. I see behavioral economics as having evolved out of the early cognitive psychology work with the emphasis being on fundamental decision making, decision making under uncertainty, and related issues. It is also characteristic of some of the work in experimental economics. Most experimental economists that came in the 1960s and 1970s were not primarily focusing on individual decision making under uncertainty. They were first focusing on decision making in markets and market exchange situations and the work was primarily interested in the performance of markets. If you go back before Danny Kahneman, you find people like Sidney Siegel Ward Edwards, who were early psychologists but were not part of the cognitive psychology development. They were seen as the “Skinner behaviorists” – that view the human mind as a kind of black box. The idea was “Let’s just do experiments with it (the human mind) and just see what comes out of it.” Even though I have interacted with Danny Kahneman over the years, we’ve always been interested in different questions and issues. Danny is more utilitarian than I am and that sounds odd – because we economists ought to be utilitarians. Right? I thought the work I did with auctions and auction theory organized the data so well that I was pretty fond of utility theory. What lay behind it ought to be an important part of the picture. Odean: Your early experiments confirmed that markets can work surprisingly well for setting prices that maximize social welfare when participants have private valuations or costs. Can you tell us about these early findings? Smith: In my early work, we began to see and realize in experiments that what we got from utilitarian and formal analysis, the predictions of behaviors in different auction formats, wasn’t holding up. In particular, if you had multiple units, people would get into what they called “jump bidding” and in that jump bidding process, they would miss out on all of the payments.4 It was that sequential nature of the bidding that led to the problem. You can’t believe how many experiments we did to understand that. We ended up just letting a clock raise the price and on each round asked: “Are you in or are you out?” That basically solved the problem. Now in English: This gives you outcomes that are completely efficient. You’re not suppose to jump bid. A person behaving rationally in the English auction should always raise the bid, if the starting bid is less than the value; never bid again yourself – that is, don’t raise your own bid. You should only bid by the minimum increment. If you bid more than that, you have the danger of paying more than you need to. People don’t follow these rules. They have various rationalizations about it. They’ll tell you: “My desire to bid was, by jumping the bid, I would advertise — They (other traders) had better get out because I was going to win.” But as far as we can see, that is completely ineffective. It doesn’t work. It doesn’t bluff anyone out. There is no way this is going to work. With the Federal Communication Commission (FCC) auctions, the increments were $5 and $6 million. Small stakes.5 Some people tried to come up with elaborate theories of what was going on. They thought there is some sort of signaling going on.

4 5

A jump bid is a bid higher than necessary to reach the next bidding level. This refers to FCC auctions of the electromagnetic spectrum. See: wireless.fcc.gov/auctions/

118

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

It was irrational behavior. As far as I could see, it led nowhere. It was jump bidders. The companies participating in the auctions had behaviorists advising them. In my early work on markets, the first ones I ran performed so well I didn’t believe the results. Odean: Did you do that in your classes at Purdue? Smith: Yes. It was what eventually got me hooked into experimental economics. It turns out that people without any

People have this ability, using institutions that somehow survived in our society using those rule systems. They have a capacity to do well for the group while doing well for themselves. Given my traditional economic training, this came as an astonishing surprise.
training in economics, without any understanding of supply and demand, without any sophistication at all walk into a room and they find the equilibrium of these markets. This is robust all types of subjects. They don’t have any idea of that’s what they did, because they don’t have a concept of what there means to be an equilibrium. They don’t have the vocabulary. They will deny there is any kind of model that can predict the convergence properties. It is hard for them to imagine anyone modeling what it is they have done. They also will report, if you ask them: “Is there anything they could have done to increase their earnings in this market?” They are certain there is some way they could, even though by definition of an equilibrium, they can’t. From what everyone else is doing, you can’t improve your position. Many people will say this is a great victory of economic theory. Not exactly, because economic theory never predicted the weak conditions under which it prevailed. The early work on isolated single markets extended to multiple markets – markets in which what you are willing to pay for commodity A depends on the price of B and vice versa. You can only describe the equilibrium with simultaneous nonlinear equations. People find those equilibria too. People have this ability, using institutions that somehow survived in our society using those rule systems. They have a capacity to do well for the group while doing well for themselves. Given my traditional economic training, this came as an astonishing surprise. Odean: In contrast to the early experiments in which markets performed surprisingly well, you’ve studied experimental asset pricing markets that lead to speculative bubbles. Can you tell us more about your work in this area and what conditions lead to bubbles? Smith: We didn’t really start to look at multiple commodity trading markets until we became computerized to handle the mechanics. This led to the idea of looking at asset trading markets in the laboratory by the early 1980’s. Traditional theory from financial economics was that all the information in the markets gets quickly incorporated into the price. One of the things we were interested in doing was studying the possibility of price bubbles. There are a lot of stories about bubbles in history going back to the South Sea Bubble and the Tulip Bubble. We see what happens to the bubbles in the stock market all the time. We thought: “Let’s see what we can learn in the laboratory?” The original idea was to begin with an environment that was transparent; that people would trade at the fundamental value because of the information we gave them. And we would disturb them to see if we could produce a bubble. Well, that research program never got off the ground. Because right from the beginning, we were getting bubbles. I think the important thing we learned was that the common information wasn’t sufficient to give you common expectations. That is actually an experiential process. We brought people back a 2nd and a 3rd time, where the same people would come back before we got convergence — observations where people were trading at fundamental value (the intrinsic value based on the dividend and information that you had from the drawings of that distribution). We had this incredible contrast, in one case – markets worked far better than you expected based on the theory. It really goes back to William Stanley Jevons’ writing in the early 1870’s that contributed the basic supply and demand

ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. SMITH INTERVIEW

119

theory. He posed the question — How would the real market actually approach such an equilibrium? People would have to know basic principles of supply and demand. There was no theorem, no result that said if people have complete information, the market converges; and if they don’t, the market doesn’t converge. It was, I think, simply the theorist saying: How might this actually happen in the world? Someone would have to know what I do and do pencil and paper calculations. That never inspired confidence in it being a very believable model. Essentially, I began doing experiments in the mid to late 1950’s to the mid 1980’s in which we found that this would extend to the asset markets, that performed very badly compared to our expectations. By then, we expected markets to work better than the theory.

We still don’t know what it is about asset trading that leads to the contagion of bubbles that we observe, even though we have ways of modeling them. You can model it by postulating there are two kinds of investment: fundamentalists and momentum traders. The fundamentalists buy in proportion to the discount from true value (fundamental value) and proportion to the premium. The momentum traders are the type of trader who simply buys in proportion to the rate of change in the price. This gives you a differential equation model that gives you bubbles.
Odean: So your early work confirmed how markets worked better than we expected but then your later work in asset markets worked worse than expected, and you got bubbles. Can you share more on this work? Smith: We thought somehow, they (the traders) did not understand the instructions. What we did was to compute for them and then remind them in each period, what the remaining dividends left were and the holding period. They ignored that. It is a beautiful example that if subjects want to dissatisfy the experimenter, they completely falsify his result. People didn’t believe those results (results that were too bad to be true) like my earlier results (results that were too good to be true). We still don’t know what it is about asset trading that leads to the contagion of bubbles that we observe, even though we have ways of modeling them. You can model it by postulating there are two kinds of investment: fundamentalists and momentum traders. The fundamentalists buy in proportion to the discount from true value (fundamental value) and proportion to the premium. The momentum traders are the type of trader who simply buys in proportion to the rate of change in the price. This gives you a differential equation model that gives you bubbles. We are able to come up with models that lead to testable propositions. The momentum traders are very much influenced by the amount of money that is around them. If you model that we have two groups, and group A has a larger endowment of cash, you get bigger bubbles. The model predicts this. This helps us understand why the market is sensitive to monetary policy and what the Federal Reserve is going to do. We’ve tried all kinds. Since people have different endowments of cash and shares, some may be willing to sell for less than fundamental value because they get a more balanced portfolio and are less exposed. We quickly shot that down with people who had exactly the same endowments. It is really common in stock market bubbles to put on price change limit rules. Odean: Many of the world’s exchanges have these such as Taiwan and the Chinese exchanges. Smith: Most of the time, they are wide enough that they are not binding. Odean: They do get hit occasionally. I’ve seen a couple of papers looking at what happens when you hit the limit. Smith: In experimental markets, what we think is going on is people feel there is a downside limit. It is still is an experimental process and the traders have to come back for a 3rd time to finally create near fundamental value. The bubbles can go longer and can carry further.

120

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Odean: In experimental markets, people sometimes believe that there is some downside protection. That is similar to the housing market in 2005 and 2006. People talked about how you can’t lose money in the real estate market. Real estate prices just went up. Or people talked about real estate and said “There is only so much land, so real estate has to go up in price.” You get this belief in the markets. Smith: I think it is a little different in the sense that there is not actually a ceiling or floor on the amount that the price will change. I think what you are getting there is self-reinforcing beliefs about price change. In past housing bubbles, the

I really believe that an important distinction in this housing bubble than previous ones, is that we had a tax law change in 1997.
bubbles have gone on for maybe 3 or 4 years and then the prices decline. This is by far the largest bubble on record. The index started up in 1997. Odean: This was the largest bubble here in the U.S. There was Japan. Smith: You have the phenomena of boom towns where the real estate goes crazy, such as local ones as in Alaska where the oil pipeline was built. That was a speculative bubble and these types of bubbles have been local. The current bubble has hit national proportions and is astonishing — the movement from 1997 to the peak in the CaseShiller index during 2006. My colleague, Steve Gjerstad, had been looking at the structure of those prices — at the low priced, mid-tier, and high priced homes. The low priced homes consistently in market after market, went up faster and further and more sharply. So the housing bubble had components. The lower tier had the biggest bubble component. These are exactly the people that are much more vulnerable to the decline in prices because whatever the wealth they had, it was in those homes. Odean: I think of this in terms of liquidity. At the low end, the ability to buy or not buy was most effected by liquidity. So basically the subprime mortgages created liquidity for people who otherwise wouldn’t have been able to buy and who didn’t have any margin of safety when things went badly. Smith: Yes, I don’t see the financial magic that was being done in the subprime as so much a cause, but as a thing that was a collateral effect. People had been expecting prices to rise and then the subprime lending industry created so much difficulty for the whole financial system. I really believe that an important distinction in this housing bubble than previous ones, is that we had a tax law change in 1997. This tax law change allowed home buyers to receive capital gains from their home that were tax free up to $500,000. This capital gain wasn’t restricted to once; you could do it more than once. There was a two year holding period. This is one of the things I want to look at incidentally, of course. My conjecture is, that if you give favorable tax treatment, say to 1/3rd of U.S. wealth (1/3rd of all US wealth is in a form of homes, about another 1/3 is in all listed securities) and you are taking one of them and giving special capital gains tax-free treatment. And you can’t take the loss on homes against income as you can with stocks Simkins: So you think that the tax law change in 1997 is partially responsible for the crisis we are in now? Smith: Yes. I think if there was a spark that explains why this bubble was bigger than our previous ones, that would be one of the things that I would want to look at. In national statistics, it is without a question the largest housing bubble. There is much more than the tax change of course. Now – you never had as much trouble with FreddieMae and FannieMac before, than this last bubble this time around. There is evidence that there is pressure on FM&FM to take substandard mortgages, both from above (politically) and below (industry), because a lot of people in the industry realize that some of these assets are risky enough…. If we could just get the government to take them over. I think these two things together. We’ve long tried to create conditions for people to own their own wealth through home ownership. It has never succeeded so dramatically as in this last bubble.

ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. SMITH INTERVIEW

121

So what you have is simply people who have a home on the market longer – that happens well before the price – break in prices. Actually, you see that in our experimental markets. Because you see in experimental markets The boom phase, there’s lots more bids than there are asks and you can just look at excess bids (the difference between bids and asks), the bids become an uncertain forecaster of the term and what will happen is that the asks will start to thicken up and prices start to rise and it takes awhile for the bid-ask activity to get back into balance. In experimental markets, it turns again if it over reacts. The bids start to become thicker and the asks start to thin out. The markets actually generate information that people don’t pick up on and immediately incorporate into some sort of rational calculation. This sort of thing can make short selling very hazardous because it is not enough to know that something is overvalued, you must be good at timing.

Somehow, we have to account for why it is we are now crashing from the mother of all housing bubbles. The policy right now is trying to keep prices from falling – but prices have to fall. Unless prices get back to some reasonable levels, we are going to have the problem that buyers are not attracted back – and buyers are the only source of sustainability in this market.
Odean: I learned that the hard way. Smith: You short a stock and it is too soon. It continues to go up and you get a bad stomach ache and you get people buying the cover, and now the market is about to turn down. When we allow short selling in our experimental markets and people can cover their shorts, it can exeaberate bubbles. We see this in our experimental markets. Odean: So you find that you’ve rekindled some bubbles? Smith: I already know that when people had larger endowments in cash, they tended to get bigger bubbles. We could reignite the bubbles which we did, but there weren’t quite as many believers. It took pretty extreme treatments to do that (reignite the bubbles). Which tells you that once people had this experience (the same group) they are reluctant to get into – to repeat that behavior. Of course in the real world, you continue to have new assets coming in. Odean: In the real world, there are lots of investment options — internet stocks in the late 1990’s and real estate stocks a few years later. Smith: People can always tell stories that “This case is different. This situation is different.” But all our evidence indicates that “It’s not!”. We haven’t seen the last housing bubble but we’re just not going to see another one for awhile. I remember being concerned about the change in the tax laws. Not because I was opposed to reduce capital gains. I will make no distinction between capital gains and income. If I would make anything deductible, it would be savings and investments. But there is no way you’re going to see that today, when people are concerned that there is adequate consumption spending. On the other hand, a negative consumption tax, so people with lower income would get rebates; this is an incentive for the poor to accumulate. Odean: You mention the change in the tax law may have contributed to the bubble. Do you see policy implications to your research? Smith: No, but I think it is important that the investment decision not be biased by differential tax treatment. It may turn out that this tax law change is less important (or will be thought to be less important) than I think it is. But it is certainly something I would point out right away. Somehow, we have to account for why it is we are now crashing from the mother of all housing bubbles. The policy right now is trying to keep prices from falling – but prices have to fall. Unless prices get back to some reasonable levels, we are going to have the problem that buyers are not attracted back – and buyers are the only source of sustainability in this market.

122
Simkins: What market conditions discourage bubbles?

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Smith: There is an old fashioned mortgage rule in capitalism – if you are going to borrow money, you have to put up a substantial amount of equity, and that protects not only you but also the lender against the possibility that prices can decline. When we introduced margin buying in a laboratory stock market, it greatly exacerbated bubbles. And tight money and less liquidity reduced the probability of bubbles. This conflicts with the idea of home ownership, where the normal pattern is that

There is an old fashioned mortgage rule in capitalism – if you are going to borrow money, you have to put up a substantial amount of equity, and that protects not only you but also the lender against the possibility that prices can decline. When we introduced margin buying in a laboratory stock market, it greatly exacerbated bubbles. And tight money and less liquidity reduced the probability of bubbles. This conflicts with the idea of home ownership, where the normal pattern is that you have family formation: people save for a period of time while they are renting. They save enough for a down payment and they move into the low end of the housing market. They accumulate more wealth and then they move up into a higher priced house. Anything you do to make it easier to buy a home (by using subsidies or creating an expectation that when you sell, your capital gains are tax free) means you can move people out of this period where they are accumulating and into the next home earlier. What we had in this last housing bubble was a pretty dramatic movement of people out of this traditional way people get started in home ownership. That increases prices and construction costs. Everyone (the buyers, the sellers, the lenders, the mortgage repackagers) believed that prices would go up. So who’s to blame?
you have family formation: people save for a period of time while they are renting. They save enough for a down payment and they move into the low end of the housing market. They accumulate more wealth and then they move up into a higher priced house. Anything you do to make it easier to buy a home (by using subsidies or creating an expectation that when you sell, your capital gains are tax free) means you can move people out of this period where they are accumulating and into the next home earlier. What we had in this last housing bubble was a pretty dramatic movement of people out of this traditional way people get started in home ownership. That increases prices and construction costs. Everyone (the buyers, the sellers, the lenders, the mortgage repackagers) believed that prices would go up. So who’s to blame? Simkins: There has been a debate regarding the large increase in oil prices during the Summer 2008 and the subsequent drop in prices, as to whether it is due to speculation or market fundamentals. What do you think? Smith: I am puzzled by the increase in crude oil if it is not due to speculation. It is very hard to find any justification for the price increase in terms of supply and demand. I think the price increase was very likely due to a lot of speculative capital. Hedge funds were going into crude oil and running the price up to $147 per barrel in a short time. It was a sharp peak and

ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. SMITH INTERVIEW

123

not sustainable. Now it has come back down to around $40. With crude oil and gasoline prices back down, what’s not to stop people from buying bigger cars again? They did it before in the 1970’s. We had a run up in oil prices that almost destroyed the motor home industry and the big gas guzzlers. Then prices dropped back down. Bigger cars are what people wanted to buy. Regarding how to prevent bubbles, I think that is as mysterious as ever. The sparks that actually lead to these bubbles, that precipitate the movement up or the current movement down. I think it is very hard to put your finger on what that is. We can talk about things that affect the severity of the bubble but it doesn’t really give us an idea of what the root cause is. I’m not sure we ever know the root cause. 

124

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

The 2008 Federal Intervention to Stabilize Fannie Mae and Freddie Mac
W. Scott Frame

Fannie Mae and Freddie Mac are government-sponsored enterprises that play a central role in US residential mortgage markets. In recent years, policymakers became increasingly concerned about the size and risk-taking incentives of these two institutions. In September 2008, the federal government intervened to stabilize Fannie Mae and Freddie Mac in an effort to ensure the reliability of residential mortgage finance in the wake of the subprime mortgage crisis. This paper describes the sources of financial distress at Fannie Mae and Freddie Mac, outlines the measures taken by the federal government, and presents some evidence about the effectiveness of these actions. Looking ahead, policymakers will need to consider the future of Fannie Mae and Freddie Mac, as well as the appropriate scope of public-sector activities in primary and secondary mortgage markets.

Fannie Mae and Freddie Mac are enormous governmentsponsored enterprises, or GSEs, that play a central role in US secondary mortgage markets.1 Together, as of mid-year 2008, the two institutions held or guaranteed about $5.5
W. Scott Frame is a Financial Economist and Policy Advisor for the Federal Reserve Bank of Atlanta in Atlanta, GA. The views expressed do not necessarily reflect those of the Federal Reserve Bank of Atlanta, the Federal Reserve System, or their staffs. I would like to thank Michael Hammill for research assistance and Mark Flannery, Diana Hancock, Wayne Passmore, Mario Ugoletti, Larry Wall, and Larry White for providing helpful comments on an earlier draft. “Fannie Mae” and “Freddie Mac” are widely used nicknames for the Federal National Mortgage Association and the Federal Home Loan Mortgage Corporation, respectively.
1

trillion in US residential mortgage debt – slightly more than the $5.3 trillion in publicly held US Treasury debt at that time. Both Fannie Mae and Freddie Mac have been the subject of a great deal of attention and controversy in recent years. Each GSE has: faced accounting scandals, been criticized for not sufficiently targeting their activities toward low-andmoderate income communities and households, and had policymakers voice concerns that they posed a systemic risk to the global financial system.2 At the heart of these (and other) issues is the GSEs’ incentive structure. Fannie Mae and Freddie Mac are publicly traded financial institutions that were created by Acts of Congress in order to fulfill a public mission. These charter Acts imbue the two GSEs with important competitive advantages (most notably, implied public-sector support for their obligations) and define the scope of their permissible activities.3 Over time, Fannie Mae and Freddie Mac became exceptionally large, profitable, and politically powerful. Recently, however, Fannie Mae’s and Freddie Mac’s singular exposure to US residential mortgages – coupled with a thin capital base – resulted in both of these GSEs facing financial distress. US housing markets became increasingly stressed through 2007 and resulted in severe disruption to mortgage markets. Secondary market liquidity for mortgages not backed by Fannie Mae and Freddie Mac almost entirely

For a discussion of the accounting problems at Fannie Mae and Freddie Mac, see the results of special regulatory examination reports at: <http:// www.ofheo.gov/Regulations.aspx?Nav=199>. For an analysis of the GSEs funding of mortgages for low-income borrowers and underserved areas earlier this decade see, for example, Brown (2001) and Bunce (2002). Former Federal Reserve Chairman Greenspan (2005), among others, described the systemic risks posed by the GSEs in testimony before the US Congress.
2

The charter acts may be found at 12 USC. § 1716 et seq. (Fannie Mae) and 12 USC. § 1451 et seq. (Freddie Mac).
3

124

FRAME — THE 2008 FEDERAL INTERVENTION

125
was first known, was created within the federal government in 1938. Its business was to purchase mortgages insured by the Federal Housing Administration, or FHA, from financial institutions around the United States.6 Fannie Mae was subsequently spun-off in 1968 as a publicly traded company as a way to reduce the federal debt during the Vietnam War.7 By contrast, Congress in 1970 created Freddie Mac, which was owned by the 12 Federal Home Loan Banks and the savings and loans that were members of these Banks.8 Freddie Mac became publicly traded in 1989 as part of the thrift crisis resolution.9 Hence, today Fannie Mae and Freddie Mac are quasipublic/quasi-private financial institutions. On one hand, each GSE was created by an Act of Congress and is broadly charged with providing liquidity and stability to the secondary residential mortgage market, with a particular emphasis on housing for low- and moderate-income households and/or in areas viewed as historically underserved (central cities and rural areas).10 On the other hand, Fannie Mae and Freddie Mac have been funded with private capital and their shares are traded on the New York Stock Exchange. This unusual governance arrangement has resulted in two, sometimes opposing, corporate objectives: fulfilling certain social policy goals (and assisting related political constituencies) and maximizing shareholder value. By law, Fannie Mae and Freddie Mac are limited to operating in the secondary conforming mortgage market and

dried-up, and GSE-backed mortgages saw liquidity pressure as evidenced by unusually wide yield spreads. These developments resulted in a significant reduction in the availability and cost of mortgage credit for homeowners. As a result of these developments, the federal government was compelled to intervene to stabilize both GSEs and mortgage markets more generally. On September 7, 2008, Fannie Mae and Freddie Mac were placed into conservatorship by their federal regulator: the Federal Housing Finance Agency (FHFA). Concurrent with this action, the US Treasury entered into “senior preferred stock agreements” with each institution obligating the federal government to inject up to as much as $100 billion each in Fannie Mae and Freddie Mac. The Treasury also established a mortgage-backed securities purchase facility and a standing credit facility in order to support the residential mortgage market. The actions of the FHFA and the Treasury last September stabilized Fannie Mae and Freddie Mac by effectively guaranteeing their debt and mortgage-backed obligations.4 A subsequent announcement by the Federal Reserve that it would purchase substantial quantities of Fannie Mae and Freddie Mac debt and mortgage-backed securities during 2009 has further acted to improve liquidity in those markets and bring yield spreads back to historical norms.5 The remainder of this paper will proceed as follows. Section I provides some background information about Fannie Mae and Freddie Mac and Section II describes the sources of financial distress facing these two GSEs. Section III outlines the steps taken by the federal government to stabilize these systemically important institutions and also presents some evidence relating to the effectiveness of these and other recent federal interventions into secondary mortgage markets. Some concluding remarks are offered in Section IV.

I. Who are Fannie Mae and Freddie Mac?
Fannie Mae’s roots stem from the Great Depression. The National Mortgage Association of Washington, as Fannie Mae
By law, the obligations of Fannie Mae and Freddie Mac must state that they are not guaranteed by the federal government. See 12 USC. § 1719(b),(d)-(e) (Fannie Mae) and 12 USC. § 1455(h)(1) (Freddie Mac). Nevertheless, as discussed further below, financial markets have long viewed the GSEs’ obligations as carrying an “implicit” government guarantee. The federal government’s recent actions were intended to send a strong signal to financial markets that the US would protect the interests of holders of Fannie Mae and Freddie Mac obligations on an ongoing basis.
4

According to Frame and White (2005), by issuing debt and purchasing and holding FHA-insured residential mortgages, Fannie Mae was able to expand the available pool of finance to support housing and also to provide a degree of unification to mortgage markets. During this time, mortgage markets were localized for technological reasons as well as for reasons rooted in laws that prohibited interstate banking and restricted intra-state bank branches in many states during most of the twentieth century.
6

Fannie Mae was replaced within the federal government by the Government National Mortgage Association, or “Ginnie Mae,” an agency within the Department of Housing and Urban Development, or HUD, that guarantees mortgage-backed securities that have as their underlying assets residential mortgages that are insured primarily by the FHA or by the Department of Veterans Affairs (formerly the Veterans Administration, or VA).
7

See Flannery and Frame (2006) for a history and overview of the Federal Home Loan Bank System, which those authors refer to as the “other” housing GSE.
8

According to Frame and White (2005), a major motivation for the conversion of Freddie Mac to a publicly traded company was the belief that a wider potential share-holding public would raise the price of the shares held by the then ailing S&L industry and thus improve the balance sheets of the latter.
9

The maturities of new debt issues by Fannie Mae and Freddie Mac also increased as a result of these policy actions. Nevertheless, the GSE’s access to long-term finance remains limited as it has been for all corporate borrowers.
5

Fannie Mae’s mission or “statement of purpose” can be found at 12 USC. § 1716. A similar statement for Freddie Mac is located at 12 USC. § 1451 [Note].
10

126

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

their activities take two broad forms.11 The GSEs’ “credit Second, the Secretary of the Treasury has the authority to guarantee” businesses involve the creation and credit purchase up to $2.25 billion of Fannie Mae’s and Freddie Mac’s enhancement of mortgage-backed securities, or MBS. This securities, which is often referred to as their federal line-of-credit. is most often done through each institution’s “swap Third, the GSEs’ issue “government securities,” as classified programs,” whereby mortgage originators present pools of under the Securities Exchange Act of 1934, which in practice qualifying mortgages and then exchange them for MBS that means that their securities are eligible for use as collateral for represent an interest in the same pool. The GSEs agree to public deposits, for purchase by the Federal Reserve in openensure the timely market operations, and for payment of principal and unlimited investment by interest on the MBS in federally insured Over time, Fannie Mae and Freddie exchange for a monthly depository institutions.14 Mac became exceptionally large, premium known as a Fourth, Fannie Mae and “guarantee fee”. (This Freddie Mac use the profitable, and politically powerful. process is commonly Federal Reserve as their Recently, however, Fannie Mae’s and referred to as fiscal agent, which means “securitization” although that their securities are Freddie Mac’s singular exposure to US the credit enhancement issued and transferred using structure is much simpler the same system as US residential mortgages – coupled with a than that typically used Treasury borrowings. thin capital base – resulted in both of by investment banks for The features of Fannie similar transformations Mae’s and Freddie Mac’s these GSEs facing financial distress. of loan pools into federal charters, coupled tradable securities.) with some past government GSE-backed MBS are very liquid (relative to other asset- actions, has long served to create a perception in financial markets backed securities and loan pools) and this liquidity facilitates that the federal government “implicitly guarantees” the GSEs’ more efficient balance sheet management for financial financial obligations.15 This belief, in turn, allows Fannie Mae institutions. and Freddie Mac to issue debt at interest rates that are far more Fannie Mae’s and Freddie Mac’s second line of business favorable (better than AAA) than their stand-alone financial is “portfolio investment”. This involves the two GSEs holding strength ratings would warrant.16 This borrowing advantage has MBS that they have purchased in the open market, whole been estimated empirically to be about 40 basis points, although mortgages (purchased from originators under their “cash such estimates vary depending upon the maturity and credit rating programs”), and liquid fixed-income investment securities. of the comparison bonds and the sample period studied.17 The Fannie Mae and Freddie Mac largely fund these assets with so-called “Federal Agency” debt. The two GSEs have historically been highly leveraged with total accounting 14A further implication is that they are exempt from the provisions of many state investor protection laws and the registration and reporting requirements (book) equity equal to less than 4% of total assets.12 and fees of the Securities and Exchange Commission (SEC). Notably, Fannie While Fannie Mae’s and Freddie Mac’s federal charters limit Mae voluntarily registered its stock with the SEC in March 2003 and Freddie the scope of their business activities to the secondary residential Mac did the same in July 2008. mortgage market, they also provide them with a number of 15 This perception arises despite explicit language on each GSEs’ securities advantages that result in lower operating and funding costs.13 that they are not obligations of the federal government. US General First, both GSEs are exempt from state and local income taxes. Accounting Office (1990, 90–91) discusses two past episodes during which
11

See 12 USC. 1719 (Fannie Mae) and 12 USC. 1454 (Freddie Mac). Conforming mortgages are those with balances below the legal limits on the size of residential mortgages that Fannie Mae and Freddie Mac can buy. For 2009, the conforming loan limit for single-family properties is $417,000, but can be as high as $625,500 in certain high-cost areas. See <http://www.ofheo.gov/Regulations.aspx?Nav=128>.

the federal government assisted troubled GSEs. First, during the late 1970s and early 1980s, Fannie Mae was insolvent on a market value basis and benefited from supervisory forbearance. Second, in the late 1980s, the Farm Credit System (another GSE serving the agricultural sector) required a taxpayer bailout totaling $4 billion. Fannie Mae and Freddie Mac long received AA- ratings from Standard and Poor’s in terms of their “risk to the government”. However, these ratings incorporated whatever government support or intervention the entity typically enjoyed during the normal course of business. See Frame and Wall (2002) for a discussion.
16

By contrast, commercial banks (as a group) maintain a ratio of total equity to total assets of about 10 %.
12

See, for example, US Congressional Budget Office (1996, 2001) for further discussion.
13

See Ambrose and Warga (1996, 2002), Nothaft, Pearce, and Stevanovic (2002), and Passmore, Sherlund, and Burgess (2005).
17

FRAME — THE 2008 FEDERAL INTERVENTION

127
equity holders thus perceive a greater-than-normal benefit from risk-taking. In order to maximize benefits to homebuyers and minimize taxpayer risk, the federal government imposes a two-part regulatory structure on Fannie Mae and Freddie Mac. The US Department of Housing and Urban Development, or HUD, long regulated the GSEs for compliance with their mission of enhancing the availability of mortgage credit by creating and maintaining a secondary market for residential mortgages. HUD was also responsible for establishing goals (and monitoring compliance with the goals) for Fannie Mae’s and Freddie Mac’s financing of housing for low- and moderate-income families, housing in central cities, and other “underserved areas”. Congress formally established a safetyand-soundness regulatory and supervisory regime for Fannie Mae and Freddie Mac in 1992. The Office of Federal Housing Enterprise Oversight, or OFHEO, was authorized to set riskbased capital standards (subject to important statutory limitations), conduct examinations, and take enforcement actions if unsafe or unsound financial or management practices were identified.20 Unfortunately, OFHEO’s structure and authorities proved deficient in many respects.21 GSE regulatory reform was an active legislative item this decade following the accounting scandals at both Fannie Mae (2004) and Freddie Mac (2003). However, it was not until the GSEs came under serious financial strain that reform was passed as part of the Housing and Economic Recovery Act of 2008. The new law created the Federal Housing Finance Agency (FHFA), which consolidated the mission and safety and soundness oversight for Fannie Mae, Freddie Mac, and the Federal Home Loan Bank System.22 The establishment of the FHFA reflects an improvement in GSE safety-andsoundness supervision and regulation since the new regulator (among other things): (1) no longer requires Congressional approval for its budget, (2) has authority to set minimum leverage and risk-based capital requirements, and (3) has receivership powers.
Prior to the Federal Housing Enterprises Financial Safety and Soundness Act of 1992, HUD maintained exclusive regulatory oversight responsibilities over Fannie Mae and (for 1989-1992) Freddie Mac. Prior to the passage of the Financial Institutions Reform, Recovery and Enforcement Act of 1989, Freddie Mac was the responsibility of the Federal Home Loan Bank Board.
20

perceived implied guarantee also affects the interest rates on MBS that Fannie Mae and Freddie Mac issue, although the advantage is difficult to estimate.18 The perception of an implied federal guarantee conveys a subsidy on Fannie Mae and Freddie Mac, part of which is translated into lower mortgage rates for consumers. In particular, Fannie Mae’s and Freddie Mac’s activities result in conforming mortgages’ carrying lower interest rates than “jumbo mortgages” with principal amounts above the conforming loan limit. Several econometric studies estimated the effect of GSEs on conforming mortgage rates, typically finding the interest rate differential to be about 20-25 basis points with variation in the estimates depending on the empirical specification, data sample, and time period studied.19 Fannie Mae and Freddie Mac have been largely free from market constraints on their size and risk because of the market perception of an implied federal guarantee of their obligations. The GSEs have become enormous financial institutions – both in absolute terms and relative to the mortgage market as a whole. As of June 30, 2008, Fannie Mae and Freddie Mac together held almost $1.8 trillion in assets (almost entirely MBS and whole mortgages) and had another $3.7 trillion in net credit guarantees outstanding – i.e., net of those held in their own portfolios. This $5.5 trillion in obligations represented almost half of all residential mortgage debt outstanding at that time. The two GSEs have also grown much more rapidly than the residential mortgage market as a whole over the past three decades. For example, in 1980, Fannie Mae’s and Freddie Mac’s share of residential mortgage debt outstanding was only 7% (Frame and White, 2005). The perceived implied federal guarantee also distorts the GSEs’ risk-taking incentives in a way that may increase the probability of financial distress. (A similar situation is wellunderstood in the context of federally insured depository institutions.) The idea is that a federal guarantee induces debt holders to accept artificially low interest rates irrespective regardless of a GSE’s true default risk. A GSE can then increase the riskiness of its activities – which promise high returns to equity holders if the risks turn out well – without needing to share those rewards with debt holders in the form of higher coupon rates on their debt. The GSEs’

US Congressional Budget Office (1996, 2001) reported an MBS advantage of 30 basis points, but Passmore (2005, p. 9) critiques the approach that generates this estimate and alternatively argues that the advantage is in the range of 0-6 basis points. See also Heuson, Passmore, and Sparks (2001) and Passmore, Sparks, and Ingpen (2002) for theoretical analyses of the relationship between GSE securitization and mortgage interest rates.
18

For an introduction to this literature, see US Congressional Budget Office (2001), McKenzie (2002), Passmore (2005), and Ambrose, LaCour-Little and Sanders (2004), and the references in these papers.
19

As discussed in Eisenbeis, Frame, and Wall (2007), OFHEO supervised only two institutions making it prone to regulatory capture. The agency was also an independent arm of HUD, which is more focused on promoting housing than contending with GSE safety-and-soundness. OFHEO was also subject to Congress’ annual appropriations process and sometimes fell victim to political meddling. With respect to supervisory tools, OFHEO lacked the authority both to adjust minimum capital standards and to resolve a failure of either Fannie Mae or Freddie Mac.
21 22 By doing so, the FHFA succeeds the OFHEO, HUD’s GSE mission regulation, and the Federal Housing Finance Board.

128 II. Sources of Financial Distress

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

mortgage-related risk and leverage were a consistent theme raised by Federal Reserve officials throughout this decade US housing and mortgage markets became increasingly (e.g., Greenspan 2005, Bernanke 2007). As of mid-year 2007 (and prior to the beginning of the stressed during 2007 and 2008, largely as a result of house financial crisis), Fannie Mae and Freddie Mac maintained price declines in many parts of the country. Between 2007:Q2 and 2008:Q3 house prices declined 18.0% on a nationwide book equity values of $39.7 billion and $25.8 billion, basis based on the S&P/Case-Shiller composite index. By respectively. This combined $65.5 billion in equity stood against almost $1.7 trillion in combined assets (3.9% capitalcontrast, over the same to-assets ratio) and another period, the OFHEO $3.2 trillion in net off-balance nationwide house price Given the ongoing decline in house sheet credit guarantees. One index fell 4.5%. While the year later, the two GSEs had magnitudes of decline in prices, mounting foreclosures, and expanded to almost $1.8 these repeat-sales indices trillion in combined assets and the weakening global economy, it differ – owing to coverage $3.7 trillion in combined net differences by geography, is likely that the GSEs credit losses off-balance sheet credit loan size, and loan quality – guarantees, but their capital will remain elevated for some time. this national decline in cushions had begun to erode. 23 house prices is unusual. During those four intervening House price declines quarters, Fannie Mae posted $9.5 billion in losses (although resulted in a large number of borrowers having mortgage balances that exceeded the value of their homes — a condition it did raise $7.0 billion in new equity) and Freddie Mac lost often referred to as “negative equity”. Economic theory and another $4.7 billion. Moreover, mark-to-market accounting evidence suggest that negative equity is a necessary condition losses on ‘available-for-sale’ mortgage-backed securities for mortgage default. 24 Borrowers may face income substantially reduced equity through negative entries to disruptions that temporarily limit their ability to pay and have ‘accumulated other comprehensive income’ on the GSE’s neither sufficient savings nor home equity to cover monthly balance sheets. As of June 30, 2008, Fannie Mae and Freddie living expenses. Other borrowers may default after finding Mac reported book values of equity of $41.2 billion and $12.9 themselves in a situation where their expectations of future billion, respectively. Perhaps more telling was that the GSEs’ house prices are such that they see little hope of attaining self-reported fair values of equity (i.e., the market value of positive equity in the foreseeable future. In any event, the assets less the market value of liabilities) as of the same date house price declines witnessed in 2007 and 2008 have resulted were $12.5 billion (Fannie Mae) and -$5.6 billion (Freddie 25 in a tremendous wave of mortgage defaults and foreclosures Mac). During 2008, significant problems emerged in both of that, in turn, has imperiled financial institutions with Fannie Mae’s and Freddie Mac’s business lines – credit significant credit exposure to US residential real estate – particularly exposure to rapidly declining markets and/or to guarantees and portfolio investment. The credit guarantee riskier subprime borrowers and investors. Fannie Mae and businesses incurred rapidly increasing expenses, largely Freddie Mac certainly fit this bill, as did thrift institutions owing to loan loss provisions. During 2006, Fannie Mae operating on a nationwide basis like Countrywide and and Freddie Mac together incurred about $1.1 billion in creditrelated expenses. These expenses rose to $1.6 billion during Washington Mutual. Fannie Mae and Freddie Mac were not only singularly the first half of 2007 alone, and then jumped markedly to exposed to US residential mortgages, but also operated with $6.5 billion during the second half of that year. For the first a high degree of leverage, owing to a statutory minimum half of 2008, Fannie Mae and Freddie Mac again saw creditcapital requirement of only 2.5% for on-balance-sheet assets related expenses almost double to $12.8 billion, and for and 0.45% for net off-balance sheet credit guarantees. 2008:Q3 alone they totaled $15.3 billion. Given the ongoing Concerns about the GSEs’ concentration of residential decline in house prices, mounting foreclosures, and the
Information about the S&P/Case-Shiller house price index can be found at: <http://www2.standardandpoors.com/portal/site/sp/en/us/page.topic/ indices_csmahp/0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0.html>. Information about the OFHEO house price index can be found at: <www.ofheo.gov>. See Leventis (2007) for an analysis of the differences between the two indices.
23 24

See Foote, Gerardi, and Willen (2008) and references therein.

According to Financial Accounting Statement (FAS) Number 157, “fair value” is the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. For the GSEs’ fair value balance sheets, see: <http:// www.fanniemae.com/media/pdf/newsreleases/q22008_release.pdf> (Page 17) and <http://www.freddiemac.com/investors/er/pdf/2008fintbls_080608.pdf> (Page 13).
25

FRAME — THE 2008 FEDERAL INTERVENTION

129
and Freddie Mac. Figure 2 illustrates how the GSEs’ share prices fell during that time (and following an even more dramatic decline during the fall of 2007). Debt investors also sought clarity from the federal government about whether bondholders would be shielded from any losses that might ultimately arise. Figure 3 shows prices for credit default swaps on Fannie Mae and Freddie Mac senior and subordinated debt between January 2007 and July 2008. Of particular note are the spikes in March 2008 (just prior to the Bear Stearns rescue) and then again during June and July 2008. Holders of Fannie Mae and Freddie Mac senior debt (and MBS) appear to have only reacted modestly to the widespread perception of GSE financial distress. However, one especially significant and risk-averse investor constituency, foreign official institutions, began decreasing their holdings of Federal Agency obligations at that time. Figure 4 presents relevant weekly data based on holdings in custody accounts at the Federal Reserve Bank of New York.29 In response to increasing concerns that Fannie Mae and Freddie Mac would be unable to rollover their debt, former US Treasury Secretary Henry Paulson requested that the federal government be given broad authority to invest in the two GSEs. That provision was included in the Housing and Economic Recovery Act that passed in July 2008.

weakening global economy, it is likely that the GSEs credit losses will remain elevated for some time. Fannie Mae’s and Freddie Mac’s portfolio investment businesses also suffered from mark-to-market losses on mortgage-backed securities held either in trading accounts or as “available for sale”. (Under Generally Accepted Accounting Principles, or GAAP, securities classified as “hold-to-maturity” are not marked-to-market unless there is an “other than temporary impairment” to value.) This was caused by an unusual and unforeseen widening of the yield spread between Fannie Mae and Freddie Mac-guaranteed MBS and 10-year Treasuries. Figure 1 presents the current coupon spreads on 30-year fixed-rate mortgages (to 10-year Treasuries) for Fannie Mae and Freddie Mac between January 2007 and July 2008. The observed widening is believed to be primarily caused by the financial market turbulence, which led to a heightened demand for US Treasury obligations that was reflected by lower Treasury yields. However, the aforementioned credit problems at Fannie Mae and Freddie Mac also likely played a role by pushing-up required yields on the GSEs’ MBS. Mark-to-market losses also occurred in each GSEs’ holdings of “private-label” mortgage securities backed by subprime and Alt-A mortgages.26 As of mid-year 2007, the two GSEs held $252.7 billion in mortgage securities backed by subprime and Alt-A mortgages — virtually all of which were rated AAA.27 The GSEs’ holdings of such securities likely reflected at least two factors. One is the distorted risktaking incentives faced by Fannie Mae and Freddie Mac because of the perceived implied federal guarantee of their obligations. Another factor was the HUD-regulated affordable housing goals that mandated a certain percent of each institution’s business devoted to affordable housing.28 Private-label MBS held by Fannie Mae and Freddie Mac were typically backed by a greater concentration of affordable housing goal-eligible loans than their own MBS. During the summer of 2008, investors became increasingly concerned about the financial condition of both Fannie Mae
Private-label mortgage securities are those not guaranteed by Fannie Mae, Freddie Mac, or Ginnie Mae. Subprime mortgages refer to those loans made to borrowers with riskier credit characteristics, such as measured by credit scores, loan-to-value ratios, or debt-to-income ratios. Alt-A mortgages refer to loans made with little or no documentation.
26

III. Federal Intervention
According to former Treasury Secretary Henry Paulson (2009), immediately following the passage of the new housing legislation, the Treasury began a comprehensive financial review of Fannie Mae and Freddie Mac in conjunction with the FHFA, the Federal Reserve, and Morgan Stanley.30 The GSEs believed in their solvency and thought that any capital deficiency below regulatory minimums could be rectified by significant asset sales. Given that US mortgage markets had already been disrupted for almost one year at that time, the prospect of Fannie Mae and Freddie Mac retrenching was not an appealing policy option. In early August 2008, both Fannie Mae and Freddie Mac released their second quarter earnings. As of June 30, 2008

Data provided in US Office of Federal Housing Enterprise Oversight (2008) indicate that, of this amount, Fannie Mae accounted for $81.4 billion and Freddie Mac $174.3 billion.
27 28 The GSEs’ housing goals were established in the Federal Housing Enterprises Financial Safety and Soundness Act of 1992. The law required HUD to set annual housing goals for Fannie Mae and Freddie Mac and to monitor the GSEs’ performance in meeting those goals. This responsibility was transferred to the FHFA as part of the Housing and Economic Recovery Act of 2008.

Fannie Mae and Freddie Mac account for about 70% of all Federal Agency obligations outstanding. The other key issuer is the Federal Home Loan Bank System. The data comes from a memorandum to the Federal Reserve’s H.4.1 release: Factors Affecting Reserve Balances. Quarterly Flow of Funds data (Table L.107) corroborates the trend illustrated by showing that foreign official holdings of Federal Agency obligations peaked in 2008:Q2. Interestingly, the same data indicates that holdings in foreign private accounts peaked sooner – as of 2007:Q4.
29

Morgan Stanley was hired by the Treasury to provide market analysis and financial expertise in connection with its authorities to invest in Fannie Mae and Freddie Mac (e.g., Solomon and Paletta, 2008).
30

130

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Figure 1: Federal Agency 30-year Current Coupon MBS Spread to 10-year Treasury

basis points
250

200

150

100

50

Fannie Mae Freddie Mac
0 Jan-07 Mar-07 May-07 Jul-07 Sep-07 Nov-07 Jan-08 Mar-08 May-08 Jul-08

Source: Bloomberg

Figure 2: Fannie Mae & Freddie Mac Stock Prices

80 70 60 50 40 30 20 10 0 Jan-07

Fannie Mae Freddie Mac Mar-07 May-07 Jul-07 Sep-07 Nov-07 Jan-08 Mar-08 May-08 Jul-08

Source: Bloomberg

FRAME — THE 2008 FEDERAL INTERVENTION

131

Figure 3: Fannie Mae & Freddie Mac Credit Default Swaps

p
basis points, 5-year (senior & subordinated debt) 300 Freddie Mac Subordinated Freddie Mac Senior 250 Subordinated 200 Fannie Mae Subordinated Fannie Mae Senior

150

100

50 Senior 0 Jan-07 Mar-07 May-07 Jul-07 Sep-07 Nov-07 Jan-08 Mar-08 May-08 Jul-08

Source: Bloomberg

Figure 4: Marketable Federal Agency Securities Held for Foreign Official & International Accounts

$ billions
1,100 1,000 900 800 700 600 500 400 Jan-07

Apr-07

Jul-07

Oct-07

Jan-08

Apr-08

Jul-08

Oct-08

Source: Federal Reserve Board

132
both GSEs were both technically solvent insofar as the book value of their equity capital was positive. (At that time, Fannie Mae had $41.2 billion in book equity and Freddie Mac $12.9 billion.) However, there was a compelling case that – on an economic basis – both were actually insolvent. First, as mentioned previously, the GSEs’ reported fair values of equity were much lower – and in Freddie Mac’s case fair value was actually negative. Second, both institutions were carrying relatively large “tax deferred assets” to allow them to reduce future income taxes. These amounts were $20.6 billion for Fannie Mae and $18.4 billion for Freddie Mac. If Fannie Mae and Freddie Mac were subject to the bank regulatory standard for tax-deferred assets – and in light of their extremely weak near-term earnings prospects – those assets would have been written-off and taken total book equity down to $20.6 billion (Fannie Mae) and -$5.5 billion (Freddie Mac).31 These facts, taken together with deteriorating mortgage market conditions and a view that the GSEs had been especially conservative in estimating expected future losses, made a compelling case for swift federal action.32 And on September 7, 2008, FHFA Director James Lockhart, Treasury Secretary Henry Paulson, and Federal Reserve Chairman Ben Bernanke outlined a plan to stabilize the residential mortgage finance market. This included: (1) placing both Fannie Mae and Freddie Mac into conservatorship, (2) having the Treasury enter into senior preferred stock agreements with both GSEs, and (3) establishing two new Treasury-operated liquidity facilities aimed at supporting the residential mortgage market — a mortgage-backed securities purchase facility and a standing credit facility. The reasoning for the imposition of the conservatorships was that both Fannie Mae and Freddie Mac were financially distressed and could not perform their public missions – that is, providing counter-cyclical support to mortgage markets and financing affordable housing. By becoming a conservator, the FHFA assumed the responsibilities of the directors, officers, and shareholders of both Fannie Mae and Freddie Mac with the purpose of conserving each GSEs’ assets and to rehabilitate them into safe-and-sound condition. New CEOs were named to act as agents of the conservator. Concurrent with the conservatorships, the Treasury entered

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

into a senior preferred stock agreement with each GSE.33 The purpose of the agreements is to ensure that Fannie Mae and Freddie Mac maintain positive net worth going forward. If the regulator determines that either institution’s liabilities exceed assets under GAAP, the Treasury will contribute cash capital equal to the difference in exchange for senior preferred stock. Each of these agreements is of an indefinite term and for up to $100 billion. After its 2008:Q3 earnings release, Freddie Mac drew $13.8 billion. Both GSEs are expected to require significant Treasury capital infusions after the announcement of their respective year-end 2008 financials. Preliminary figures suggest that Fannie Mae will require as much as $16 billion and Freddie Mac as much as another $35 billion (Kopecki 2009). The senior preferred stock accrues dividends at 10% per year, a rate that steps up to 12% if in any quarter dividends are not paid in cash. Also, in exchange for the senior preferred stock agreements, the Treasury received from each Fannie Mae and Freddie Mac: (1) $1 billion of senior preferred shares, (2) warrants for the purchase of common stock representing 79.9% of each institution on a fully diluted basis, and (3) a quarterly commitment fee (starting March 31, 2010) to be determined by the Treasury and the FHFA (as conservator) in consultation with the Federal Reserve. The senior preferred stock agreements require each GSE to begin shrinking their retained investment portfolios in 2010 at a rate of 10% per year until they each fall below $250 billion. This provision was intended to assuage policymaker concerns about the GSEs’ investment portfolios, which had become widely viewed as posing a systemic risk to the financial system and providing little social welfare benefit.34 The senior preferred stock agreements also included various covenants. Specifically, Treasury approval is required before: (1) purchasing, redeeming or issuing any capital stock or paying dividends, (2) terminating conservatorship other than in connection with receivership, (3) increasing debt to greater than 110% of that outstanding as of June 30, 2008, and (4) acquiring, consolidating, or merging into another entity. The Treasury’s GSE credit facility is for Fannie Mae, Freddie Mac, and the Federal Home Loan Bank System and is operated by the Federal Reserve Bank of New York.35 As of year-end 2008, no credit had been extended through this program. The MBS purchase program, by contrast, had
33

While acceptable under GAAP, bank regulators require institutions to write-off all but the lesser of: (1) the amount of tax deferred assets the institution expects to realize in the next 12 months, or (2) 10% of Tier 1 capital. For example, for state member banks, see: 12 C.F.R. 208 Appendix A, Section II(B)(4).
31

See <http://ustreas.gov/press/releases/reportspspa_factsheet_090708% 20hp1128.pdf.

See Eisenbeis, Frame, and Wall (2007) for an overview of the policy concerns and the related literature.
34

Morgenstern and Duhigg (2008) report that Morgan Stanley (working on behalf of the Treasury) concluded that both GSEs had overstated their financial condition by postponing various types of losses.
32

Credit must be collateralized, can be extended for one-to-four weeks, and is priced at LIBOR plus 50 basis points. Eligible collateral is limited to Fannie Mae and Freddie Mac mortgage-backed securities and Federal Home Loan Bank advances.
35

FRAME — THE 2008 FEDERAL INTERVENTION

133
facilities to: 1) purchase up to $500 billion in mortgagebacked securities guaranteed by Fannie Mae, Freddie Mac, and Ginnie Mae, and 2) purchase up to $100 billion in debt obligations of Fannie Mae, Freddie Mac, and the Federal Home Loan Bank System. Figures 5 and 6 also show a positive market response to these announcements.

accumulated $71.5 billion. 36 Consistent with the GSE investment provisions in the Housing and Economic Recovery Act of 2008, credit extensions and MBS purchases must be made by year-end 2009 (although previously purchased securities may be held beyond that time). The intent of the senior preferred stock agreements and Treasury liquidity facilities was to provide comfort to Fannie Mae’s and Freddie Mac’s senior and subordinate creditors and holders of mortgage-backed securities.37 By extension, these actions were expected to lower and stabilize the cost of mortgage finance. Figures 5 and 6 illustrate the announcement effect for Fannie Mae and Freddie Mac 5-year debt spreads and current coupon MBS spreads, respectively. The tighter spreads on mortgage-backed securities, in turn, resulted in conforming mortgage rates falling by about 50 basis points. Of course, the two agreements had significant negative consequences for the GSEs’ common and preferred stockholders. Fannie Mae and Freddie Mac common shares quickly fell below $1, down from $60 just 12 months earlier. Indeed, as a result of trading at such low levels, the two GSEs now face delisting.38 Preferred shares suffered a similar fate. Indeed, several community banks became financially distressed themselves as a result of having to write-down the value of their holdings of GSE preferred stock.39 The positive bond market reaction, coupled with a relatively smooth operational transition, suggested that the imposition of conservatorships at Fannie Mae and Freddie Mac was, so far, a success. However, by November 1, 2008, mortgage rates and yields on Fannie Mae and Freddie Mac obligations had climbed back to pre-conservatorship levels because of worsening financial market conditions. Policymakers then searched for additional tools to lower and stabilize the cost of mortgage finance. In response, the Federal Reserve announced on November 25 that it was establishing new
36

IV. Conclusion
Fannie Mae and Freddie Mac play a central role in the US residential mortgage finance system. As real estate prices fell and mortgage defaults and foreclosures mounted, the two highly leveraged GSEs became financially distressed. In response, Fannie Mae’s and Freddie Mac’s federal regulator placed both institutions into conservatorship and the US Treasury entered into senior stock purchase agreements with each GSE and introduced new liquidity facilities aimed at supporting the institutions and mortgage markets more generally. The federal intervention into Fannie Mae and Freddie Mac has been successful insofar as it improved the confidence of creditors and stabilized residential mortgage markets. However, the current arrangement of government ownership and control over these two enormous financial institutions will likely be revisited by the Congress in the months ahead. Today’s consensus appears to be that the previous publicprivate business model is inherently flawed and unstable. Indeed it is unclear what role Fannie Mae and Freddie Mac will ultimately play in the US housing finance system, and the reasons for this uncertainty do not solely rest with the two GSEs. The financial distress at Fannie Mae and Freddie Mac has occurred along with significant and well-publicized problems at a host of mortgage originators, private mortgage insurance companies, and monoline bond insurers. Hence, the federal government may need to redefine its role in supporting primary and secondary mortgage markets. Federal Reserve Chairman Bernanke (2008) and former Treasury Secretary Paulson (2009) have offered some initial thoughts about various policy options. Nevertheless, additional research and policy analysis should commence quickly about the public-sector’s role in mortgage markets, the efficacy of the GSE model of financial intermediation, and the future of Fannie Mae and Freddie Mac.

See <http://www.fms.treas.gov/mts/index.html>.

On September 11, 2008, the Treasury issued a press release intended to clarify the status of the senior preferred stock agreements. See <http:// www.ustreas.gov/press/releases/hp1131.htm>. The Treasury affirmed that the agreements are permanent and that legislative efforts to abrogate them would give rise to government liability to parties suing to enforce their rights under the agreements. The senior preferred stock agreements may only be terminated by either: 1) full funding by the Treasury ($100 billion), 2) GSE liquidation, or 3) GSE satisfaction of all liabilities. In some sense, the senior stock purchase agreements have become an appendage to the GSE charters.
37

The NYSE Listing Manual (Part 802.01C) notes that a company will be deemed to be below compliance standards if the average closing price of a security is less than $1.00 over a consecutive 30-day trading period. Once notified, a company has six months to bring its share price and average share price above the $1.00 threshold. Fannie Mae and Freddie Mac were each notified in November 2008.
38 39

See McGeer (2008) and Blackwell and Flitter (2008) for some discussion.

134

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

Figure 5: Federal Agency 5-year Debt Spread to 5-year Treasury

180 Conservatorship 150 120 90 60

basis points

Fed Agency Purchase Announcement 30 Fannie Mae Freddie Mac 0 Aug-08
Source: Bloomberg

Sep-08

Oct-08

Nov-08

Dec-08

Figure 6: Federal Agency 30-year Cuurent Coupon MBS SPread to 10-Year Treasury

basis points 250

200

150 Conservatorship 100

50 Fannie Mae Freddie Mac 0 Aug-08
Source: Bloomberg

Fed MBS Purchase Announcement Oct-08 Nov-08 Dec-08

Sep-08

FRAME — THE 2008 FEDERAL INTERVENTION

135
Frame, W.S. and L.J. White, 2005, “Fussing and Fuming over Fannie and Freddie: How Much Smoke, How Much Fire?” Journal of Economic Perspectives 19 (No. 2), 159-184. Greenspan, A., 2005, “Testimony before the Committee on Banking, Housing, and Urban Affairs,” United States Senate (April 6), http://www.federalreserve.gov/boarddocs/ testimony/2005/20050406/default.htm. Heuson, A., W. Passmore, and R. Sparks, 2001, “Credit Scoring and Mortgage Securitization: Implications for Mortgage Rates and Credit Availability,” Journal of Real Estate Finance and Economics, 23 (No. 3), 337-363. Kopecki, D., 2009, “Fannie, Freddie Funding Needs May Pass $200 Billion, FHFA Says,” Bloomberg (February 10). Leventis, A., 2007, “A Note on the Differences between the OFHEO and S&P/Case-Shiller House Price Indexes,” http:// www.ofheo.gov/media/research/notediff2.pdf. McKenzie, J., 2002, “A Reconsideration of the Jumbo/NonJumbo Mortgage Rate Differential,” Journal of Real Estate Finance and Economics 25 (No. 2), 197-213. McGeer, B., 2008, “Preferred Exposure Fallout,” American Banker (September 9). Nothaft, F.E., J.E. Pearce, and S. Stevanovic, 2002, “Debt Spreads Between GSEs and Other Corporations,” Journal of Real Estate Finance and Economics 25 (No. 2), 151172. Passmore, W., 2005, “The GSE Implicit Subsidy and the Value of Government Ambiguity,” Real Estate Economics 33 (No. 3), 465-486. Passmore, W., S. Sherlund, and G. Burgess, 2005, “The Effect of Housing Government-Sponsored Enterprises on Mortgage Rates,” Real Estate Economics 33 (No. 3), 427463. Passmore, W., R. Sparks, and J. Ingpen, 2002, “GSEs, Mortgage Rates, and the Long-Run Effects of Securitization,” Journal of Real Estate Finance and Economics 25 (No. 2) 215-242. Paulson, H., 2009, “The Role of the GSEs in Supporting the Housing Recovery,” Remarks to the Economic Club of Washington (January 7). Solomon, D. and D. Paletta, 2008, “Treasury Hires Morgan Stanley for Advice on Fannie, Freddie,” Wall Street Journal (August 6).

Ambrose, B.W., M. LaCour-Little, and A.B. Sanders, 2004, “The Effect of Conforming Loan Status on Mortgage Yield Spreads: A Loan Level Analysis,” Journal of Real Estate Economics 32 (No. 4), 541-569. Ambrose, B.W. and A. Warga, 1996, “Implications of Privatization: The Costs to Fannie Mae and Freddie Mac,” In US Department of Housing and Urban Development, Studies on Privatizing Fannie Mae and Freddie Mac Washington, D.C., HUD, 169-204. Ambrose, B., and A. Warga, 2002, “Measuring Potential GSE Funding Advantages,” Journal of Real Estate Finance and Economics 25 (No. 2-3), 129-150. Bernanke, B.S., 2008, “The Future of Mortgage Finance in the United States,” Remarks to the UC Berkeley/UCLA Symposium: The Mortgage Meltdown, the Economy, and Public Policy (October 31). Bernanke, B.S., 2007, “GSE Portfolios, Systemic Risk, and Affordable Housing,” Remarks to the Independent Community Bankers of America, (March 6). Blackwell, R. and E. Flitter, 2008, “Regulators and Bankers at Odds Over GSE Seizure,” American Banker (September 11). Brown, J., 2001, “Reform of GSE Housing Goals,” In Peter J. Wallison, Editor, Serving Two Masters Yet out of Control, Washington D.C., AEI Press, 153-165. Bunce, H.L., 2002, “The GSEs Funding of Affordable Loans: A 2000 Update,” US Department of Housing and Urban Development Housing Finance Working Paper HF-013 (April), http://www.huduser.org/publications/hsgfin/ workpapr13.html. Eisenbeis, R.A., W.S. Frame, and L.D. Wall, 2007, “An Analysis of the Systemic Risks Posed by Fannie Mae and Freddie Mac and An Evaluation of the Policy Options for Reducing Those Risks,” Journal of Financial Services Research 31 (No. 2), 75-99. Flannery, M.J. and W.S. Frame, 2006, “The Federal Home Loan Bank System: The ‘Other’ Housing GSE,” Federal Reserve Bank of Atlanta Economic Review 91 (Q3), 3354. Foote, C.L., K. Gerardi, and P.S. Willen, 2008, “Negative Equity and Foreclosure: Theory and Evidence,” Journal of Urban Economics 64 (No. 2), 234-245. Frame, W.S. and L.D. Wall, 2002, “Financing Housing through Government-Sponsored Enterprises,” Federal Reserve Bank of Atlanta Economic Review 87 (Q1), 29-43.

136
US Congressional Budget Office, 1996, Assessing the Public Costs and Benefits of Fannie Mae and Freddie Mac, Washington, D.C., CBO. US Congressional Budget Office, 2001, Federal Subsidies and the Housing GSEs, Washington, D.C., CBO.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

US General Accounting Office, 1990, GovernmentSponsored Enterprises: The Government’s Exposure to Risks, Washington, D.C., GAO. US Office of Federal Housing Enterprise Oversight, 2008, Report to Congress, Washington, D.C., OFHEO.

Book Review: Ending the Management Illusion: How to Drive Business Results Using the Principles of Behavioral Finance
By Hersh Shefrin, McGraw Hill: 2008, vii + 317 pages

Someone whose business responsibilities or research areas do not overlap with behavioral finance will find Shefrin’s latest book an interesting and efficient way to learn about how behavioral principles can be extended from the realm of investment decisions to the arena of corporate financial management. Most chapters, which can easily be digested in separate sessions, are concise and interesting. The text is written at the middle-management level and will also be useful for academics that teach or consult for executives. The first chapter of the text is one of the strongest. It uses excellent real-world illustrations to explain the psychological arguments and traits that induce educated, powerful individuals to make biased decisions and introduces Shefrin’s recommendations for improving the way corporations are managed. Improving access to financial information and financial literacy for all of a firm’s employees are cornerstones of Shefrin’s recommendations for ways to eliminate behavioral biases, and he provides several examples of how firms have successfully attained these laudable goals. The second chapter discusses the financial metrics the author would include in each employee’s training, but the level of discussion is a bit advanced, even for finance professionals if they do not use valuation principles on a regular basis.

The middle chapters of the text are its weakest components. Chapter 3, “Narrow Financial Focus on Projects and Financing: Traditional Approach”, contains a litany of stories of financial mis-steps made by managers at a variety of familiar firms that do not really serve to advance the educational recommendations the author is making. It also lavishes praise on firms that use managerial techniques of which the author approves in a way that sounds self-serving and diverts attention from the author’s main points. The discussion of the difference between Free Cash Flow to the Firm and Free Cash Flow to Investors remains difficult to parse after several readings. Chapter 4’s reliance on accounting measures of return and investment is puzzling given earlier discussions of the importance of distinguishing between accounting profit and actual cash flow. Chapter 5, “Involving the Workforce in Financial Planning”, is quite dense. Corporate managers could probably convinced of the importance of using simulated games and other experiential techniques to teach employees about the crucial value drivers of the business they are in without a long pretend “conversation” at an environmental firm. Chapter 6, “Motivating the Workforce Through Smart Carrots and Sticks”, returns to the text’s strongest feature, which is the link between business decisions and behavioral

137

138
psychology via a discussion on constructing appropriate bonus and incentive plans. This issue is intriguing for an academic, but the real-world focus here is quite narrow, with most of the examples being drawn from a single local firm. The chapter could have easily been combined with chapter 7, “Sharing Information Throughout the Organization”. This chapter on information sharing has a fascinating explanation of how hard the management of the equity research practice at Union Bank of Switzerland works to be sure that the analyst reports they issue are not subject to typical psychological glitches that could bias the recommendations. The text ends with chapter 8, “Integration: The Whole Ball of Wax”, that summarizes most of the author’s ideas and contains an invaluable table (8.1) that lists the various psychological pitfalls and biases human beings could fall prey to when making financial or managerial decisions. Throughout the text, Shefrin discusses various types of information that firms post on their walls to incentivize employees. Table 8.1 is a list of forces that should appear in every conference room in corporate America where managerial decisions are made.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

In summary, Shefrin’s latest literary effort is a worthwhile attempt to educate practitioners and academics about how the principles of behavioral psychology can be applied to corporate decision-making. Its main messages are 1) that sharing information, (especially financial information) with employees at all levels improves productivity and efficiency, 2) that all stakeholders should be educated about how financial information is presented and evaluated, and 3) that everyone makes biased decisions but that these biases can be identified and corrected. The rationale behind these recommendations is presented and defended using real-world examples to develop a treatise that is worthwhlle reading for anyone who needs to motivate, manage or train individuals to battle the “gremlins”, (a favorite term the author uses to describe typical behavioral glitches that bias human decisions) that confront us all every day. Andrea Heuson Professor of Finance University of Miami

Book Review
The Venturesome Economy
By Amar Bhidé, Princeton University Press: 2008, ix + 499 pages

 In The Venturesome Economy (2008, Princeton University Press), Amar Bhide offers an in-depth, refreshingly optimistic, and insightful counterpoint to the increasingly loud voices decrying the trend of economic globalization. Dr. Bhide gives particular attention to what he terms the techno nationalists – those who believe, “Leadership in science and technology gives the United States its comparative advantage” (p. 262). He describes techno nationalists as people who promote a national agenda to divert scarce resources in order to facilitate an increase in the number of native born engineers, mathematicians, scientists, statisticians, and researchers. The fundamental premise of the techno-nationalist argument is that the competitive advantage of the US and other developed economies lies in their scientific and technological acumen – that they are the world’s preeminent innovators because of their mathematic and scientific excellence. The ability to “out innovate” the rest of the world drives the economy. Techno-nationalists argue that if that competitive advantage erodes, so will the economy. Former Chancellor of the Exchequer in the UK, Gordon Brown, summed up the argument thusly (as Dr. Bhide reports on p. 264), “Every advanced industrial country knows that falling behind in science means falling behind in commerce and prosperity.” The techno-nationalists believe this is already occurring. The techno-nationalist position is supported by numerous statistics, some of which Dr. Bhide outlines in his book. He cites the work of respected scholars such as Clyde Prestowitz and Richard Freeman to outline the empirical foundation of the techno-nationalist movement. Freeman’s work is especially representative of the cause. He notes that in 1970 US educational institutions granted over half of all doctorates awarded in science and engineering fields. By 2001, European Union countries were granting 54% more of these

doctorates than the US. Over that same period, China went from granting almost no doctorates to granting roughly 1/3 as many Ph.Ds as the US. Freeman also notes the decline in publications and citations of US researchers. For instance, he quotes a New York Times article that states, “The share of papers (by US researchers) counted in the Chemical Abstract Service fell from 73% in 1980 to 40% in 2003” (p. 262). Freeman and Prestowitz’s arguments Dr. Bhide outlines are consistent with the much publicized recent studies showing the US school children slipping in international academic comparisons. According to the techno-nationalists, the result of this decline in the US’s scientific and technological competitive advantage is a weakening of the US economy. Here too, they offer empirical support. On this point, Dr. Bhide devotes more discussion to alarming forecasts than actual data. For instance, he cites a 2006 study by Alan S. Blinder in which Blinder claims that 40 million jobs in the service sector were at risk of being off-shored. Dr. Bhide writes of a similar report from Forrester Research in 2002 that estimated 3.3 million jobs in the service sector would be moved to other countries by 2015. Surely, the recent troubles of the Detroit automakers, along with the economy-wide meltdown of the second half of 2008, add fuel to the techno-nationalist fire. Dr. Bhide, however, rejects the techno-nationalist paradigm and rebuts the fundamental thinking behind this technologically protectionist movement. While it’s an injustice to the author to attempt to summarize his work in a single paragraph, I believe the following excerpt from the book captures the spirit (if only partly) of his message: …the United States should welcome more research from China and India, because an increase in the supply of highlevel know-how helps mid-level innovators

140
based in the United States develop products that increase productivity and wages in the United States (p. 59). His book is both timely and relevant. Regarding the continued prosperity of the United States, it is possible that there is no more important issue than economic globalization. The much publicized trend of outsourcing and off-shoring of US jobs – now extending far beyond the manufacturing industry – has created a groundswell of support for increased economic protectionism, further intensified by the collapse of the US auto industry. But the US’s attitude toward and response to economic integration on a global scale is something we must get right; we cannot afford to take the wrong position on this issue. For that reason, I believe Dr. Bhide’s book should be required reading for anyone seeking to join the debate on what policies the US should embrace related to free trade and international economic integration, regardless of whether the reader accepts or rejects Dr. Bhide’s thesis. The Venturesome Economy is a very well-written and thoughtfully organized two-part analysis of the effects of innovation and globalization on the US economy. The first part (Book I) builds a foundation for the second part and is primarily devoted to qualitative analysis stemming from an extensive study he performed involving 106 venture capital (VC)-backed businesses in the US. The first part is supplemented by numerous comparisons and contrasts of Dr. Bhide’s current work with his previous work focusing on firms from the well-known Inc. 500 List. The second part of the book (Book II) is devoted to a discussion of policy debates related to the tide of globalization, especially the debate regarding the importance of developing high-level innovation on the US soil (and by the US citizens). The book is written in easy to digest modules, which readers will appreciate. Readers can choose to read the foundation in Book I and then proceed to the prescriptions of Book II, or they can simply jump directly to the policy debates of Book II and use Book I as reference material. Similarly, each chapter is written to allow readers to spend as much (or as little) time as necessary. I found the chapter prefaces and conclusions to be concise and content rich, while the bodies of the chapters were persuasive and intriguing. Readers will benefit greatly from the bountiful mini-case studies throughout. Literally, almost every page in the book contains an example that relates the points Dr. Bhide is articulating to the real-world companies he studied. And although his study is intentionally short on econometrics (a point I return to at the end of my review), what he lacks in numerical analysis Dr. Bhide makes up for with enlightening qualitative analysis springing from countless hours of interviews with CEOs of important mid-level companies. In Book I, Dr. Bhide uses his study to make the case the

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

innovation and economic prosperity related to innovation are not as simple as many purport. He stratifies innovation into three levels: high, mid, and ground-level innovation. High level innovation is the kind that deals with molecular or physical advances. Mid-level innovation is the application of a high-level innovation. It represents an innovation per se, but not to the same degree. Ground-level innovation represents the application of a mid-level innovation, and again represents an innovation per se, but again, not to the same degree as the mid-level innovation. The example he gives is that the invention of the silicon-based micro-processor represents a high-level innovation, the creation of the motherboard is a mid-level innovation, and the development of the personal computer represents a ground-level innovation. Dr. Bhide argues in Book I, through his interviews and surveys of CEOs of VC-backed businesses, that mid-level innovators play a key role. They take the somewhat difficult to understand and difficult to apply high-level innovation (and other mid-level innovations) and begin the critical process of transforming that innovation into a marketable and beneficial product that helps to foster economic progress and prosperity. He goes on further to argue that mid-level innovators are relatively unconcerned with the location from which the high-level innovation emanates, so long as they are able to transform it into something marketable and beneficial to their consumers. He also argues that the transformation of high-level innovation into mid and groundlevel innovation is critically influenced by the willingness of consumers to try new technologies and by the interaction of mid and ground-level innovators with their customers. The first point highlights the importance of an economy filled with “venturesome” consumers who are willing to try new innovation. Such venturesome consumers attract all levels of innovation. On this point, Dr. Bhide believes the US represents an innovation Mecca. It is the country where all innovators seek to land their products and services. On the latter point, most mid-level innovators require a close relationship with their beta customers that allows for an iterative development process to fine-tune products to meet customer demand. This iterative development process requires, for the most part, geographic, language, and cultural proximity. In other words, Dr. Bhide is convinced that midlevel innovation is not currently and will not in the near future be something that can be moved off shore. These points present critical ramifications: 1.) having consumers that embrace new technology makes an economy the natural magnet for all levels of innovation, and 2.) mid-level innovation (the adaptation of high-level innovation into marketable products) is something that has been and will continue to beneficial for domestic economies. Regardless of where the high-level innovation comes from, the mid-level

WRIGHT— BOOK REVIEW OF THE VENTURESOME ECONOMY

141
the means and gusto to buy new, high-tech products. This willingness of consumers creates a magnet market for all levels of innovation. As all levels of innovation come from various sources, from Mumbai to Paris, mid-level innovators in the US will use the imported or domestically produced innovations to develop even more innovations, which will not be outsourced or off shored. (While he concedes ground level innovation and basic production can be and is off-shored, he is adamant that mid-level innovation is difficult to offshore.) In short, innovation breeds innovation. The innovation cycle creates jobs. But equally important, the innovation cycle creates products and services from which consumers, not innovators, reap the lion’s share of the benefits. Dr. Bhide even provides convincing statistical support for this rosy cycle of innovation and economic prosperity. For instance, he cites a study by Gene Epstein in which Epstein notes that from 2002 to 2006, employment in the two sectors most vulnerable to off-shoring had increased 7.7%, while employment in the overall service sector increased 4.5%. Dr. Bhide argues that pursuing a policy of protectionism and isolationism will do nothing but hinder this process. The subject of the US’s role in and response to global economic integration is so critically important to the future of the US’s prosperity that we cannot afford to get it wrong. Any decisions we make and any policies we enact must be done only after careful and measured investigation. Given the importance of the subject, I believe Dr. Bhide’s work would have benefited from a more rigorous econometric approach to his study. He is very clear about why he chose not to do that. But such an analysis used as a supplement could not have hurt the veracity of his book and likely would add increased credibility to his conclusions. In spite of this one perceived deficiency, which was a conscious choice by Dr. Bhide, I found the book well written and highly informative and believe it makes a substantial contribution to a critical topic that is currently being hotly debated in the US. Colby Wright Assistant Professor of Finance Central Michigan University

innovation that results will benefit the country where it occurs, both by creating jobs and by providing its citizens with a beneficial new product or service. Most of Dr. Bhide’s discussion on this topic is derived from his own work with the CEOs of VC-backed businesses, which he describes as being almost universally mid-level innovators. Further, it is important to recognize that much of Dr. Bhide’s conclusions on the topic are derived from qualitative analysis and critical thinking. There is little in the way of rigorous econometric analysis. Nevertheless, his reasoning and qualitative evidence is sound and convincing. These keystone points provide support for his argument in Book II that the US should not fear high-level innovation from foreign countries or from foreign workers in the US. Instead, the US should embrace high-level innovation from all sources without diverting scarce resources to attempt to secure a position of superiority in the creation of those highlevel innovations. The US is best served by pursuing a policy agenda that (a) invites high-level innovation from all locales and nationalities and (b) that facilitates mid-level innovation within the US itself. Regarding the second point, facilitating mid-level innovation, Dr. Bhide provides evidence from his study that Ph.D.’s are not necessary for mid-level innovation. He even quotes one CEO who states frankly that most Ph.D.’s at his company consider the attainment of their Ph.D. to have been a career mistake (it was unnecessary in performing the roles they fill inside the company). Dr. Bhide ultimately concludes that fostering an environment that invites high-level innovation into the country and encourages mid-level innovation within the United States likely does not require any significant policy initiative to increase the number of engineers, mathematicians, scientists, statisticians, and researchers. In fact although I don’t see this explicitly stated in the book, I come away feeling as though Dr. Bhide is suggesting a relatively lassies faire rebuttal to the techno nationalist push to address the shrinking number of American born hard scientists in the United States. The overarching message of the book is one of optimism. Perhaps the most important factor in enjoying economic prosperity spurred by technological innovation is to simply have a country full of venturesome consumers – people with

142

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008

With the previous issues of JAF, we began publishing Financial Puzzles by Stewart C. Myers. This set (problems 14-15) is the fifth installment. The solutions for the previous set are also given below. —The Editors

Financial Puzzles
Stewart C. Myers

14. This problem considers how corporate income taxes can affect IRRs for capital investment projects. Assume a constant future tax rate T. (a) In what circumstances is the IRR of after-tax cash flows exactly equal to the IRR of pre-tax cash flows multiplied by 1 – T? (b) What tax system always generates the same IRR for both pre-tax and after-tax cash flows? (c) What tax system always generates the same IRR for any pattern of after-tax cash flows? Hint: the IRR is negative. 15. A convertible bond emerges from call protection with the conversion option well in the money. The issuing company can call and force conversion now, but does not do so. Are the convertible bondholders necessarily better off because of the decision not to call? Does the price of the bond necessarily increase? Assume that post-conversion dividends would equal the coupon payments on the bond, so that there is no cash-flow advantage or disadvantage from conversion.

Proposed Answers for Problems 11 – 13
In the Fall/Winter, 2007 issue of the Journal of Applied Finance, p. 131 11. Book return equals the true rate of return (IRR) in the following cases. (a) Book depreciation always equals economic depreciation. Economic depreciation would be calculated as the change in the PV of project cash flows using the IRR as the discount rate. (b) The firm settles into steady-state growth and the growth rate equals the IRR. Condition (b) was first derived in Solomon and Laya (1967).1 12. Think of a project balance sheet, with PV(Revenue) on the asset side and PV(Fixed costs) and PV(Variable costs) on the liability side. NPV = PV(Revenue) – PV(Fixed costs) – PV(Variable costs). Project A’s costs are fixed. Project B’s costs are fixed, but with diversifiable noise added. Project C’s costs are variable, with  = 0.5. Net cash flows and NPV are safer (lower  ) for project C than for A and B. C’s variable costs are a partial hedge against C’s uncertain revenues. A’s and B’s costs do not provide such a hedge. A project amounts to a long position in PV(Revenue) and short positions in PV(Fixed costs) and PV(Variable costs). The overall position is safer (lower  ) if the short position is in a positive- asset. Thus NPV(C) has a lower  and a lower

E. Solomon and J. Laya (1967), “Measurement of Company Profitability: Some Systematic Errors in Accounting Rates of Return,” in A. A. Robichek, ed., Financial Research and Management Decisions, John Wiley & Sons, Inc., New York.
1

142

MYERS — FINANCIAL PUZZLES

143

cost of capital than NPV(A) and NPV(B). C is the more valuable project. 13. (a) NPV > 0 when x is close to zero. If the institutional investor only has to put in $1 for a $100 million project, then the investor is almost certain to earn the return r = RCAP immediately. The return r includes a risk premium, but the payoff is risk-free, so NPV > 0. NPV may be even larger if the first month’s cash flow is greater than 1 + RCAP and the investor can keep all of that month’s cash flow. NPV < 0 when x = 1. When x = 1, the institutional investor contributes $100 million to a project with PV = $100 million. However, the investor may not get all future cash flows. There is another residual claim that gets all subsequent cash flows if and when IRR reaches RCAP. This residual claim has positive value, so the value remaining to the institutional investor must be less than $100 million. (b) NPV at first increases as x increases and then declines. NPV > 0 for small values of x, for the reasons given in (a). But a larger x also means that the investor has to put in more money and wait longer to earn IRR = RCAP. The risk that the investor will not earn RCAP increases, and value finally declines. NPV < 0 at x = 1. (c) Once the project is up and running, poor performance can actually benefit the institutional investor. Suppose that IRR after 60 months is just shy of RCAP. A large project cash flow for month 61 could be just enough to achieve IRR > RCAP and put the investor out of the game. A slightly lower cash flow could leave IRR < RCAP, assuring the investor of cash in both months 61 and 62 and possibly in later months. Two or more cash flows can be better than one, even if the first cash flow is not as high as it could have been. The ideal outcome for the investor would be a series of cash flows, each as large as possible without bringing IRR > RCAP. This would be a disappointing outcome ex ante, at the start of month 0, but not ex post, when earlier cash flows are already money in the bank. (d) Financing from the institutional investor could require RCAP > r, in order to give the investor additional upside to offset the negative NPVs noted in (a). The value of the investor’s claim would have to be calculated from a Monte Carlo simulation of certainty-equivalent cash flows. The simulation is necessary because the test for IRR > RCAP in any given future period is path dependent. The residual claim that gets all cash flows after the institutional investor is out of the game is also a path-dependent option.

JOURNAL OF APPLIED FINANCE
STYLE NOTES FOR PROSPECTIVE AUTHORS
Mission Statement: The mission of the Journal of Applied Finance is to disseminate information and to foster debate on the practice and pedagogy of finance. The journal is devoted to the publication of original manuscripts that are accessible to a broad audience, including practitioners, academics as teachers, and students. Of particular interest are manuscripts with an applied orientation falling into one of the following broad categories: traditional research articles (empirical, practical, survey, and synthesis), clinical studies (characterizations of real world situations using unique sources of data), and education (wellmotivated, scientifically sound manuscripts that represent major contributions to the field of financial education). Submission Procedure: Manuscripts should be submitted electronically as a PDF file at www.fma.org-Journal of Applied Finance. Each paper must be accompanied by a submission fee for manuscript evaluation: $100 for FMA members, $200 for non-FMA members, and $130 for doctoral students who are not FMA members. If paying by check, please make checks payable to Financial Management Association. If you prefer to pay by credit card, please include in your submission email the following information: type of credit card, cardholder's name, credit card number, and expiration date. The non-member submission fees include a one-year membership in FMA for the submitting author.

       

Manuscripts are evaluated anonymously. Names of authors should not appear on the article itself. Attach a separate cover page that includes the title, authors, and title and affiliation of each author to one copy of the article. Double space the text with ample margins. Authors should ensure that their identity does not appear under the “Properties” of the file they are submitting. The cover page of the manuscript should show only the title of the paper and a short, one-paragraph (approximately 100 word) abstract of the articlethat provides a brief overview of the paper. Avoid tedious mathematical expressions. When algebraic terms do appear in the text, accompany them with a clear explanation. Each equation should be numbered consecutively, with the number in parentheses and flush with the right margin. Place derivations and proofs in an appendix. When submitting a paper for review, please provide supplemental sheets showing all steps in algebraic derivations so that the reviewers do not have to re-create them. Tables and figures should appear on separate pages labeled in numerical order and grouped at the end of the text. Label tables at the top and follow the heading with a description of the table in sufficient detail so that it is capable of standing alone. Label figures at the bottom. Include marginal notation in the article for the approximate placement of all tables and figures. Minimize extensive content footnotes. When preparing accepted papers, place all footnotes, double-spaced, at the end of the manuscript. Place references in an unnumbered, alphabetical list at the end of the manuscript. Provide all relevant publication information available (i.e., season/month, year, city and state, author(s) full names, etc.). Examples of references are provided below: References
Baldwin, Carliss Y., 1991, “The Impact of Asset Stripping on the Cost of Deposit Insurance,” Harvard Business School Working Paper 92-053 (December). Commerce Clearing House, 1993, 1994 U.S. Master Tax Guide, Chicago, IL. Weston, J. Fred, 1994, “A (Relatively) Brief History of Finance Ideas,” Financial Practice and Education 4 (No. 1), 7-26. Myers, Stuart C., 1993, “Finance Theory and Financial Strategy,” in D.H. Chew Jr., Ed., The New Corporate Finance, New York, NY, McGraw-Hill, 90-97. Smith, Clifford W. Jr. and Charles W. Smithson, 1990, The Handbook of Financial Engineering, New York, NY, Harper Business.

 

Cite references in the text by citing the author(s) name(s) and then the year of publication in parentheses. Authors of accepted papers must supply an MS Word copy of their article. Any questions about disk preparation should be directed to the Managing Editor, Financial Management Association International, University of South Florida, College of Business Administration, Tampa, FL 33620-5500, TEL 813-974-2084, FAX 813-974-3318, E-mail:[email protected]