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: publications@fma.org; 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

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

Rizzi Roundtable 97 University of Rochester Roundtable on Bankruptcy and Bailouts: The Case of the US Auto Industry Panelists: Thomas Jackson. and Sotiris K.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. 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. James Brickley. and Paul Sengmueller Michel Rakotomavo Morris G. Staikouras Ryan Garvey and Fei Wu Anne Dorn. Shaffer Edward C. Myers . Jean L. Gulnur Muradoglu. and Clifford Smith Moderator: Mark Zupan Interview 116 Pioneers in Finance: Vernon Smith Interview Terrance Odean and Betty J. Lawrence Practitioner Contribution 84 Behavioral Basis of the Financial Crisis Joseph V. Joel Tabas. and David R. Hersh Shefrin. Simkins Case Study 124 The 2008 Federal Intervention to Stabilize Fannie Mae and Freddie Mac W. Daniel Dorn. Danielson. Heck. Charles Hughes.

In this issue our focus is on behavioral finance. Professor Smith shares his perception of how experimental economics and behavioral economics are related. especially roundtables. We would like to encourage our readers to be actively engaged in these types of submissions. 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. and Paul Sengmueller. 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”. 4 . Our continued success though depends on our ability to attract submissions in each of these categories. 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. Shaffer (“Shareholder TheoryHow Opponents and Proponents both Get It Wrong”). We thank the University of Rochester for sponsoring the roundtable. The behavioral finance theme continues in our “Pioneers of Finance” interview feature. Our roundtable features a time topic: Bankruptcy and Bailouts: The Case of the US Auto Industry. we are more than happy to work with you to develop submissions in these categories. It is based on a panel discussion held at the FMA-Europe meetings in Prague. who also authored the piece. and “Why Do People Trade?” by Anne Dorn. and Sotiris Staikouras. “The Effects of Institutional Risk Control on Trader Behavior” by Ryan Garvey and Fei Wu. 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. case and clinical studies. The case study also continues the subprime crisis theme from our first issue. We thank Terry for his contribution. In This Issue As we mentioned in our first issue. The second is by Amar Bhidé. recipient of the 2002 Nobel Prize in Economics.4 Letter from the Editors This is our second issue as editors of the Journal of Applied Finance (JAF). Daniel Dorn. Scott Frame provides insights into the Fannie and Freddie debacles that only an insider can provide. The panelists included Werner De Bondt. surveys. One is titled Ending the Management Illusion: How to Drive Business Results Using the Principles of Behavioral Finance written by Hersh Shefrin. As a Federal Reserve insider. In this interview. and book reviews.” This article also continues a theme from our first issue on the subprime crisis. The panelists feature several prominent auto industry executives and faculty members from the University of Rochester. another prominent author who is an authority on strategy. As your editors. 2008. one of the pioneers in behavioral finance. Heck. case and clinical studies. This issue concludes with the solution to the Financial Puzzle that appeared in the Fall/Winter 2007 issue. The review was written by Andrea Heuson. The interview was conducted by Terry Odean and Betty Simkins. In addition. surveys. and interviews. There is also a provocative article on shareholder theory by Morris G. and David R. Mark Zupan for moderating. along with two new puzzles by Stu Myers. Our academic contributions also include several articles on trading behavior: “The Long-Term Value of Trade Informativeness” by Michel Rakotomavo. roundtable discussions. The book is titled The Venturesome Economy and is reviewed by Colby Wright. Danielson. interviews. and Don Chew for editing it. We will strive to keep these same features consistent across issues. Professor Vernon Smith. Gulnur Muradoglu. Jean L. 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. one of our goals is to have one or two themes for each issue. We received positive feedback on the new layout of JAF with sections on academic and practitioner contributions. Hersh Shefrin. Our book review section features two books. 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.

corporate restructuring. and dividends and share buybacks. Simkins Williams Cos. Wise Chair Oklahoma State University Email: ramesh. we would like to highlight themes/topics we are considering for future issues. Professor of Business Oklahoma State University simkins@okstate. We welcome your contributions and any suggestions you may have for JAF.rao@okstate.com .5 In Closing In closing. Sincerely.edu Betty J. Rao Paul C.edu Charles Smithson Founding Partner Rutter Associates LLC csmithson@rutterassociates. These include valuation. Ramesh P.

6 .

Tversky and Kahneman. UK. CA. We describe the move from the standard view that financial decision making is rational to a behavioral approach based on judgmental heuristics. departures from neoclassical finance theory. Hersh Shefrin is a Professor of Finance at Santa Clara University in Santa Clara. Werner De Bondt and Werner De Bondt is a Professor of Finance at DePaul University in Chicago. IL. prospect theory. Daniel Kahneman. mental frames. Robert Shiller. The paper is written for a wide spectrum of readers. Five or six years later. 1989). economists model behavior in terms of rational individual decision-makers who make optimal use of all available information. Our paper aims to provide an over-arching view of the field. and then 7 . Most everyone agrees that it is problematical to discuss these dramatic episodes without reference to investor psychology. among others. A new class of asset pricing models. The term “behavioral finance” has a variety of meanings. and examines what the consequences are for financial markets and institutions. biases. with spillover effects on marketing.Behavioral Finance: Quo Vadis? Werner De Bondt. and sometimes explains differences in policy recommendations on such issues as financial regulation.. Staikouras Behavioral finance endeavors to bridge the gap between finance and psychology. the bubble in Japan during the 1980s. with consequent excess volatility in stock and bond prices. Behavioral finance emerged as a field in the early 1980s with contributions by. this small group of financial economists was meeting regularly with psychologists — including Paul Andreassen. It is a summary of a panel discussion. Now an established field. the Asian crisis of 1997. behavioral finance studies investor decision processes which in turn shed light on anomalies. is proposed. Examples include the stock market crash of 1987. Investment portfolios are frequently distorted. Hersh Shefrin. biases. Now behavioral finance is poised to replace neoclassical finance as the dominant paradigm of the discipline. and Sotiris K. mental frames. including financial practitioners. judgmental heuristics.e. Soon. or the privatization of social security. From its beginnings as a fringe movement. Proponents of behavioral finance argue that poorly informed and unsophisticated investors might lead financial markets to be inefficient. the dot-com bubble. There is ample evidence that the rationality assumption is unrealistic. Gulnur Muradoglu. Richard Thaler. It begins by reviewing the foundations of finance and it ends with a discussion of the future of behavioral finance and a self-critique. i. which adds behavioral elements to the standard framework. and the financial crisis of 2008. Lola Lopes. and new theories of choice under risk. Meir Statman. This paper is the summary of a panel discussion. Hersh Shefrin. game theory. and SP/A theory has provided new foundations for financial economics. political science and law. the demise of Long-Term Capital Management. UK. behavioral finance moved to a middle-ofthe-road movement. and Amos Tversky — at the Russell Sage Foundation in New York. corporate governance. management. David Dreman. Traditionally. reviews some fundamental questions. and others on bounded rationality. Gulnur Muradoglu is a Professor of Finance at Cass Business School in London. It had immediate impact worldwide including emerging markets (Muradoglu. experimental economics. the National Bureau of Economic Research began organizing semi-annual meetings. Staikouras is a Senior Lecturer in Finance at Cass Business School in London. Behavioral finance studies the nature and quality of financial judgments and choices made by individual economic agents. The debate between neoclassical and behavioral finance is wide ranging. The path-breaking work of Herbert Simon. Sotiris K. It begins by examining the current state of finance.

JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 I. Modern (or neoclassical) finance is the paradigm that has governed thinking in academic finance since the late 1950s.8 introduces behavioral concepts. The first pillar is the concept of “beautiful people”. The main question is: What do people do and how do they do it? The research methods are mostly (but not exclusively) inductive. The real economy is where goods and services are produced and consumed. and the financial institutions involved in them (e. the heuristics and biases literature pioneered by Tversky and Kahneman (1974) and Kahneman and Tversky (1979). and so on).” The psychology of decision-making can be explored in various ways. The world of finance is mostly seen as a sideshow. field studies.. What is the role of institutional factors such as market organization. in neoclassical finance? To a first approximation. 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. Consider. there is none. the banking system. 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. insurance companies. His writings say that the basic functions of finance are the same. There are no opportunities left for rational arbitrage. is what much of finance is about. various asset pricing theorems are derived. Section IV discusses key building blocks of the behavioral approach. Salih. respectively. competitive. It flows from a philosophical tradition (the 18th century Enlightenment) that aims to reconstruct society with individual rational action as its centerpiece. rating agencies. Bayesian updating. complete markets. II. 1989) and they did not run experiments.g. pension funds. Modern finance is built on two pillars. risk aversion. saving and investing. and Mercan. He spoke about institutions as potential distortions. Section VII concludes. organizational architecture. Section V explores some new ideas in behavioral asset pricing and behavioral corporate finance. 2005). defined as logical. and rational expectations. perfect. modern finance theorists rarely administered surveys (Muragdoglu. price equals value. but the link with the Miller-Modigliani theorems and the work of Ronald Coase is obvious. Even so. A quarter-century ago. Certainly. and why banking in Switzerland is different from banking in Egypt. the theory of value. autonomous agents characterized by expected utility maximization (over time). etc. What does change is the technological and regulatory environment. Based on these two concepts as well as the mutual adjustment of demand and supply (plus an assortment of auxiliary assumptions). The second pillar is the concept of “beautiful markets” i.) and organize them into a number of “superfacts. Of course. What Is Behavioral Finance? Behavioral finance does not assume rational agents or frictionless markets. soon realizes that the central unifying concept is asset valuation. the process may take time. contract design.1 In the past. Anyone who contemplates the functions of finance. “Behavioral finance: Quo Vadis?” Sections I and II review modern and behavioral finance. Miller’s comments were often formulated in the context of regulatory barriers to financial innovation. Investment portfolios are mean-variance efficient. Of course. we usually make a distinction between the two. Only systematic non-diversifiable risk is priced. money management firms. and where wealth is created. liquid. 1 . In equilibrium. and risk management. tax systems etc. 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. What Is Finance? Let us start by defining finance. Rational agents work around institutional frictions and thereby render them immaterial to market outcomes. That is why banking in 2008 is different from banking in 1908. always and everywhere. most effort went into cognition. It suggests that the institutional environment is vitally important. Still. regulatory framework. although this is starting to change(Muragdoglu. Conditional on what is known about the future. Behavioral researchers collect “facts” about individual behavior (based on experiments. all agents reach their optimum. Robert Merton’s views are similar. surveys.” That economic and financial intuition is fragile may clash with our aspirations for mankind. depending on the problem-at-hand. Section VI provides a self-critique.e. many financial economists believe that the swaying power of data cannot match the power of logic. 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. though ultimately neutral mutations. Merton Miller made this type of institutional arbitrage a favorite lecture theme. and comparisons of price and value. Even though the real economy and finance are linked. Behavioral finance is the study of how psychology impacts financial decisions in households. for instance. Section III briefly delves into the efficient markets literature. The starting point is bounded rationality. 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. markets and organizations. finance serves important functions such as the payment system. the pooling and transferring of funds.

Second. behavioral illuminate how the human mind works. behavioral procrastination. In Belgium. we find predictable variables that are critical. Specific errors many studies rely on surveys or depend on context. 3 Our economist friends emphasize incentives. Everyday we learn more about committee decision-making (e.g. As president of the American Finance Association. In plus the well-known fact that people tend to stick with the addition. donated. The this finding appear in Nudge. etc. Michael Jensen asked that we break open the black box called the firm. Shefrin (2005) focuses on behavioral corporate finance. 2002) and Thaler (1993). SHEFRIN. evidence. it studies emotion (mood. 2 This type of research is especially relevant to the study of organizations. 1999). does not wipe out inefficiencies.. experiments teach us that subjects do not update beliefs in since most people’s investment horizons are short. Daniel default solution is the opposite. Here. & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS? 9 questions such as: How do people think? How do they decide? difference? For an answer. from multiple sources. 2008a). Price and Value 15% whereas in Belgium it is over 95%.2 There are three default is a signature. Dreman (1995). the law psychology (especially herding behavior). The U. matched with market anomalies (when social Investor sentiment matters. laboratory research Intuition is fragile.e. for example. only that it experiment “at home. their trading records confirm fiascoes is informative. It also explains why researchers also make use of conventional market-level price financial judgment is fallible. affect) and social status-quo. However. Apart from agency and asymmetric information problems. and the pros and cons of bureaucratic formalities and red tape. 1993. often so convincing? One Behavioral finance is based on predictable mistakes such as reason is that “good” behavioral overconfidence in judgment.K. the role of top managers in the creation of corporate wealth. financial behavior. Widely-shared misconceptions interaction allows fine-tuning) produce a powerful body of (that may be self-reinforcing) cause transient price bubbles.3 Here too.g. more sophisticated traders will These works lay emphasis on investment and asset pricing.. arbitrage Bayesian fashion (De Bondt.5 Third.” Further. Note that it permits any reader who doubts preferences. 2005. trading records) in a natural research examines psychological mechanisms which environment (e. 1998. 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. instance.. there Why is behavioral research is a catalog of biases. research depends on support three main building blocks.DE BONDT. wishful thinking. i. 2003). and volume data. 1985).K. investor overreaction. Note that in either country all it takes to modify the Shefrin (2001a. Numerous specific applications of reversals in share prices (De Bondt and Thaler. at the market level. the main insight is that the systematic errors of unsophisticated Decision anomalies (in the laboratory). De Bondt (2002. Certainly. In case of a fatal car accident in the U. This “one-two-three punch. 2002). Odean. First. Take. even though naive investors may push security prices away from intrinsic values. myopia.4 classes of findings. the finance are described in Barberis and Thaler (2003). Lastly. there are behavioral costs that obstruct the corporate value maximization process. but are systematic nonetheless. processes shape decision outcomes. the study of Regardless of what investors say. we look to the decision process Current work continues to draw on cognitive research. and market results appear to Thaler and Cass Sunstein (2008). since it guides us to decision process the bias.” we believe. environment). One striking example has be connected.g. boards). large and small. the driver’s organs are et al. 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. Behavioral finance is contributing to that effort. participation rate in organ donation is approximately III. when asked. rational arbitrage matters too but. and limits to is not alleged that financial the results to replicate the arbitrage. 1985). Muradoglu. What explains this Milton Friedman (1953) and Eugene Fama (1965) argue that. For namely sentiment. can break.e. 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. a book authored by Richard laboratory. even if noise traders create a great deal of risk. to do with organ donation (Johnson and Goldstein. intuition is broken. i. 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. superior to what is typical for research in modern finance. 5 . The observe individual behavior (e. MURADOGLU. Certainly. (2002).. matched with investors (“noise traders”) create profit opportunities for anomalies in the behavior of individual agents (in a natural experts. 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.

153-4) observes how “day-to-day fluctuations in the profits of existing investments …tend to have an altogether excessive. investing in low PE stocks is a profitable contrarian strategy. Consider the share price movement of Royal Dutch/Shell Group. 1999. 1999.. 1988) offers systematic evidence of market volatility. has been examined by many scholars.g. Roll (1984. There are other widely documented phenomena which are difficult to reconcile with efficient markets. Dreman (1977. • Earnings momentum: Stock prices “underreact” to annual and quarterly announcements of corporate earnings causing a post-announcement drift in returns.. 1990. 7 Other price-scaled ratios. Grinblatt and Han (2005) and Frazzini (2006) suggest that momentum can be explained by the disposition effect. 1990) were among the first to establish this effect. An example is the case of “Siamese twins” stocks (Rosenthal and Young. For a critique of these findings. 1993) work on equity volatility. It is probable that the phenomenon is also partly explained by earnings momentum. their prices are not predictable on the basis of publicly available information. momentum is often followed by reversals.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. markedly for firms with low institutional shareholdings (Bartov et al.” This comment anticipates Shiller’s (1981. but the research goes back to Ball and Brown (1968).. See Chen and De Bondt (2004) and De Bondt (2008b) for details. There. e. competitive rational arbitrage guarantees that. not associated with information arrival. e. and many others who believe that stocks with low price-to-earnings (PE) ratios are undervalued and stocks with high PE ratios are overvalued. 2000). A vivid example was the 22. (1995) on share price repurchases. and the gradual dissemination of news. the spread between the return on equities and fixed income US government Corporate news that is not directly related to earnings also predicts returns. investor underreaction. • Equity premium puzzle: Historically. Based on market efficiency.g. 6 Trends are also visible in stock indexes of US industries and investment styles.8 Yet. these authors suggest. Bernard and Thomas (1989.. the book-to-price ratio. Basu (1977). where Royal Dutch stock trades in the US/Netherlands and Shell stock trades in the U. pp. Price momentum may be due to positive feedback trading.g. at all times. there is also evidence that noise trader risk is a significant impediment to arbitrage (Scruggs. 1999). Jegadeesh and Titman (1993. De Bondt and Thaler (1987) and Fama and French (1992). the arbitrage activity of rational traders will prevail (over irrationals) as long as securities have close substitutes. dividends) cannot possibly account for the observed aggregate price movements. Poterba and Summers (1988) obtain analogous reversals for national stock price indexes. are overly pessimistic about the prospects of low PE stocks. 2000). Nicholson (1968). 1993. Lee and Swaminathan. In other words. past winner stocks remain winners. (1995) on dividends or Ikenberry et al. First.e. Muradoglu. and even absurd. Among others. The efficiency of security prices has also been challenged by Graham (1949). see Fama (1998). and past losers remain losers. See. influence on the market. 8 . the inflow of funds causes prices to rise further.6% drop in the Dow Jones Industrial Average on October 22. Investors. finance academics have made two main assumptions about security valuation. beyond one year. 2007). Michaely et al. Small firms feature more momentum than large firms (Jegadeesh and Titman.7 • Price momentum: For holding periods up to one year. and second. Over the decades. it is suggested that fluctuations in economic fundamentals alone (e. i. This contradicts “the law of one price. Since the crowd avoids them.10 find it worthwhile to correct any mispricing. 1987. a concept introduced by Shefrin and Statman (1985) whereby investors sell winners too early and hold losers for too long. Grinblatt and Moskowitz. also forecast stock returns. the efficient markets hypothesis. Fama (1965) argued that the competitive activity of arbitrageurs will bring security prices into line with fundamentals. Behavioral finance has provided evidence which contradicts the notion of efficient markets. European and emerging markets exhibit similar patterns (Rouwenhorst. Froot and Dabora. (1991) show that during periods with “no” major news announcements equity prices experience some of their largest one-day moves. 1998. Consider the following examples: JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 • Price volatility that is not linked to news: Cutler et al. 1980). Thus a ratio of 1.g. See.. e. when large increases in stock prices pull in new investors. Froot and Dabora (1999) and Lamont and Thaler (2003) find that the relative price ratio ranges from 15% overvalued to 35% undervalued.5 (price of Royal Dutch relative to Shell) should have been achieved in order for the prices to reflect fundamentals.. The two companies’ original merging interests were on a 60:40 basis for Royal Dutch and Shell respectively. securities have an intrinsic value based on their fundamentals. That is. 2000). this perspective.K. 2001) and others show trends in share prices of individual stocks. Thus. and in stock indexes of foreign equity markets.” In relation to these stocks. 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. • Excess volatility: Keynes (1936.

1993. We next describe each of these building blocks in greater detail. rational information traders exploit the behavioral inconsistencies of irrational noise traders. Reinganum. loss averse investors may be inconsistent towards risk.” 9 . In the next section. By sentiment is meant investor error. Key Building Blocks Behavioral finance is based on three main building blocks. Proponents of behavioral finance suggest that there are limits to the process of arbitrage. i. In neoclassical finance. and in so doing lead prices to be efficient. but blame failure on bad luck. • Size and calendar effects: Small firms earn anomalous high returns. (1998. Roll. Keim. and are not sufficiently adjusted afterward.. Behavioral finance offers promising. Investor overreaction is partly rooted in representativeness. that is easily recalled from memory or that corresponds to a future scenario that is easy to imagine. and the turn of the year (Siegel. we describe some of the key psychological building blocks of the behavioral framework. and the value function is defined over gains and losses. perhaps because it is was available first. affect theory. change of process theory. Barber and Odean. 1983. regret theory. behavioral preferences. 1983. • Overconfidence implies that individuals overvalue their knowledge or abilities. forthcoming). The main point of the above examples is that business fundamentals alone do not explain the structure and dynamics of asset prices. the value of each outcome is multiplied by a decision weight. Tversky and Kahneman (1992) argue that people weight losses twice as much as gains of a similar magnitude. the source of the problem is cognitive. 1983). Psychology shows that people’s beliefs are often predictably in error. people attribute success to their own skills. Errors originate at the level of the individual but can manifest themselves at the level of the market. and self-control theory. 2001) suggest that investors suffer from a combination of overconfidence and self-attribution bias. they underreact. 1998. For instance. In many cases. That is.. People may prefer Fellner (1961) introduces the concept of decision weight to explain ambiguity aversion. the problem is a function of how people think. and as a result prices need not be efficient. MURADOGLU. 1994).e. Investor preferences constitute the second key element of financial models. overconfidence may lead investors to underestimate risk or to overestimate their ability to beat the market. By and large. “Conservatism” is closely related.e. they lead to systematic error. IV. People are likely to remember events that receive a lot of attention by the media and this influences their behavior (see. In this regard. Some psychological mechanisms have been modeled as heuristic rules of thumb. Investors who regard recent time-series trends as representative of an underlying process are vulnerable to extrapolation bias.DE BONDT.. For example. Overconfidence bias may also cause excessive trading. There is also ample literature on calendar effects. • Anchoring is a form of bias where beliefs rely heavily on one piece of information. there are curious patterns in equity returns related to weekends.9 Other behavioral preference frameworks include SP/A theory. The best known is prospect theory. It has many consequences. Daniel et al.g. However. A few biases in beliefs are described below. e. 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. Decision weights are inferred from choices between prospects much as subjective probabilities are inferred from preferences in the Ramsey-Savage approach. For instance. plausible alternative explanations for some of these phenomena. the turn of the month. developed by Kahneman and Tversky (1979) to describe the manner in which people systematically violate the axioms of expected utility theory. namely sentiment. Behavioral preferences capture attitudes about risk and return which do not conform with the principles of expected utility theory. heuristics perform well but. Muradoglu and Onkal. 1985) since it implies that the representative investor is exceedingly risk-averse. The “gambler’s fallacy” is also connected to representativeness but leads investors to make unwarranted predictions of reversal. & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS? 11 securities has exceeded 6%. e. sometimes. The following list describes some of the most important features of behavioral preferences: • Loss aversion portrays investors’ reluctance to realize losses. there are several behaviorallybased preference frameworks. SHEFRIN. investor forecasts may anchor on the price at which they bought a security (De Bondt. and limits to arbitrage. Prospect theory differs from expected utility theory in that probabilities are substituted by decision weights.g. i. • Availability bias means that investors overweigh information that is easily accessible. Unlike what is assumed in neoclassical finance. 1986. not final wealth. It is difficult to reconcile the magnitude of this premium with modern asset pricing theory (Mehra and Prescott. • Representativeness is overreliance on stereotypes. Kahneman and Tversky (1979) state: “In prospect theory.. decision weights are not probabilities: they do not obey the probability axioms and they should not be interpreted as measures of degree or belief. Investors may place excessive weight on past information relative to new information.

Narrow framing series of financial market poses dangers. the people exhibit self-control with respect to saving behavior. where mainly elderly investors have a Shefrin (2005. Academics. • Myopic loss aversion combines time horizon-based Mispricing has been the focus of many studies. margin payments. a basic tenet of modern account. 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). research has uncovered a framing. 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. 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.e.12 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 to avoid risk in order to protect existing wealth. 2007). based on this idea. to self-control when they explain how investors deal with the thereby presenting a coherent. Behavioral Analogues to Neoclassical losses twice as much as gains. • 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. integrated framework for impulse to hold onto losing investments for too long (see describing how markets are impacted by psychological Lease et al. 2000). Such regrets evaluate financial outcomes (Henderson and Peterson. • Self-control refers to the degree to which people can Asset pricing theory and corporate finance are in the control their impulses. etc. current investors will bet on good assets.b) argues that in the future finance will preference for dividends. when their time horizon is short than when it is long (Haigh (2002). 11 Yet. Investors are more averse to risk and Liu (2001). broadly understood as erroneous beliefs about future cash flows and risks (Baker and Wurgler. investors prefer securities with actions of noise traders (i. arbitrage can be risky (Shleifer. At the moment. Shefrin and Statman (2000) is always carried out. and List.. Koening (1999) argues that wealth in three mental accounts: current income. or Mitchell et al. As a result. limited arbitrage greatest from the current assets. . risks. which in turn could possibly wealth. may also encourage investors to hold on to loser stocks Shefrin and Thaler (1988) assume that people categorize (Shefrin and Statman. and policy makers have all contributed 11 In the traditional approach. investors want to avoid the regret • Mental accounting refers to how people categorize and of having sold shares that later went up in price. investors judge a new gamble via its to the question of what caused the crisis. Shefrin and Statman (1985) refer combine the best of neoclassical and behavioral elements. usually called narrow return. Sentiment impacts the prices of all assets. asymmetric distributions that combine downside based on fundamental information) may cause prices to be inefficient.” Smith’s observation captures regard. and drives the 10 In The Theory of Moral Sentiments.10 dividend puzzle if. and drives the difference plays a crucial role in income account and smallest between what behavioral and behavioral asset pricing. ex ante. consider the global financial crisis that began in 2008. 1976. yet may superior decision (ex post). Benartzi and Thaler (1995) argue that the size of the equity premium suggests that investors weigh V. in the MMA version. behavioral approach is somewhat piecemeal. the account. and future income. In this enjoy when we rise from a worse to a better. Therefore. the propensity to consume is Finally. Schill and Zhou (2001). protection (in the form of a floor) with upside potential. 1992). 2005). For this reason. taxes. It trigger some sort of herding is furthermore assumed that Sentiment impacts the prices of all behavior. traders with biased beliefs. in order to avoid regret. not non-normal. 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. Regret helps to explain the assume risk in order to avoid sure losses. media. e. In the SMA to exploit mispricing. phenomena.. whereas the Shefrin and Statman (1984) develop a theory of dividends neoclassical approach is more coherent and integrated. In a New York Times contribution to total wealth. To from the future-income repeat. and that they evaluate their APV and SDF-based Pricing portfolios on an annual basis. Thaler and Shefrin (1981) analyze how process of becoming behavioralized. for empirical evidence). the modern notion of loss aversion.. investors integrate their portfolios into a single mental transaction costs. 1985). 2008a.g. Cornell framing and loss aversion. Markets are not frictionless because of version.

which is part of the “black swan” phenomenon emphasized by Taleb (2006). Shefrin (2008a. the risky asset used to construct behavioral mean-variance portfolios features the use of derivatives.14 Behavioral mean-variance portfolios satisfy the two-fund separation theorem. during the first half of the 1990s. consider figure 2. However. The return to a neoclassical mean-variance portfolio is essentially linear. Barrone-Adesi et al. For example. Tomorrow’s managers should understand why people. FamaFrench multifactor model. To see how different behavioral and neoclassical mean-variance portfolios can be. 1985). Although an extensive discussion is beyond the scope of the present study. will be associated with betas and factor pricing models. a point worth addressing is whether behavioral assumptions alter the basic neoclassical relationship between the SDF and mean-variance frontier. In neoclassical theory. and how as managers they should deal with market inefficiencies. Notably. 2005). 12 If investors underestimate the probability of extreme negative events. where investors collectively commit errors in their judgments of probabilities. 14 .” 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. psychological obstacles may prevent organizations from putting them into practice (Shefrin. Behavioral risk premiums. using survey expectations data. make mistakes.12 The new approach should be specific.DE BONDT. mergers and acquisitions. This figure displays a mean-variance return pattern whose shape is that of an inverse U.13 Mean-variance analysis is very useful for bringing out the implications of behavioral phenomena for the pricing of all assets.b) introduces the concept of “behavioral adjusted present value. Shefrin (2005. capital structure. To illustrate this point. SHEFRIN. In this regard. This figure contrasts the equilibrium returns to two mean-variance portfolios. the theorem still holds but a negative position with respect to the risk-free asset is held. 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. the return to a behavioral mean-variance portfolio oscillates with economic growth. analysts attached too low a probability of a major decline in real estate prices. Moreover. In contrast. Notably. like their neoclassical counterparts. and models for the yield curve and option prices all appear as special cases of a general SDF framework. they need to be augmented so that there is a narrowing in the gap between what academics preach and what managers do. & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS? 13 article. Yet. MURADOGLU. and corresponds to the return from combining the risk-free security and the market portfolios. 13 In the case of leveraged portfolios. reflecting the impact of investor sentiment. thereby leading some derivatives and their underlying assets to be mispriced. behavioral risk premiums serve as compensation for bearing both sentiment and fundamental risks. His approach offers a unified treatment. Lohr (2008) discussed the failure of financial engineering to incorporate the human element. The behavioral approach adds sentiment risk. Many normative aspects of traditional corporate finance remain intact. including themselves. One of the most important points made in behavioral corporate finance is that although the principles taught in traditional corporate finance have great value. one neoclassical and the other behavioral. However. 2008) predicted that the shape of the SDF would change during 2001-2004. However. Risk can take any form in financial markets but. Shefrin (2008a. in broad terms. the CAPM corresponds to the special case when the SDF is a linear function of the growth rate of aggregate consumption in the economy. b) suggests that an appropriate starting point for discussing the asset pricing paradigm transition is the book written by John Cochrane (2005). It is a well-established fact that investors require compensation to assume risk. payout policy. with a decline in the left portion displayed in Figure 1. Aït-Sahalia and Lo (2000) and Rosenberg and Engle (2002) find that. the neoclassical framework focuses on fundamental risk. and focus on how to make decisions about capital budgeting. The weakness of the neoclassical SDF approach is that its underlying assumptions are behaviorally unrealistic. the capital asset pricing model (CAPM). the SDF features an oscillating shape that supports the predictions based on behavioral assumptions. Therefore. This type of situation is typical of events that take place in a behavioral SDF model. such a shape is implied Behavioral corporate finance emphasizes organizational heuristics and biases. Cochrane’s excellent work is built around the concept of a stochastic discount factor. the SDF is monotone declining. (2008) report that during 2002-2004 the left portion of the SDF does indeed feature a flat shape. as functions of aggregate consumption growth in the economy. What they do is alter the shape of the SDF and the ingredients of mean-variance portfolios. In particular. and corporate governance. The theoretical outcome of such combination is known as the two-fund separation theorem (Tobin. the behavioral stochastic discount factor (SDF) approach developed by Shefrin incorporates the human factor into financial engineering. then the SDF will typically be upward sloping in its left tail. Notably. Such heuristics and biases were endemic to financial firms involved in the global financial crisis that began in 2008. They do not. not general. consider figure 1. The construction of efficient portfolios under the neoclassical paradigm is done by combining an investment in the risk-free asset and the market portfolio.

28% Consumption Growth Rate g (Gross) 106.97 Return to a Combination of the Market Portfolio and Risk-free Security 0.87% 101.95 95.03 1.74% 102.96 0. the mean-variance function is exactly linear.98 0.49% 104. Gross Return to Mean-variance Portfolio: Behavioral Mean-Variance Return vs Efficient Market Mean-Variance Return 1.16% 100.19% 106% .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.01% 100.48% 98.82% 96.02 Neoclassical MV Portfolio Return 1.64% 97.31% 99.38% 105.01 Mean-variance Return 1 0. 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 1.61% 103. This figure also shows that the neoclassical mean-variance return is approximately linear. as functions of aggregate consumption growth in the economy. In the CAPM.99 Behavioral MV Portfolio Return 0. when behavioral mean-variance return function has the shape of an inverse-U.

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

Yet. entrepreneurs and investors tend towards unrealistic optimism – an error with perilous consequences. Still. 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. one may ask. For example. and this can be very costly. Recall the Y2K problem. spontaneous or deliberate. The questions that we would ask in relation to these evil acts are as follows: How did the perpetrators comprehend the world. the concept of error is elastic. 1990). 21 . 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. 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. Economists’ chief concern is efficiency. People enact roles. US data suggest that CEOs. cultural and historical factors. It is enough that we are able to read the time. These economists espouse the private enterprise system but call attention to the fact that its assumed virtues (innovativeness. experts make decisions (e. Yes. 1993).” Consider. aside from efficiency. Error. The truth may be nearly the opposite. there are other criteria of economic and financial organization such as sustainable development or equity and fairness. has to do with the September 2001 attacks in New York. Similarly. persuade them and motivate their actions? Investment bankers. technology is coupled with labor specialization. Smart technology and organization are human-centered (Reason. research in behavioral finance should examine the tangible content of people’s thought processes.” that personal freedom is lost when either law regulates what we do or large corporations – e. technology greatly extends man’s cognitive capabilities.g. behavioral finance must move beyond the narrow micro-level study of typical “mistakes. The secret is encapsulated by the motto of the 1933 Chicago World’s Fair: “Science finds. and conception of purpose. and economic organization are central. collective beliefs and norms often make all the difference. We lament that man must “conform. would be interested in answers to these questions. An extravagant illustration. We sell 99% of what we produce. First. James March and Chip Heath (1994) draw a useful distinction between the economic “logic of consequences” and the more broadly applicable “logic of appropriateness. Because we forget. 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. client relationship and financial marketing managers. how do Swiss citizens (who in majority rent) think about home ownership. so far.. structure and style of intuitive economic stories. for instance. These may be “protected values. Over longer periods. Third. administrative parsimony) have no solid basis in microeconomic theory. people reject all trade-offs for money. we buy 99% of what we consume. That in modern society the balance between individual and institutional forces has shifted often gets on our nerves. responsiveness. and we lead better lives for it.e. what sorts of economic arguments (true or false) sound plausible to investors. among others. they say. because of network externalities – control our choice options.” If not. Why are institutions so crucially important? The reason is that everyone in society depends on everyone else... of pragmatic behavioral research. In the short run. however. However.” i. this issue cannot be resolved without reference to social. habits. However. but may also make little practical difference.e.16 that of a stage actor. Hence. Organization produces It is difficult to interpret human action without knowing first how people think about a problem.21 Finally. Their motives. administrative organization can be effective or ineffective. we do not need to know. we use a notepad or we access the Internet. far removed from finance. people and machines have to work together. So. and likewise how do Americans? In general. Third. to what purpose and under which specifications. this is a matter of design. outlook and self-image are shaped by what is expected from them in society. His norm violations may be embarrassing.g. and how did they understand their self-interest so that they wanted to be suicide-pilots? 19 20 Austrian. Technology can be easy or difficult to use. For example. Finally. in relation to the nation’s supply of electricity) that tens of thousands are incapable of making for themselves. mainly in corporate bureaucracies and government. industry applies. society forgets why some systems or technologies were designed the way they were. Yet. good products or ideas spread quickly. someone who breaks the rules of etiquette. product and service standardization. man conforms.19 Evidently. Rationality and well-being derive from organization. technology embodies knowledge. man is limited by his brainpower. Occasionally. too much behavior remains unintelligible. 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. For example. technological artifacts often allow cheap replication. But. behavioral finance has little to say.” Technological artifacts make us smart for several reasons (Norman. Few of us know exactly how the watches on our wrists function. i. perhaps inexcusable. institutional and evolutionary economics do not.20 It blames social evils on dysfunctional incentives and disarray. Second. We need to look more into the content. Incessant technological progress. it is the outcome of an evolutionary process. Fortunately. Why are we collectively so strong.

22 Of course. and more broadly. “Testing Asset Pricing Models with Coskewness. and that they are intelligently reconstructed to meet social changes and to take advantage of technological progress. Urga. Retirement saving plans and asset allocation tools can be made more effective. history requires that economic and financial systems are continually updated. Behavioral finance holds the potential to create much value for society but it also has a great deal of work to do. that is to say their imperfections and bounded rationality. 307-344. 2). Handbook of the Economics of Finance Elsevier. want to make wise decisions. one can only hope. Rules and regulations coordinate society while reducing the individual’s need to think. G. What is required is “financial ergonomics. and T. Chances are that “the new paradigm” will combine neoclassical and behavioral elements. Stulz.. B. A unified SDF framework may also provide the basis for behavioral explanations of option pricing. and J. and R. our understanding of finance has increased a great deal. they must take into account the true nature of people. we hope. Take. A major paradigm shift is underway. Wurgler. Vishny.” Journal of Business and Economic Statistics 22.” Journal of Econometrics 94. In every instance. This is fundamental. “What Drives the Disposition Effect? An Analysis of a Long-Standing Preference-Based Explanation.M. the ATMmachine. Barone-Adesi. It will replace unrealistic. Shleifer. References Aït-Sahalia. 163-168. 1998. financial technology is often customerfriendly and performs brilliantly. N. Barone-Adesi.” Journal of Economic Perspectives 21 (No. Asset pricing theory. 9-51.” Journal of Financial Economics 49 (No. 2004. will combine a new realism in assumptions with methods and techniques first developed in neoclassical finance. “A Survey of Behavioral Finance. the term structure of interest rates. Xiong. “Investor Sentiment in the Stock Market. MURADOGLU. “A Model of Investor Sentiment. N.” Review of Financial Studies. heroic assumptions about the optimality of individual behavior with descriptive insights tested in laboratory experiments. 129-151. the industry will learn. for instance. if academicians are to succeed in understanding financial institutions and actors. N. and if the agents themselves. and P. Still. VII. Barber. G. forthcoming. Harris and R. 2007. markets and organizations remain a grey area waiting for behavioral researchers to shed light on it. Gagliardini and G. 2). Talwar. & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS? 17 predictability. 3). Baker.” a discipline that engineers financial products and services according to human needs and that optimizes well-being and overall system performance. Barberis. See Shefrin and Statman (1993b). 2009. The US mortgage debt crisis of 2007-2008 is a gigantic drama from which. 2003.A. Y. and W.DE BONDT. SHEFRIN. Thaler. and other asset prices. (Behavioral mean-variance portfolios may explain risk premiums. 2007. M. 1983. On the whole. it is easy to come up with counter-examples. Odean. Constandinides. P. Behavioral finance also offers a framework for understanding financial market regulation. 2). Conclusion Over the last few decades. “All that Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors. 22 . 474-485. “Nonparametric Risk Management and Implied Risk Aversion. yet there are countless questions begging for answers. Barberis.” Journal of Business and Economic Statistics 1 (No. as well as policy-makers. financial decision making processes in households. Barberis.” Journal of Finance 64 (No. M.” in Ed. the solution of these problems starts with the recognition that people are human. and A. “Market Models and Heteroscedasticity of Residual Security Returns. and R. by G. It is clear that. Lo. 2000.) Finally. 751-784. 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.

D.. W.H. 2008b. 2002. “The Parent Company Puzzle: When is the Whole Worth Less Than One of its Parts?” Journal of Corporate Finance 7. “Myopic Loss Aversion and the Equity Premium Puzzle. 2008. J. S. 3). Dittmar. 3)..F.M. 4). R. Teoh. 1). Thaler. 2005. De Bondt. “Investor Psychology and International Equity Markets.. Black. 2001. “Post-earnings Announcement Drift: Delayed Price Response or Risk Premium?” Journal of Accounting Research 27 (Supplement).” Journal of Finance 53 (No.. 1980. De Bondt. Bartov. 1977. Kurtosis Preference. 305–340..N. Thomas. Chen. and International Policies. Liu. “Investment Performance of Common Stocks in Relation to their Price-Earnings Ratios: A Test of the Efficient Market Hypothesis. V. 1).M. Asset Price Bubbles: Implications for Monetary. “Investors Psychology in Capital Markets: Evidence and Policy Implications”. “Bubble Psychology. 1985. D. W. 3). E.” Journal of Finance 32 (No. Basu. W.D. Subrahmanyam. 1986. Fama. “Do Security Analysts Overreact?” American Economic Review 80 (No.” Journal of Accounting and Economics 13.).. 557-581. Bernard. K. 1989. Hirshleifer and S. G. Cochrane. I..” In Advances in Behavioral Finance. AIMR. and J.D. 1990. Daniel.M. “Speculative Dynamics. Thaler. Amacom.” Journal of Finance 56 (No.” In Behavioral Finance and Decision Theory in Investment Management. MIT Press.F. and R. . Bernard. 6). 1991. Arbitrage. “Investor Sophistication and Patterns in Stock Returns after Earnings Announcements. Daniel. 1995. Bernard. 1998. R.. 1990. 139-209. D. Random House.” Journal of Finance 42 (No. Thaler. Mancini.). Benartzi. Regulatory. and Equilibrium Asset Pricing. Edited by R. B.-L.. E. Edited by Arnold S.K. Princeton University Press. “Stock Price Reactions to Earnings Announcements. Kaufman (eds. Hirshleifer and A. and Evidence from the Cross Section of Equity Returns. De Bondt. 1993. D. V. Thaler. “The Behavior of Stock Market Prices. K.M. and Q. “Style Momentum Within the S&P 500 Index.M.and Over-Reactions. D. 663–682.. W.18 Barone-Adesi. 1995. K. “Further Evidence on Investor Overreaction and Stock Market Seasonality. De Bondt.F.2). Dreman. Russell Sage Foundation.. 2002.” in W. 369-403.F. 2004.” Journal of Finance 41 (No.. Radhakrishnan. 1-48. 355–371. 3). Edward Elgar.” Journal of Finance 57 (No. Krinsky and S. S. 793808. W. 3). Poterba. Psychology and the Stock Market.M. D.M. De Bondt.F. Dreman. De Bondt. Journal of Monetary Economics 49 (No. F... “Does the Stock Market Overreact?” Journal of Finance 40 (No. “Betting on Trends: Intuitive Forecasts of Financial Risk and Return. 1). “Exploiting Behavioral Finance: Portfolio Strategy and Construction. Hunter and G. H. “Overconfidence. 43–63. Dreman. The New Contrarian Investment Strategy. 52– 57.” The Journal of Business 38 (No.F. “Noise. 529– 546. 1992.. and L.M. 2005. Summers. and W. 2001.” Review of Economic Studies 58 (No. “Investor Psychology and Security Market Under.” in Alan Lewis (ed. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Daniel.” Review of Financial Studies 21(No.N.. Wood... 921-965. 2008a. Engle and L. 107-121. “Stock Prices: Insights from Behavioral Finance. “Evidence that Stock Prices do not Fully Reflect the Implications of Current Earnings for Future Earnings. 341-366. 75-92. Cornell. 529-543.N. and J.” Accounting Review 75. and R. 18391886.” Quarterly Journal of Economics 110 (No. De Bondt. 4). 1987. De Bondt. 3). Cambridge University Press.K. 1965. Asset Pricing.. and R. D. “A GARCH Option Pricing Model with Filtered Historical Simulation.. W. “Nonlinear Pricing Kernels.F. 1977. 3). De Bondt. W. 34–106.. Thaler.” Journal of Empirical Finance 11(No. Cutler. V. 2002. W. The Cambridge Handbook of Psychology and Economic Behavior. 3). Thomas. J. Subrahmanyam. 12231258. Hirshleifer and A. 1).F.” Revue Economique et Sociale 66 (No. The Psychology of World Equity Markets. 2000.H.F. 483-507. and R. D.” International Journal of Forecasting 9 (No.

2). Grinblatt. Fellner.. Wermers. and E. “The Case for Flexible Exchange Rates. 1983. J. Ikenberry. 5). MURADOGLU. Schlarbaum. S. 2003. and K.. 2017-2046 Friedman.C.” Journal of Finance 55 (No. R. “Price Momentum and Trading Volume. Grinblatt.. 1998.. B.” Journal of Financial Economics 49 (No. 699–720. 92-117. 13-32. The Intelligent Investor: A Book of Practical Counsel.DE BONDT. 311-333. 427–465. 2000. 6). Han. C. and R. 263-291. J.. 1).A.” Financial Analysts Journal 32.” New York Times. 53-60.J. and A. B. 2017–2069.” The American Economic Review 85.” Journal of Finance 55 (No. Porter. “Market Underreaction to Open Market Share Repurchases.S.F.” Trust and Investments 69. 97-113. M. 17-23. and B. D. 2008. 3). 1). E. University of Chicago Press. and R. and T. 1976. List. 1949.. Keim. 12631295.” Quarterly Journal of Economics 75. Titman and R. “Do Professional Traders Exhibit Myopic Loss Aversion? An Experimental Analysis. 54-60. 1953. Tversky. 670-690. 181–208. 2). S.G. “Can the Market Add and Subtract? Mispricing in Tech Stock Carve-outs.” Journal of Finance 56 (No. Jegadeesh. and A. . “The Cross-Section of Expected Stock Returns. and G. Harvey. and T. and P. Working Paper No. 1999. “Momentum Investment Strategies.” Journal of Financial Economics 12.. “In Modeling Risk. Kahneman. Keim. Thaler... N.. 1999.M. 1995. Portfolio Performance and Herding: A Study of Mutual Funds Behavior. Titman. & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS? 19 Fama. Lease. 2).. W. E.” UCLA Anderson School. Peterson. “Size-Related Anomalies and Stock Return Seasonality: Further Empirical Evidence. O. “Does Prospect Theory Explain the Disposition Effect?” University of Zurich working paper. and J. Vlcek. Johnson. Jegadeesh. November 5. French. “The General Theory of Unemployment. 2001. Lee. “Market Efficiency. “Do Defaults Save Lives?” Science 302 (No. Goldstein. Dabora. “Profitability of Momentum Strategies: An Evaluation of Alternative Explanations. Moskowitz. “How are Stock Prices Affected by the Location of Trade?” Journal of Financial Economics 53 (No. 2005. N. “Distortion of Subjective Probabilities as a Reaction to Uncertainty.. 2003. 65-91. 2006. and S. A.G. and M.” Journal of Political Economy 111 (No.” Journal of Portfolio Management 12 (Winter).” Organizational Behavior and Human Decision Processes 51 (No. Vermaelen. 1).” Econometrica 47. Long-Term Returns. “Prospect Theory. Interest and Money.H.” Journal of Financial Economics 39 (No. 1088-1105. 1981. T. and Behavioral Finance. 227-268. Graham.F. 523-534. Fama. 24-99. 1993. Lohr. “The Disposition Effect and Underreaction to News. 1995. “Mental Accounting and Categorization. reprint of the 1936 edition. 1964. Hens. P. Froot. 1992. SHEFRIN. E.A. “Market Segmentation.. Grinblatt. 1999. 2). M. “Conditional Skewness in Asset Pricing Tests. Lakonishok. Frazzini. Swaminathan. S. W. “Dividend Yields and the January Effect. Lamont. and S. 1986.” Econometrica 49. “Behavioral Finance: Examining Thought Processes for Better Investing. 2-3).. 1961. 2005. Lewellen. K. Titman.” Harcourt Brace Jovanovich. “Prospect Theory: An Analysis of Decision Making under Risk. D. M. LeRoy. 1979. Keynes. Haigh. D. 3). Henderson.. 283–306. and M. 189-216.” Journal of Finance 60 (No.” Journal of Finance 41 (No. 2). 1992. “Stock Price Volatility: A Test Based on Implied Variance Bounds. and D. “Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. Siddique. the Human Factor Was Left Out. “The Cross-Section of Expected Returns and its Relation to Past Returns: New Evidence.D. Mental Accounting and Momentum.” Essays in Positive Economics. 5649). Harper. Koening. 2000. D. M. M.” Journal of Finance 47 (No.” Journal of Financial Economics 78 (No. 2005. 1338-1339.” Journal of Finance 48 (No. C.

541– 566. 1988.” Journal.. “Noise Trader Risk: Evidence from the Siamese twins. H. Shefrin. 1775-1798. M. 2008a. Roll./PLC. Behavioral Finance. K. McGrawHill.. G. Shefrin. Elsevier Academic Press (2nd Ed. 1). “An Exploratory Analysis of Portfolio Managers’ Probabilistic Forecasts of Stock Prices. 2005. 1987. 2002. H.” Journal of Finance 54 (No. Shefrin. “Factors Influencing Stock Demand in Turkey. 105-109. 2002. 2001a. G. 1990. Rouwenhorst. J. Roll. “Risk and Return in Behavioral SDFBased Asset Pricing Models.).G... “Portfolio Managers and Novices’ Forecasts of Risk and Return: Are There Predictable Forecast Errors?” Journal of Forecasting 21 (No. Rouwenhorst. and C. Shefrin. 573–608. and E. . 1983. Bogazici University. R. 89-104. Perseus Books.Mercan.. 1990.. 3). 551-584. J. 18-28. “The Anomalous Stock Market Behavior of Small Firms in January: Empirical Tests for Tax-Loss Selling Effects. March. 395416.V. F.” Journal of Forecasting 13 (No. L. 1989.. of Finance 57 (No. Edward Elgar. 1439–1464.” Journal of Financial Markets 10 (No. 5). 2000. 1994.. 2001b. 565–578. Onkal. 341372. “Between Hope and Fear: The Psychology of Risk. R. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Reason.20 Lopes. 6). “A Behavioral Approach To Efficient Portfolio Formation” The Journal of Behavioural Finance 6 (No. R. Salih. A Primer on Decision Making: How Decisions Happen.. 27–59. Free Press. “Are Investors Reluctant to Realize Their Losses?”Journal of Finance 53 (No. 113-124. 1-19. H..T. 3). Muradoglu.... Thaler.M. T..” Journal of Applied Corporate Finance 14 (No.. Muradoglu. 1984. Nicholson. Rosenberg. Heath. 1279-1298. Young.. “Limited Arbitrage in Equity Markets.” Journal of Finance 53 (No. G.” Journal of Financial Economics 12.. Keim. 2002. Muradoglu. “Local Return Factors and Turnover in Emerging Stock Markets. “The Turn-of-the-Year Effect and the Return Premia of Small Firms. Cambridge University Press.” Journal of Financial Economics 26 No. and C. “Behavioral Corporate Finance. 4). Roll. 1988. “International Momentum Strategies. 5). 1999. Ziemba. Womack.. Muradoglu. 1994. 7). Norman. Pulvino. 1). L.” In Security Market Imperfections in Worldwide Equity Markets. M. R. Turkey. 861–880. 2). K. 2007. 2002. 2005.” Financial Analysts Journal 24 (No.G. J. 3). Edited by D. and K. Scruggs. H. Shefrin. “The Seemingly Anomalous Price Behavior of Royal Dutch/Shell and Unilever N. Oxford University Press. Michaely. Behavioral Corporate Finance. H. Cambridge. 1). and L. “Do Investors Trade Too Much?” American Economic Review 89 (No. Istanbul. D. 1968.. 123–142. 202-212. 1995.H. Engle. and R. “Orange Juice and Weather. 2).. G. T. Poterba. Odean. G.” Journal of Investment Management 6 (No. 3). Reinganum. A. 1998.H. 1999. and W. T. 1993. “R2.. 267–284.” Advances in Experimental Social Psychology 20. and D. “Price Ratios in Relation to Investment Results. 76-105. “Price Reactions to Dividend Initiations and Omissions: Overreaction or Drift?” Journal of Finance 50 (No... “Mean Reversion in Stock Returns: Evidence and Implications. H. Cambridge University Press. 2008b..” Journal of Financial Economics 64 (No. Beyond Greed and Fear. “Empirical Pricing Kernels.” Journal of Financial Economics 22.A.” Journal of Finance 43 (No. Muradoglu.” Journal of Portfolio Management 9 (Winter). A Behavioral Approach to Asset Pricing. 4). Summers. R. 1998. J. Human Error. “Turkish Stock Market: Anomalies and Profit Opportunities. Stafford. 1). 5).” Unpublished thesis. and M.” American Economic Review 74 (No. 255-295. Odean.. J. Rosenthal. Shefrin. Things That Make Us Smart. Mitchell.. 1983.

“Pricing an Emerging Industry: Evidence from Internet Subsidiary Carve-outs. 421–436. and M. Random House. and D..” Review of Economic Studies 25 (No. “The Behavioral LifeCycle Hypothesis. 1985.. Statman. 2008. P. 392-410.J. 253-282. 1993.. M. .” Journal of Finance 27 (No. (February). 609643. 1958. Statman. “Judgment under Uncertainty: Heuristics and Biases. “Explaining Investor Preference for Cash Dividends. “Behavioral Aspects of the Design and Marketing of Financial Products. Advances in Behavioral Finance. 2008. A. 21-29. 4). Shleifer. H.H. Statman. “An Economic Theory of Self Control. Slovic.J.” Journal of Financial Planning.. 777-790.” Financial Management 30 (No. Thaler.” Financial Analysts Journal 49 (No.. 1972. Inefficient Markets. 3). and M. Tversky. Thaler. New York. 2001.. The Black Swan: The Impact of the Highly Improbable. and H. R. 297-323. 2). 1124-1131. 67). SHEFRIN. Liberty Fund (published in 1982). “Behavioral Portfolio Theory.. 2007.J. 1981. Schill. Russell Sage Foundation. Shefrin. McGraw-Hill. Siegel. Shiller. Fairness and Efficiency in Financial Markets. MURADOGLU. 4157). Shiller.” Financial Management 22 (No. and Happiness. 1988. Kahneman. 2). Stocks for the Long Run. 1992. H. R. R. 5-34. Shefrin. Wealth. Thaler. Smith. 779-799. 4).” Journal of Finance 40 (No. R. 3). and M. 2000.J. Taleb. and C. and M. and D. 6). “Advances in Prospect Theory: Cumulative Representation of Uncertainty. 1998. J. Statman.” Science 185 (No. 1759. “Liquidity Preference as Behavior Towards Risk. Shefrin. “The Disposition to Sell Winners too Early and Ride Losers too Long: Theory and Evidence. and C. J. Nudge: Improving Decisions about Health. “The Expressive Nature of Socially Responsible Investors. Statman. Yale University Press. Market Volatility. & STAIKOURAS — BEHAVIORAL FINANCE: QUO VADIS? 21 Shefrin.R. and M. N. A. The Theory of Moral Sentiments. Oxford University Press. 1993a. 1981.” Journal of Financial Economics 13.” Journal of Financial and Quantitative Analysis 35 (No. A. M.. Kahneman. Tversky. “Do Stock Market Prices Move too Much to be Justified by Subsequent Changes in Dividents?” American Economic Review 71. and R.. 1974.” Journal of Political Economy 89 (No.H. 123-134. 127-151. 2). Statman. Sunstein. 1993. 1993b. Zhou. Shefrin.DE BONDT. “Ethics. H. 4). R. “Psychological Study of Human Judgment: Implications for Investment Decision Making. A. Shefrin. 1984. H.” Economic Inquiry 26 (No. Thaler. Shefrin. 65-86. Tobin.” Journal of Risk and Uncertainty 5 (No. H. MIT Press. H.. 2000.

who work on behalf of a National Securities Dealer. Much of the academic literature has been devoted to uncovering behavioral biases in various settings and among different types of market participants. and 2) to examine how employees respond to control mechanisms. Odean (1998). Indeed.  A considerable amount of research has uncovered behavioral biases among financial market participants. 1 better decisions. Yet. PA. and Coval and Shumway (2005). 2 22 . While our results are useful for firms implementing or planning to implement risk control mechanisms. Grinblatt and Keloharju (2001).2 In our paper. Consequently. When trade price is heavily controlled yet trade size isn’t. they often hold onto their losing trades until the very last moment. 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. they have a tendency to hold their losing trades too long because they want to recover from their losses. We are also grateful to executives at the US Securities Firm for providing proprietary data. It is well known that market professionals often have difficulty coming to terms with their losses. 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. 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. This desire to get even is quite persuasive in financial market settings. some of the greatest Statman and Caldwell (1987) discuss risk control and behavioral biases in the context of capital budgeting decisions. We analyze how proprietary stock traders. of course. New Zealand. a necessary first step. Our findings highlight the difficult balancing act firms face with getting market professionals to realize their losses without impeding their trading strategies. and it is inevitably ingrained in many of the everyday decisions that traders make. When traders are forced to liquidate their inventory at a pre-designated time. for example. 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. We would like to thank the Editor. Despite this. Uncovering these biases is.1 In order to circumvent these biases and help employees make Ryan Garvey is an Associate Professor of Finance at Duquesne University in Pittsburgh. and an anonymous referee for helpful comments and suggestions on a prior draft. Ramesh Rao. our results also provide a step forward for the academic literature.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. we examine such a setting. See. Fei Wu is a Senior lecturer in Finance at Massey University in Palmerston North. some trading biases are well known and firms have been grappling with ways to control them for decades. react to institutional control mechanisms that are primarily intended to get them to realize their trading losses.

reviewed trader manuals. trading performance traders. on average. Locke considerably varies and traders let their losses run longer on and Mann (2005) and Garvey and Murphy (2004)). periods of time and performance would be a lot worse. our sample firm implemented several control mechanisms. However. And they hold losing trades longer than winning trades. This 4 We sat in on the firms training sessions for traders. Thus. trade size usually that. the larger size trades. 5 Our results highlight the complexities involved with For example. Garvey and Murphy (2004) examine proprietary stock traders who mainly offset their positions intraday. who work on behalf of a firm implemented other significantly declines as the National Securities Dealer. 1985). they are likely to which traders are required to liquidate their inventory. Locke and Mann (2005) assume commodity traders traders’ aversion to realizing losses. Consequently. In our setting. Institutional market participants trade in large trade longer than their winning trades. or are not aware of. and is why it is so important for financial firms to implement observed traders trade so that we could better prepare this paper.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. traders hold their losing trades significantly is not. near. Their tendency to hold losses for longer periods of time resulted in lower performance. the end of the simply unwilling to take a loss. 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. or largest losses resulting from this behavior. traders are forced to exit of holding losses too long. If firms implement end each day flat when determining trader gains and losses. realized. lower performance. trade size longer than their winning trades (see.e. activity throughout the day. Leeson incurred over $1. they are more unprofitable professional traders observed in prior studies were not when they trade in larger trade sizes. The firm even hired an on-site Inventory liquidation requirements ensure losses get 4 psychologist who was readily available to meet with traders. These longer holding times coincide with professional traders need flexibility with respect to trade size. so they gambled in an attempt trading day) and our main focus is on how traders holding to recover from their loss. required to close out of all of their positions by a These traders were constantly being drilled on the dangers 5 predetermined time. If the 3 firm did not require traders to close out of their positions by One of the more notable cases. occurred with Nicholas Leeson. We find controlled in institutional trading settings. they clearly employed a trading manager who closely monitored trading do have an influence on trader behavior. the firm last moment. 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. for example. The Moreover. While trade price is often heavily to have difficulty coming to terms with their losses. had several discussions with management and traders. . Statman.3 times and performance vary across the day leading up to the In order to help traders come to terms with their losses. but the traders can and do hold implementing optimal risk control mechanisms to circumvent positions overnight. presumably losses would be held for even longer old Barings PLC. For example. 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. traders still appear mechanisms to do the job. which led to the demise of his employer. control mechanisms. ensure loss realization. any mandatory time period in control mechanisms that are too stringent.4 billion in the end of the day. react to control mechanisms including liquidation deadline draws institutional control mechanisms an emphasis on price control. 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. dangers of holding losses too to hold onto them until the very long. but they tended to hold onto their losses despite the warning. inventory liquidation deadline. 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. 232 yearmeasures. the authors do not report. Mr. or if they did not implement any control trading losses in 1995. but firms (traders) also rely on price control Despite all of these control measures.

a behavior that is commonly known as the “disposition effect”.925 shares. the execution time. Each trader’s sole objective was to generate intraday trading profits utilizing firm capital. the market where the order was sent. Related Research Kahneman and Tversky’s (1979) prospect theory provides a descriptive framework for decision-making under risk. While most researchers examine traders’ unwillingness to take a loss through holding times and focus on decisionmaking in a single period setting. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 I.. regret aversion. the type of trade (marketable versus limit order).g. as our research shows. The disposition effect has proven to be quite pervasive in US markets. Dhar and Zhu (2007) find that individuals’ income level. Grinblatt and Keloharju. and we examine trader holding-time decisions in a singleperiod setting. Much of the behavioral finance literature focuses around how market participants make decisions when they are confronted with the prospect of a loss. Like much of the prior research. the price. and self-control. are important factors in indicating who is more susceptible to the disposition effect.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). Garvey and Murphy (2004) and Heisler (1994)) all exhibit signs of this behavior. Shapira and Venezia.g. Feng and Seasholes. 2002 through May 30. China (e. but they need to be very careful with how they enforce this because traders align their strategies with control mechanisms. 7 Researchers have also found strong evidence of the disposition effect in Finland (e. larger size trading. Because psychological biases are so intertwined in many people’s decision-making processes. occupational status. the location of the trader (the traders were located in five branch offices). firms are unlikely to eliminate these biases. the volume. (2007) find traders who have experienced prior morning losses engage in subsequent afternoon risk-taking as measured through increased trading activity. Our motivation is not to examine if traders suffer from the get even behavior that underlies the disposition effect. research shows that individual investors (e. For example. our findings imply that firms do need to implement control mechanisms. Barber and Odean (1999) and Odean (1998)). Israel (e. Traders. short. think about stock purchases within separate mental accounts (see Thaler.g. if firms do nothing they open themselves up for considerable risks. Data Our data originates from a National Securities Dealer. A central theme in their research is the role of loss.g. and various other trade execution information.g.24 conflict with traders’ overall strategies and objectives because. Taiwan (e.. For every transaction. 1985) then apply prospect theory decision rules to each mental account. the 150 traders combined to execute 2. Coval and Shumway (2005) and Garvey et al. During this oneyear period. Frazzini (2006) and Scherbina and Jin (2005)). 2001). II. 6 . we examine trader resistance to loss realization through holding-time decisions. mutual fund managers (e. who exhibit behavior that is consistent with the disposition effect. some other studies have looked elsewhere. These factors together help explain theoretically why traders have a tendency to hold their losing trades much longer than their winning trades. Shefrin and Statman (1985) were the first to apply prospect theory to a financial market setting.g.5 billion shares through 1. the US stock markets were open for 251 days. etc. Thus. the contra party on the trade (if given)..3 million transactions on 693 securities. The data covers June 3. sell. They also placed prospect theory in a wider theoretical framework that includes mental accounting. and cover). These findings are consistent with the same behavioral tendency that leads traders to hold their losing trades too long. For example. 2007) and other countries. They examine a trader’s reluctance to realize losses using other risk measures and focus on decisionmaking in a multi-period setting. The average trader executes 75 per day and the average executed trade size is for 1. Some of the more recent studies identify individual trader characteristics that are correlated with the disposition effect.6 For example. the data reveals the identity of the trader. 2003. The firm had several trading operations and our focus is on the firm’s proprietary trading operation for US equities. the traders trade often and they also trade in large trade sizes. In total. but rather our motivation is to examine how (if) traders respond to institutional control mechanisms to prevent this behavior. and professional traders (e. On the other hand.g. 2005). Consequently. This average trade size is more than three times the average trade size in US equity markets. etc. trader behavior is heavily influenced by control mechanisms. but they can lessen the damage caused by them.. Barber et al. 2001). the action taken (buy.

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. mandatory close-out period. A trader could combine a closing do examine institutional trading (see.m.334 stock trades over just a one-year period. the traders had a clear incentive to the corresponding holding time for each round-trip. suppose a trader opens up a 2.8 While the efforts to get traders to realize their trading losses. is not alone with respect to their efforts in this regard. SUNW and JDSU are two of the more actively traded US stocks. 1996. prices in the various trading venues which US equities trade in.417 roundcompensation of the traders is solely tied to their trading trip trades from the 1.465 households execute 1. 10 11 The data has been used in several studies in finance literature. In order maximize their trading performance. and prior studies do not examine how opened. transaction with an opening transaction. Van Ness. The traders set market firms and their risk managers constantly grapple with.000 share position of Yahoo at 10:30:00 a. trading intensity difference A. we calculate the holding time between the intraday opening and closing transaction(s). analyze the trading Specifically. which reported its trades through the Cincinnati Stock Exchange. They trips systematically rises throughout them to accept their losses. and Wahal. Our sample traders received continual training on various position was closed out. We then calculate out performance.3% of the positions. The traders did not always open and close is focused on retail market participants. If a trader accumulates inventory before they eventually close out of their position. For our purposes. this literature largely focuses on part of an open position.33 seconds (1/3 * 20 seconds + 2/3 * 10 seconds). 8 we use a weighted average between the various opening The only stocks traded every day were Sun Microsystems (SUNW) and JDS Uniphase (JDSU). If the next trade were a sell of 6. Empirical Results To get an idea of the Traders hold losses longer than gains. 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. yet they had accumulated inventory.GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR 25 Nasdaq) on certain days.969. and Van Ness (2004) for a discussion on the reporting of Island trades on the Cincinnati Stock Exchange. who trade holding-time decisions. 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). holding-time measures are not included in the raw transaction 66. Regardless of whether trades transaction costs.000 shares of Yahoo at 10:30:20 a. Conrad. trading strategies from the firms’ management. they also set market prices often. and corresponding prices paid or considerable freedom with selecting stocks to trade. or they could lay off Johnson. While some studies positions with two trades.5% and 3. Our 150 traders data. while traders to take losses is a common problem that securities 35% of their trades take liquidity.3 million trades. we searched forward in time each day until the opening institutional traders respond to risk control mechanisms.11 annual share volume of SUNW and JDSU respectively. Methodology but these holding time patterns do not between institutional and The objective of our study retail market participants. . The firm traders trade often. often accounting for a sizeable most intriguing aspect of this particular setting is the firm’s portion of a particular stock’s trading volume. In order to determine the gains and losses for each execute 1.m. In total. Thus. The traders accounted for 1. more recently. we use an intraday round-trip matching procedure Institutional market participants dominate the trading similar to the one used in Garvey and Murphy (2005). round-trip. See Nguyen. Getting Approximately 65% of their trades provide liquidity.000 shares of Yahoo at 10:30:10 a. 9 The traders often traded on The Island ECN. or open and closed a position simultaneously. The received. through a US discount Round-trip performance and brokerage firm. We landscape in US equity markets. For example. and we kept track of execution times. The matching procedure creates 730. closed.9 III. the holding time on the round trip trade is 13.701 stock trades over a sixyear period ending in December.316. remain constant throughout the day. the to do this. 2002). and then purchases another 4.m..

which highlights how deadline effects coincide with a trader’s resistance to realize losses.898 break-even round-trips (gross roundtrip trading profits equal to zero). 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.176 seconds. so when traders’ open positions are held for extended periods of time. The individual trader results coincide with the aggregate trader results. The objective of the traders is to rapidly enter and exit positions in order to profit from small price changes. The sample period we observe was a difficult time to trade US equities. 12 overall result. They also show how important holding times are for a trader’s success. 146 out of 150 traders hold their losing round-trip trades longer than their winning round-trip trades (139 differences are statistically significant). traders would hold their losses for even longer periods of time. Traders hold losses longer than gains. 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.12 Our firm was not immune to these difficult trading conditions. In order to check the robustness of our initial result. Trader profits are highest when they hold their trades for under one minute.248 winning round-trips (gross round-trip trading profits above $0). And when they enter into a position on one side of the market. The frequency of breakeven round-trips highlights how focused these traders are on their trade purchase price (the reference point). On a mean holding time basis. Round-trip trading profits systematically decline with longer holding times.26 From the matching procedure. The absolute trading profit difference is statistically significant from zero at the 1% level. 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). This allows us to see if certain traders are driving our For example. Table II and Figure 2 provide information on round-trip holding times across the day. How Does Inventory Control Influence Trader Behavior? While the firm’s efforts do not appear to eliminate biases from traders’ decisions. holding positions for longer periods of time is generally undesirable.23. The trader’s information is short-lived. 14 . Many securities firms were reporting steep losses on their equity trading desks. 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. In Figure 1. 13 The average round-trip gain is $13.271 losing round-trips (gross round-trip trading profits below $0). On average. The traders do not hold their open positions for long. 2005). 145 out of 150 traders hold their losing round-trip trades longer than their winning round-trip trades (135 differences are statistically significant).274 seconds and their winning roundtrip trades for 568 seconds. When traders hold their trades under (over) five minutes they are profitable (unprofitable).13 In our setting. 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. For example. their risk control measures do have an influence on trader behavior. they generally seek to quickly offset their position by trading on the opposite side of the market. but these holding time patterns do not remain constant throughout the day. 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.14 These results highlight the trader’s short-term strategies. traders hold their losing round-trip trades for 1. the revenues of broker dealers with a Nasdaq market making operation fell over 70% during 2000-2004 (GAO. This influence is quite strong in the moments just prior to the inventory liquidation deadline. Our firm had a Nasdaq market making operation. the mean holding time per round-trip is 780 seconds and the median holding time per round-trip is 205 seconds. 209. we examine the holding time differences for each individual trader. and 230. there are 290. C. Presumably. The difference in holding times between losing and winning round- The average round-trip holding time among traders ranges from 122 to 3. and trader profits are lowest when they hold their trades for more than 15 minutes. we segregate the roundtrip profits into five holding time categories. 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. Despite the firm’s efforts to get traders to realize their losses in a timely fashion.70 and the average round-trip loss is $19. On a median holding time basis.

417 round-trip trades over the 251 day sample period. Average Round-trip Profit and Holding Times $6. The traders only traded US equities.417 round-trip trades over the 251 day sample period. 145 135* Median 146 139* Figure 1. who traded the capital of a National Securities Dealer from May 2002 through June 2003.00) ($4. The results are based on the trading records of 150 proprietary stock traders. The mean (median) holding time is 780 (205) seconds. Winning (losing) round-trips have a gross trading profit above (below) zero and break-even round-trips are omitted.00 $0.00) ($6. 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.10 level. The traders executed 1. Performance and Holding Times This figure displays the average round-trip trading profit for five holding time categories. Panel A.00 $4.274 568 706* Median Holding Time 377 166 141* Panel B.00 ($2.00) Less than 1-2 1 minute minutes 2-5 minutes 5-15 More than minutes 15 minutes . The traders executed 1.00 $2. Holding times are calculated in seconds.GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR 27 Table I.3 million trades which resulted in 730.3 million trades which resulted in 730. Aggregate Results Mean Holding Time Losing round-trips Winning round-trips Difference 1. 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. 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.

9% 6.m.207 49.5% 10.417 round-trip trades over the 251 day sample period.4% 12.m. 1:00-1:30 p. a.604 13.777 12.5% 7. 2:30-3:00 p. .3 million trades which resulted in 730.m 0 a.m 0 p.:00.390 12.419 53.3 million trades which resulted in 730.m 0 a.921 66. 11:30-12:00 a.816 61. .1% 25. 10:00-10:30 a.107 53.m.28 Table II. . Holding times are calculated in seconds.428 round-trip trades executed before the open (9:30 a. 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 2003.625 Percentage of Round-Trips above Mean Holding Time 3.m.:30. The 1. .) or after the close (4:00 p. 11:00-11:30 a.m.128 13.3% 24. The traders executed 1. .m.5% 6.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 .m.:00. Losing Holding Times: Time of Day Seconds 2000 1500 1000 500 0 .m. 1:30-2:00 p. .124 13.m 0 p.8% 6.m.1% 8.051 16.4% 6.m 0 p.7% 23.848 47.4% 24.554 8. Intraday Time Number of Round-Trips Percentage of Round-Trips Number of RoundTrips above Mean Holding Time 2.m.4% 9.597 53.267 75.8% 28. 12:30-1:00 p. 62.3% 7.787 50. .0% 27. 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.m.m. Winning vs.1 0 -1 0 -1 0 -1 0 -1 0 -1 :30 .:00.0% 9:30-10:00 a. . .m 0 p. 2:00-2:30 p. .093 9.3% Figure 2.:30. 3:30-4:00 p.2% 21. 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. .415 13.m. 3:00-3:30 p.094 12. 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.: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 .m.6% 24.055 51. 10:30-11:00 a. .m 0 p.3% 7.) are omitted.7% 7.537 46.m 0 a.680 12.m 0 p. 3500 3000 2500 Average Winning vs.m. 12:00-12:30 p.m 0 p.6% 31.027 13. The traders executed 1.1% 7.485 13.7% 25.8% 18.874 56. Winning round-trips have a gross trading profit above zero while losing round-trips have a gross trading profit below zero.m 0 a.417 round-trip trades over the 251 day sample period.m 0 p.

In Table III. there are competing factors There are many self-employed (retail) traders who adhere to this rule on a self-imposed basis. we compute a hypothetical performance measure assuming some trades were held overnight.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. it is interesting to theorize about what would occur if the firm did not have this control mechanism in place. In the US. 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. The sharp decline in traders’ profits in the moments just before the close. The trader’s behavior is not a desirable reaction to what the firm is trying to accomplish. or the period which is furthest away from the liquidation deadline. The Admati and Pfleiderer (1988) theory has served as a prominent explanation for these intraday volume patterns. 15 .17 (note that a few anomalous observations are dropped from the analysis). 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. Without this control mechanism in place. and other types of traders to end the day flat.332 round-trip observations. account for a very large percentage of overall daily trading volume in US equity markets. they usually close out of their positions.56. If the firm allowed traders to hold their positions overnight. 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). arbitrage. the average round-trip trading profit declines from -$3. Because trading in the very last moments of the day appears highly driven by the firm’s control mechanism. Trading profits steadily decline until noon. The magnitude of the losses that occurred in each five minute interval is highlighted in Figure 3. would this give traders greater flexibility with implementing their trading strategies and subsequently improve performance? Or.15 If traders at other firms exhibit behavioral tendencies similar to these traders. In order to do this. 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. indicates that trading is significantly driven by the firm’s control mechanism. to 4:00 p. While the end of day closeout period induces trade and forces traders to realize their losses. the lunch period serves as another period for realizing losses. Contrary to using the round-trip trade price at the end of the day to determine profits. Our firm’s inventory liquidation requirement is not unique. When traders leave their trading terminals. 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. along with the holding time patterns. we report the gross round-trip trading profits for each half hour of the trading day. but the liquidation requirement appears necessary. Under the adjusted closing price. proprietary.. these traders’ reluctance to take losses would have most likely resulted in larger trading losses. these retail day traders typically trade through direct access brokers. and statistically different from zero at the 1% level.m. Trading profits are positive after 12:00 p. and losses are generally held for shorter periods of time leading up to the midday period. 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. While our traders’ intraday trading activity closely resembles a U-shape pattern. Thus. For many traders. 2004).95 to -$4. Trading profits are highest in the initial opening period. but they steadily decline again until the close of trading. than our results provide insight into some factors that drive intraday order flow patterns.). or they significantly restrict traders ability to accumulate inventory from day to day. While the firm’s liquidation requirement imposes a constraint on traders. 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 traders lost $136. The results indicate that traders often hold losses up until the very last moment before they are forced to realize them.674 shares. who often end the trading day flat.m. we recalculate trading profits for positions closed out in the final 15 minutes of the trading day. There are 24. Many securities firms require their market making. However. the midday close-out period is not binding like the end-of-the-day period is. The average round-trip profit in the last 30 minutes is -$2. we assume traders closed their positions at the opening price on the following (trading) day.522% less than the average round-trip trading profit at other intraday times. and the average closing position is for 1. whereas the average round-trip profit at other intraday times is $0.934 in the final 30-minute period. The average round-trip trading profit in the closing 30 minutes is 1.18.m. Trading profits most likely continue to decline until noon because traders often break for lunch. The end-ofday trading losses can be broken down further. retail and institutional traders.

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

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. Fig. Winning round-trips have a gross trading profit above zero while losing round-trips have a gross trading profit below zero. Stock turnover ratios are calculated using CRSP by averaging daily stock turnover (volume/shares outstanding) over the sample period.3 million trades on 693 securities which resulted in 730.GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR 33 Figure 4. In Fig. The traders executed 1. Liquid (illiquid) stocks are stocks with a turnover ratio in the top (bottom) 50% percentile of our sample. 4B Fig. 4C . 4B and 4C. holding times results are reported separately for trades occurring on liquid stocks and illiquid stocks. 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). 4A Fig. Winning vs.417 roundtrip trades over the 251 day sample period.

05 $38.69 -$15. there are very few large trade size observations on illiquid stocks.69 -$11. groups (liquid vs.496 89. RoundTrip Gain Avg.24 $3.84 ***Significant at the 0. RoundTrip Gain -$2. Panel A. Liquid (illiquid) stocks are stocks with a turnover ratio in the top (bottom) 50% percentile of our sample.93 -$28.01 -$19.22 $9. Illiquid Stocks <250 250.65 -$189.88** -$3.40 -$44.05 Avg.<3000 3000 12. RoundTrip Loss Absolute RoundTrip Profit Difference Total Trading Profits <250 250. In Panel B & C.09 -$45. Liquid Stocks <250 250.33 $8.269. RoundTrip Loss $1.00*** $4. Although there is a sharp drop in the performance difference for illiquid stocks under the largest trade size category.78 Avg.51 Avg.172. All Stocks Trade Size (Shares) Number of RoundTrips Avg.46 -$28.803.034.07 $102.747.66 $16.<1000 1000.3 million trades on 693 securities which resulted in 730. As expected.271 163.417 round-trip trades over the 251 day sample period.22 $22.99*** $5. RoundTrip Loss $0.821.340. Our .771. performance results are reported separately for trades occurring on liquid stocks and illiquid stocks.982 154.53 -$29.684 12.05 level.65 $26.47*** $5. most trading activity occurs on liquid stocks and holding times are much lower on liquid stocks.732 79.54 $7. 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. We are able to use matching trade data from CRSP for more than 97% of the trading activity in our data.96*** $3.<3000 3000 Trade Size (Shares) 154.24 -$39.34 Table V.25*** $5.<2000 2000.46*** $7.35*** $5.803 130.93 $45.99*** Absolute RoundTrip Profit Difference -$14.12 -$24.104 $2. 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 2003.11 Total Trading Profits Panel B.00 $36.47 -$129.47 $23. **Significant at the 0.802 Number of RoundTrips $1. We compute holding time differences and performance differences for both liquid stocks and illiquid stocks according to our trade size classifications.00 $27.82*** $7.<2000 2000.71*** $5.25*** $2.39*** $4.52 Total Trading Profits Panel C.<1000 1000.808.426 Number of RoundTrips $1.69 $18.652. Stock turnover ratios are calculated using CRSP by averaging daily stock turnover (volume/shares outstanding) over the sample period.50 -$9. losing round-trips are held considerably longer than winning round-trips for each trade size category and the difference systematically widens with trade size.969 138.00 $14.46 -$33.03*** $3.97 $41.517. The traders executed 1.<1000 1000. In general.453. the absolute difference between the average trading gain and loss correspondingly widens with trade size too. RoundTrip Gain -$2.04 Avg.<2000 2000.<3000 3000 Trade Size (Shares) 139.26 -$3.057. 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.66 $7. illiquid stocks) based on their average daily turnover ratios (volume / shares outstanding) over our one year sample period.62 $14.221 14.13 $1.613 12. Volume and share outstanding data is obtained from the CRSP database.99 -$18.629. Performance and Trade Size JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 This table reports round-trip performance results segregated by five trade size categories.469 182.62 $2. For both liquid stocks and illiquid stocks.044 149.018 6.42 $92.01 level.967.85 $39.73*** Absolute RoundTrip Profit Difference -$18.

IV. However. a firm may analyze the trading decisions of its fund managers and find that overall. this behavior is undesirable and can be quite costly. 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. we find that professional traders still have difficulties accepting their losses. By all accounts. but their efforts to get employees to accept their losses has much broader implications. these are not desirable behavioral responses to the firms underlying objective. While our results highlight how difficult it is for institutions to rid psychological biases from the traders’ decisions. For example. they may exhibit a tendency to do so with their larger holdings. managerial oversight. losing trades will be held for longer periods of time and losses will surmount. traders hold losing trades more than twice as long as winning trades and these longer holding times coincide with lower trading profits. For example. Clearly. While the fund managers may not exhibit a tendency to avoid realizing their losses on an overall basis. if control mechanisms put in place are too lax. the additional trading constraints will likely begin to start conflicting with the traders’ overall strategies and trading practices. if financial institutions impose stricter control mechanisms to get their traders to realize their losses sooner. we examine whether such measures actually work and how they influence proprietary stock trader behavior. In our paper. trader access to a licensed psychologist. Traders have a tendency to hold their losing trades too long because they are predisposed to get even with their losses. Despite our sample firm’s efforts to get traders more comfortable with realizing their trading losses through training. would be insightful for both creating optimal risk control mechanisms and also for determining their overall effects in the marketplace. On the other hand. Our results show that institutional risk control can have a strong influence on trader behavior. which would pose a significant (preventable) risk that is not easily detectable though casual analyses of the overall trading data.GARVEY & WU — THE EFFECTS OF INSTITUTIONAL RISK CONTROL ON TRADER BEHAVIOR 35 results also highlight the need to devise control mechanisms. which provide institution detail on the design of control mechanisms being used at other financial institutions and how employees respond to them. Consequently. 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. and inventory liquidation. Future studies. professional traders respond by holding their losses longer on larger size trades. . our results also highlight the complexities involved with implementing efficient control mechanisms to get traders to realize their losses. professional traders respond by holding their losses up until the very last moment. their fund managers do not exhibit a tendency to avoid realizing their losses. When firms force traders to liquidate their inventory and realize their losses. discipline price control. Conclusion One of the more well known psychological tendencies that permeates Wall Street trading desks is the traders’ aversion to realizing losses. When firms heavily focus on disciplined trading and uniform price control to ensure loss realization. on a situational trade basis. among other things. Securities firms implement control mechanisms to improve performance and reduce risk.

Barber. Shumway. J. Garvey. Do Investor Sophistication and Trading Experience Eliminate Behavioral Biases in Financial Markets?” Review of Finance 9 (No. 793-822. Coval. May.” Management Science. Mann. 41-55.” Marketing Science 4 (No.” Quarterly Journal of Economics 110 (No.” Journal of Portfolio Management 33 (No. Murphy. 305-351. R. B. 1999. 589-616.” Financial Management 16 (No. Frazzini. Zhu. Pfleiderer. 93-100.M. Seasholes. D. 2005. and S.. 30-39. and I. “What Makes Investors Trade?” Journal of Finance 2 (No. “Patterns of Behavior of Professionally Managed and Independent Investors. and F. “Are Professional Traders Too Slow to Realize Their Losses?” Financial Analysts Journal 60 (No. 76 (No. 73-92. Johnson.” Journal of Applied Finance. 199-214. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Government Accountability Office (GAO). Conrad. Caldwell. 35-43. 2001. J.. “Up Close and Personal: An Individual Level Analysis of the Disposition Effect. but Trading Volumes Rise. 14 (No. and A. “Change is Good or The Disposition Effect among Mutual Fund Managers. Odean. Keloharju. H. “Professional Trader Discipline and Trade Disposition. and R. 2005. 1994. Lee. 4). B. 13 (No. “The Disposition Effect and Underreaction to News. and L. Wu. Shefrin.” Journal of Finance 61 (No. Statman. 2002. 8). 6).. Y. and M. Garvey. R. R. 2). “Applying Behavioral Finance to Capital Budgeting: Project Terminations. A. Wahal. “Do Behavioral Biases Affect Prices?” Journal of Finance 60 (No. and M. and A. Lupercio. 2). Venezia. “Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors..36 References Admati. 4).F.. Z.” Journal of Applied Finance 15 (No.. and T. A. J. “The Courage of Misguided Convictions: The Trading Behavior of Individual Investors. 7-15. 4). 777-790. Goldberg. 401-44. Full report available online at: http://www. A. M. 2001. M. 1775-1798. 3). Odean. “Mental Accounting and Consumer Choice.M. R. 4). Barber. T. Feng. 2005. 1). Scherbina. “Is the Aggregate Investor Reluctant to Realize Losses: Evidence from Taiwan.” Review of Futures Markets 13 (No. B. 2017-2046. Van Ness. 1). A. 2000. 2007. and M.M. and T. 2007. 3). 2004.” European Financial Management.A.. 7-14.” GAO Report. Odean. and A. August. Y. “Cruising at 30. 2007. 1573-1587. and N. Lu. Jin. 75-83. K. Van Ness. 3).items/d05535.” Working Paper. forthcoming.T.C. 1995.” Financial Analysts Journal 56 (No. 1985.R. Murphy. “The Disposition to Sell Winners Too Early and Ride Losers Too Long: Theory and Evidence. Shapira. Benartzi. Grinblatt. B.” Journal of Finance 55 (No.R. 1998.gov/new. “Loss Aversion in a Futures Market: An Empirical Test. 2005. “Do Losses Linger? Evidence from Proprietary Stock Traders.” Journal of Applied Finance 12 (No. “Are Investors Reluctant to Realize Their Losses?” Journal of Finance 53 (No. and T. 773-806. 1985. and D. 2).” Bear Stearns Company Report. Harvard Business School. L. Nguyen. 371-388. “Decimal Pricing has Contributed to Lower Trading Costs and a More Challenging Trading Environment. “A Theory of Intraday Patterns: Volume and Price Variability. 3). Heisler. Statman. and R. Garvey. “The Trading of Institutional Investors: Theory and Evidence. R. 2005. 2006. 3-40. “Myopic Loss Aversion and the Equity Premium Puzzle. 2004. and T.” Journal of Banking and Finance 25 (No. and S. 1987. “The Profitability of Active Stock Traders. 1). .. “The Reporting of Island Trades on the Cincinnati Stock Exchange. V. 2). 1988. 2004. 1-34. 2). 3).” Journal of Finance 40 (No. 2005. 5). Locke. S. Barber.” Review of Financial Studies 1 (No. P. Thaler. 1).” Journal of Financial Economics.gao.C.000: SemiPro Numbers Level Off. Murphy.pdf. Dhar. Odean. Thaler. and P.

Trades are necessary to get into and out of the market. Trading volume is concentrated among a small number of market participants. standard motives for trading such as saving and rebalancing explain little of the variation in trading activity across investors. The copyright on “Trading as Entertainment?” is held by INFORMS. stocks in the US change hands roughly once per year. This contribution is based on the article “Trading as Entertainment?” by Daniel Dorn and Paul Sengmueller which is forthcoming in Management Science. Paul Sengmueller is an Assistant Professor of Finance at CentER-Tilburg University in Tilburg. Daniel Dorn. and Paul Sengmueller Besides trading to save. we find.Why Do People Trade?1 Anne Dorn. 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. a substantial fraction of US individual investors with a brokerage account do not trade at all in a given year. 1 37 . Daniel Dorn is an Assistant Professor of Finance at Drexel University in Philadelphia. respondents who indicate that they “enjoy investing” and “enjoy risky propositions” trade twice as much as their peers. In a well-cited study of US discount brokerage clients. We argue that in exchange for this major source of revenue. Trading volume in financial markets is high. In a survey of 1. and to convert back into cash when the money is wanted. In contrast. people trade simply because they find it entertaining. PA. racking up trading costs like a casino patron sliding his chips across the table. and hand over their large trading fees as happily as an audiophile pays top dollar for the latest in speaker technology. 2. Barber and Odean (2000) report that the most active investors turn over their portfolio several times per year. are playing the market in a very literal sense. Some investors view trading as a hobby. Other investors. All this trading performs the Anne Dorn is currently unaffiliated. to put unneeded cash into the market. 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. Traders underperform buy-and-hold investors. to exchange one asset for another. manage risk. In this paper. and speculate. and to exploit information about future price movements. the Netherlands. we explore reasons investors trade that go beyond assembling a portfolio with the best return-risk profile. The rational investor assumed by standard theory is only interested in the return and risk attributes of his portfolio. It also (viewed from certain quarters more importantly) provides a major source of revenue for securities firms. 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.300 German discount brokerage clients. They are also needed to move money around within the market. 3. securities firms are also allowing traders to enjoy themselves in the act of trading. For example. according to recent waves of the US Survey of Consumer Finances. Trading in financial markets is an important economic activity. important social function of incorporating information into asset prices. Standard economic theory appears to be at odds with these facts. to manage risk. In contrast.

and is then tempted to go out and buy them). An undiversified portfolio of volatile stocks exposes its holder to intense stimuli in the form of extreme returns. and Biais. Grinblatt and Keloharju (2008) use self-confidence assessments from a psychological test administered by the Finnish military to infer overconfidence of male Finnish investors. executing. Lüders. camaraderie (among members of an investment club. the joy of imagining what a handsome payoff will buy. For all its intuitive appeal.” The leading answer of the behavioral camp to the question “Why do people trade?” is overconfidence. Milgrom and Stokey (1982) and Tirole (1982) formalize this intuition and show that indeed. coupled with an inherent impatience to reach their desired wealth level. Essentially. rational investors should refuse to speculatively trade with each other. may lead aspiration-driven investors to pick up and abandon trading ideas more quickly than their peers. According to Zuckerman (1994). and the willingness to take [. or experiencing the outcome of a trade. prominently advocated by Odean (1998). Deaves. return maximizing rational investors should be very reluctant to trade with one other. Perceiving investing as a diversion rather than a chore. and intense sensations and experiences. Other things equal. In general. (2005) report little or no relation between proxies for overconfidence and observed trading activity. and Pouget. they report that variation in the number of speeding tickets explains variation in trading activity in a large sample of Finnish investors. 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.” For people who trade because they like to do so. and Luo (2004). anticipated by Black (1986) who notes that “[w]e may need to introduce direct utility of trading to explain the existence of speculative markets. 2002). the proxy for overconfidence derived from the selfconfidence assessments is significantly positively related to the number of trades. De Bondt and Thaler (1995) call the observed trading volume in financial markets “perhaps the single most embarrassing fact to the standard finance paradigm. A trade can be seen as a bet that carries a “dream value.. The exposure to trading stimuli in the form of extreme returns. the overconfidence hypothesis has received only mixed empirical support. B should be suspicious that A knows something about the future price that B does not. Trading is a zero-sum game. for example).2 Entertainment trading can also be motivated by sensation seeking in the financial domain. complex. 1993) and exploited in advertising by retail brokers (Barber and Odean. Such aspirations have been used to explain lottery participation (Conlisk. They report that the univariate correlation between the self-confidence score and trading activity is close to zero. anticipating the outcome of.. Mazurier. directly lowering their marginal cost of trading. as pointed out by Grinblatt and Keloharju (2008). sensation seeking. The inadequacy of standard theory opens the door for the behavioral approach. This paper explores a different explanation of why people trade. 2 . Exposure to such stimuli by itself may trigger trading as argued by Dorn and Huberman (2007). trading can be motivated by an aspiration for riches as suggested by Statman (2002). 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. however. 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). 2008). the monetary cost of trading is offset by non-pecuniary benefits from researching. novel. If rational investor A offers to trade with rational investor B. “Sensation seeking is a trait defined by the seeking of varied. 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. hobby investors have less of a psychological hurdle to overcome when executing changes to their portfolio. In addition. overconfidence allows both parties to a trade to believe that they will win the zero-sum game of trading. None of the proxies help explain cross-sectional variation in portfolio turnover.38 and only trades if the benefits from trading justify the costs.] financial risks for the sake of such experience. 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. In trading experiments with students from different countries. Alternatively. 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. talking about. Grinblatt and Keloharju (2008) use traffic violations to proxy for thrill seeking behavior.” In a (perhaps subconscious) quest for arousal. and an aspiration for riches. they also expose themselves to more trading signals and should hence be expected to trade more than their peers. By actively following the financial markets. The intuition is simple. Hilton. 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.” that is. sensation seekers look for both intensity and novelty in experience. Motives for entertainment trading can be classified in three distinct groups: recreation.

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. Turnover due to savings. mutual funds. the typical client holds the equivalent with trading activity. 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. consider purchases and sales of individual stocks. The responses to these each month (similar to ShareBuilder in the US). our sample A. we justified by standard trading motives. That is. defined as one half the sum of 2. and during the sample period rebalancing considerations January 1995 to May 2000. term deposits. Judging from a survey I. 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 1.000 US dollars [USD] at the average USD/ standard trading motives. their birth date. Section at account opening. Individual holdings. that is. summarizes the client Turnover due to savings. 2000). Entertainment-driven investors turn DEM exchange rate of 1. liquidity. and options at We analyze the portfolios using a measure of concentration roughly twice the rate of their peers. riskier portfolio components. 2000. Stock mutual funds are assumed to consist of one hundred equally-weighted positions that do not overlap with other holdings of the investor. Entertainment appears to be a major driver of portfolio bonds.7 during the sample period. Standard motives for subjective investor attributes trading fail to explain much of detailed below. The investment or withdrawal plan that exist for dozens of survey offers responses to statements that elicit whether a individual stocks and mutual funds. each Standard motives for trading such derived from trading. 4 . 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. Entertainment appears to be a major driver of portfolio the entire account life is roughly 90. funds. is 15%. Entertainment-driven investors hold more stock trades account for 62%. statements serve as proxies In July and August 2000. and term deposits. either directly or through mutual funds (see Deutsches Aktieninstitut. month divided by the average portfolio value during that dissavings. The average portfolio size over 4. during the sample period. known as the Herfindahl-Hirschmann Index (HHI).DORN. 1995 and May 31. 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. we can infer the gender of all main account holders and the age of those who choose to report III concludes. especially turnover that cannot be justified by [DEM] or 50. fund trades account for 18%. The remainder of the paper proceeds as follows: Section I of an equally-weighted portfolio of three individual positions. over their portfolio of stocks. accounts only for about one third of total turnover. & SENGMUELLER — WHY DO PEOPLE TRADE? 39 The paper examines the hypothesis that entertainment In addition to the standard trade attributes. or within an automatic at one of the top three discount brokers in Germany. options. dissavings. The Data of Germans who hold stocks. young. Table I trading activity and the variation the observed trading activity.4 The 5. From the information provided by the client to the broker describes the data and the construction of the variables. A portfolio consisting of n equally-weighted stocks would have an HHI of 1/n. DORN. over the internet. for the entertainment benefits after the sample period.000 Deutsche Mark turnover. Section II discusses the main findings in more detail.3 The broker is labeled as a discount broker because no investment advice is given. Proxies for overconfidence are at best weakly correlated median HHI of the stock and fund portfolios during the sample period is 31%. In our turnover calculation. liquidity. Average monthly turnover.345 current and former clients at one of Germany’s three largest discount brokers between January 1. and has held the account for three years. mutual funds. bonds. portfolios and trading activity of such activity across individuals. and and option trades account for 15% of the total trading volume portfolios with more positively skewed returns. concentrated portfolios. and options obtained for a sample of 1. 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. 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%.000 clients over the phone. bonds. individual 3. 3 The HHI is defined as the sum of squared portfolio weights. The typical respondent is male.

and term deposits in the client’s portfolio.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. Agreement with statements two to four identifies respondents who enjoy risky propositions.46 between statements three and four. The investors are asked to indicate their agreement with the four statements . (3) tend to agree. in general. “Self-employed” is a dummy that is one if the respondent reports to be self-employed and zero otherwise.000 373. “Wealth” is the self-reported total net worth (including all financial assets and real estate). Dorn and Huberman (2005) describe the survey in detail. 2000 or the day when the account was closed. 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. the fascination increases with the size of the bet. 3. Turnover in a given month is the sum of the absolute value of purchases and sales of stocks.345 sample investors from the day when the account was opened until May 31. whichever comes first.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. we replace the missing value with the age recorded for the main account holder in the brokerage database). 2005). we replace the missing value with the gender recorded for the main account holder in the brokerage database). During the sample period. (2) tend to disagree. Agreement with statement one defines a hobby investor. Table II summarizes objective demographic and socioeconomic attributes of investors grouped by their responses B. “Age” is the age of the respondent (if missing. we focus on the survey items that make an explicit reference to whether or not respondents enjoy dealing with their investments or enjoy gambling. options. 1985). Average portfolio value is calculated at the end of every month across all individual stocks.000 36 88. Survey Variables To gauge which investors likely derive non-pecuniary benefits from their trading activities. one US Dollar [USD] corresponds to roughly Deutsche Mark [DEM] 1. in fact statements three and four are taken from a study on identifying compulsive gamblers (Nadler. “Gender” is a dummy variable that is one if the respondent reports to be male and zero otherwise (if missing. “College” is a dummy that is one if a respondent has a college degree and zero otherwise. Hobbyists and gamblers appear to form distinct groups. and gambling. bonds. portfolio structure. Higher values of the HHI indicate less diversification.000 325. This focus yields responses to a total of four statements (reproduced below in translation from the original German). I enjoy risky propositions. 4. I enjoy investing. and demographic and socioeconomic status.000 typical German stock market participant. investment experience and knowledge. Relative to the population of German stock market participants. risk attitudes and perceptions. mutual funds. “Income” is the self-reported gross annual income. Statements two through four flag investors as gamblers more or less consistently. (4) strongly agree. 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). bonds.7. funds.000 31% Median 7% 38. In gambling. 2. Games are only fun when money is involved. the pairwise correlation between the responses to statements two to four is quite high and reaches 0. The survey elicited information on the investors’ investment objectives. To pin down the importance of entertainment motives for different investors. we use their self-reported attitudes towards investing and gambling gleaned from a survey administered in July and August 2000. Portfolio Characteristics Average monthly portfolio turnover Average portfolio value [DEM] Average Herfindahl-Hirschmann Index Mean 15% 86. on a five-point scale ranging between (1) strongly disagree. the sample investors also appear to be more highly educated and earn higher incomes (see Dorn and Huberman. the response to the first statement is only weakly correlated with the responses to the other statements. in particular. and (5) don’t know: 1.

” “Nobs” is the number of respondents in each category. the fascination increases with the size of the bet. tend to be younger. we combine the “strongly agree” and “tend to agree” categories for statement four as only 38 investors “strongly agree. To be able to make meaningful statistical comparisons across groups.” For the same reason. 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. in particular. *Significant at the 0. the number of respondents who respond with “don’t know” ranges from 11 (for statement one) to 56 (for statement four). In Panel A.” 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 . it is interesting to note that younger age. Those who enjoy games only when money is involved.” 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 . The investors are asked to indicate their agreement with the four statements on an ordinal scale of (1) strongly disagree. and less wealth have been linked to a higher propensity to participate in legal . and less wealthy.10 level. The demographic and socio-economic variables are defined as in Table I.05 level. a lower level of education.” 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 . (4) strongly agree.Statement 4: ``In gambling. Although we have no direct information about whether our sample investors engage in gambling outside the stock market. DORN.Statement 3: “Games are only fun when money is involved." 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.” In Panel D.345 respondents. Nobs Gender Age College Selfemployed Income Wealth Panel A .01 level **Significant at the 0.Statement 1: “I enjoy investing.DORN. we have combined the categories (3) and (4) to “agree” since only thirty-eight respondents choose to “strongly agree. (3) tend to agree. the number of missing responses ranges from 10 (for statement three) to 15 (for statement one).” Male investors and wealthier investors appear to enjoy dealing with investments more than their female and less wealthy counterparts. We exclude the few investors with missing responses and investors who respond with “don’t know” — out of a total of 1. less well educated. to the above statements. & 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. we have combined the categories (1) and (2) to “disagree” since only four respondents choose to “strongly disagree. (2) tend to disagree.Statement 2: “I enjoy risky propositions.

liquidity. 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). 1997). the challenge in explaining the A. 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. 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. finer measures: normal turnover and excess turnover. past The standard deviation of normal turnover across the returns. dissavings. fund. 5 6 Barber and Odean (2002) use the terms “non-speculative” and “speculative” trades instead of normal and excess trades. investors who exhibit high excess turnover. Dorn and Sengmueller (2008) go on to document explained by savings. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 normal as they are likely motivated by liquidity and savings considerations.42 forms of gambling in Germany (see Albers and Hübl. All concentrated equity portfolios. For example. motives. or option that is followed by one or more stock. In particular. Normal turnover consists of trading that can be explained by standard B. in part because we only observe part of the portfolio for some investors. investors appear to be fairly homogenous in their desire to trade due to savings. separated by their responses to our four statements. 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. Second. or small accounts. 1999). we define an excess Table IV sets out the portfolio characteristics of investors. by contrast. For example. sale as a sale of a complete position of an individual stock. that these results are not driven by survey response bias. fund. these variations in the definition of excess turnover have little effect on our results. excess turnover is the portion of Entertainment Hypothesis total turnover that cannot be explained by these motives. We that those investors who are most excited by risk indeed hold define excess purchases as all stock. sample respondents is 4% as opposed to 29% for excess turnover. One could also argue that put purchases are used for portfolio insurance purposes. Across the turnover and lagging returns bonds. 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. or rebalancing motives. .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. liquidity. 1989). liquidity. We divide the turnover we observe in our sample into two liquidity. and rebalancing motives. Substantively. 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. they restrict their analysis to trades in common stocks. Similar to Barber and Odean (2002). they do not distinguish between savings plan and non-plan trades. Self-professed gamblers in our sample hold more purchases made within three weeks of an excess sale. them as hobby or gambling only one third of the observed trading volume can be investors. and the US (see Clotfelter and Cook. the UK (see Farrell and Walker. Third. First. or Characteristics Are Consistent With the rebalancing considerations. those who 5 other trades are classified as normal trades. and options at sample respondents. Since most trading occurs in individual stocks and call options. 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. Cross-sectional Differences in Portfolio motives for trading such as savings. those individuals turnover and 10% excess peers. or option purchases within three weeks of the sale. Therefore. We find mutual fund. and option more risky portfolios. For example. our classification differs from theirs in three ways. funds. Standard Motives Inadequately Explain the heterogeneity in trading activity across investors appears to Observed Trading Activity lie in understanding excess turnover. whose survey responses label turnover — in other words. 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.6 Our trade classification likely overstates normal turnover. or trading for entertainment rebalancing motives. an investor might sell off a complete II.

An excess sale is defined as a sale of a complete position of an individual stock. Excess turnover is defined similarly.4% 3.01 level **Significant at the 0.``I enjoy investing.32*** 22% 29% 46% 51%*** Panel C .50 0.52 0.45 0. In particular.05 level. DORN. ACV. All other trades are classified as normal trades. Mean Average monthly portfolio turnover thereof: normal turnover excess turnover 15% 5% 10% Std 32% 4% 29% Median 7." Disagree Tend to agree Strongly agree 84 403 822 30% 30% 31% 42% 42% 45% 18% 29% 41%*** Panel B . HHI. but using excess purchases and excess sales.``In gambling.9% Table IV: Portfolio Choices of Entertainment-Driven Investors HHI is the average Herfindahl-Hirschmann Index across the portfolios in the group. the fascination increases with the size of the bet. Average component volatility (ACV) is the value-weighted average volatility of the portfolio components in an investor’s portfolio." Strongly disagree Tend to disagree Tend to agree Strongly agree 148 571 492 87 26% 28% 33% 39%*** 39% 41% 48% 52%*** 0." Strongly disagree Tend to disagree Tend to agree Strongly agree 470 487 277 71 28% 30% 34% 38%*** 42% 43% 48% 53%*** 0. fund.DORN. fund.07 30% 34% 46% 52%*** Panel D . Nobs HHI ACV Realized Skewness 0." Strongly disagree Tend to disagree Agree 674 396 199 28% 32% 37%*** 42% 45% 50%*** 0. and skewness are calculated using only the individual stocks and stock mutual funds for which Datastream provides daily total return data. . An excess purchase is defined as a stock.45 0.92 0. or option that is followed by one or more stock. higher values indicate less diversification. “Options” is the fraction of respondents in a group that have traded options at some point during the sample period. Realized skewness is calculated from daily portfolio returns as in Chen et al. or option purchase made within three weeks of an excess sale.53 0.87 1. (2001).``I enjoy risky propositions.77 0. all trades in term deposits and automatic investment and withdrawal plans are classified as normal.63 0.9% 2.74 Options Panel A . or option purchases within three weeks of the sale. mutual fund.91 1. & 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.``Games are only fun when money is involved.96** 31% 42% 47%*** ***Significant at the 0.

only one out of five investors who strongly disagree these correlations between turnover and entertainment with this statement also trade options. or enjoy investing”) but not gambling (that is. liquidity. one would expect such differences to manifest portfolio components — of investors who strongly agree with themselves in terms of excess turnover. For example. and wealth. 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). Figures 1-4 illustrate investing nor gambling. anticipating. For with the statement (see Figure 3). who strongly disagree with Figures 1-4 also show that Entertainment as a motive for trading this statement. in part because of a lack of high-frequency price 5% — similar to the average normal turnover of 4. Indeed. For example. The wealth of survey responses allows us to construct three . preferring skewness (see those who enjoy investing or Proxies for overconfidence are at best Golec and Tamarkin. income. or rebalancing relative to 39% for investors considerations. talking about. 1998). 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. executing. and those who enjoy gambling offset the cost of trading.44 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 involved” turn over their portfolios at an average monthly weighted portfolio of four stocks. they disagree or experiencing the outcome of a trade.5% of data. and investors who enjoy both trade the most of Hypothesis all. the average excess turnover rate of the selfpositively skewed payoffs. they strongly agree with the statement “I researching. motives hold up under multivariate statistical analysis. members of each group. rate of 24% — twice the rate of those who strongly disagree but they also consist of individually riskier securities. Cross-sectional Differences in Turnover who enjoy investing or gambling trade more than those who Are Consistent With the Entertainment enjoy neither. For example. entertainment-motivated investors preferring securities with However. investors C. Not only are the portfolios of gamblers more concentrated. Investors who report enjoying investing also trade more aggressively than their peers. Alternatively. 19%. However. propositions” trade options during our sample period. portfolios of stocks and entertainment-driven mutual funds that exhibit investors. those who enjoy both investing and the equally-weighted average monthly turnover rates for the gambling trade the most. half of the investors professed gamblers. example. And indeed we find that even after controlling for gender. the average component volatility — the valueIf differences in trading activity were indeed driven by weighted average of the annualized volatility of the stock entertainment. 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. education. 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. is almost thrice the corresponding who strongly agree with the statement “I enjoy risky rate of their peers (see Figure 3). entertainment might amplify the effects of overconfidence or vice versa. Figure 5 illustrates that respondents who enjoy here is that some investors derive non-pecuniary benefits from investing (that is. virtually the entire difference Consistent with gamblers in total turnover between is distinct from overconfidence.5% of his self-reported gross annual income on trading commissions. options holdings and trades also point to their peers who strongly disagree with the statement. The typical respondent in the paper’s pleasure from trading both as a pastime and as a form of sample spends 0. investors who strongly agree with “Games are only fun when money is D. 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. age. 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. These results suggest that investors appear to derive Trading is costly. These benefits help strongly disagree with the statement “Games are only fun when money is involved”). turnover the statement “I enjoy risky propositions” averages 52% unlikely due to savings. employment status. The main hypothesis entertained gambling. that is. in In Dorn and Sengmueller (2008) we investigate whether contrast. more positively skewed returns.

" Figure 2: Turnover as a Function of Enjoyment of Risky Propositions 45% 40% Total turnover 35% 30% 25% 20% 15% . DORN." . 10% 5% 0% Strongly disagree Tend to disagree Tend to agree Strongly agree Normal turnover Excess turnover Agreement with "I enjoy risky propositions. & 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.DORN.

the fascination increases with the size of the bet.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% ." ." 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. 5% 0% Strongly disagree Tend to disagree Tend to agree Strongly agree Agreement with "Games are only fun when money is involved.

and to ensure that the resulting groups consist of enough members. To estimate the self-enhancing attribution bias.7 III. Moreover. However. due to my specific skills. 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. and the erroneous expectation of being able to affect chance outcomes (known as the illusion of control.DORN. I control and am fully responsible for the results of my investment decisions. 1998. above all. or relative knowledge. we consider the extent to which survey participants agree with the statement “My past investment successes were. see also Barber and Odean. 2002. we find that none of the overconfidence proxies is significantly related to excess turnover. it appears that the underconfident who do not enjoy investing trade less than their overconfident peers. 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. no additional insights were found in this interaction. I am in control of my personal finances. I am certain that they will work out. The results of interacting the investor’s agreement with “I enjoy investing” with the three overconfidence proxies are shown in Figures 6-8. overconfident hobby or gambler investors turned over their portfolios at similar rates as underconfident investors with the same hobby or gambling affinities. We also investigated how our hobbyist and gambler designations interact with overconfidence. we compute an aggregate score using the investors’ responses to four statements: 1. 7 . Daniel et al. 2001). Conclusion Some investors derive enjoyment from trading which offset In Figure 6. When I make plans. the difference in trading activity is not statistically significant. In regressions reported in Dorn and Sengmueller (2008). and Gervais and Odean. 3. the significance of our hobby and gambler proxies is maintained even while controlling for overconfidence. the tendency to overly attribute successes to skill in conjunction with past returns (known as the self-enhancing attribution bias). In general. we created binary entertainment and overconfidence proxies. 2.. To simplify the presentation. I always know the status of my personal finances. 4.” To construct a proxy for the illusion of control.

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 .

Consistent with this conjecture. Most of the variation in trading activity across individuals is variation in excess turnover. 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. The most entertainment-driven investors trade about twice as much as those who fail to take pleasure in gambling or investing. & 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. DORN. and rebalancing.DORN. Like lottery players who buy tickets with negative expected values. independently of the survey responses). Variation in excess turnover is highly correlated with our proxies for nonpecuniary benefits derived from trading. entertainment trading appears to be quantitatively important — at least for this sample of discount brokerage customers during the late 1990s. . In sum. variation in the self-reported enjoyment of investing and gambling explains variation in trading intensity even after controlling for competing explanations such as overconfidence. Relying solely on transaction records (that is. entertainment-driven investors trade even though trading diminishes the expected monetary payoff of their portfolio. liquidity.

R. Factbook 1999. and P. Tirole.” Drexel University Working Paper. P. and G. “A Simple Model of Capital Market Equilibrium With Incomplete Information. Hirshleifer. Hong. Stokey. 2005. 385-410.D. and M. 345-381. 1982. “Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. “Sensation Seeking. 2003.” Review of Economic Studies 72 (No. Golec. and M. Thaler. 1).” Journal of Finance 53 (No. 6). Not Risk. 1887-1934. Sengmueller. at the Horse Track. “Talk and Action: What Individual Investors Say and What They Do. N. T. M. B. 1).” University of Mannheim Working paper. Past Returns and Conditional Skewness in Stock Prices.T. J. and T. Maksimovic and W. Price and Profit When All Traders are Above Average. Chen.2). 1989.. 773-806. Forthcoming. 125-144.” Econometrica 50 (No. Gervais.2). Harvard University Press. 2001. and M. and A.. “Trading as Entertainment?” Management Science. Zuckerman.. 1). “On the Possibility of Speculation under Rational Expectations. V. and G.” Journal of Financial Economics 61. 261-292.C. 2001. Statman. “The Epidemiology of Pathological Gambling: Critique of Existing Research and Alternative Strategies. “Information. and G. D.. E. Barber. 2). 1-27. 5). Merton. L. Huberman.” Journal of Gambling Behavior 1 (No. B. B.. Dorn. and Common Stock Investment. Odean. Grinblatt. 1997.” Journal of Political Economy 106 (No. Barber. 3). 1999.” Working Paper. Forthcoming.M. 2008. Overconfidence. Farrell.” Review of Financial Studies 14 (No. and I. 1).50 References Albers. C. Clotfelter. B. 3).and Overreactions. Ziemba (eds). L. Keloharju. 2002.. “Financial Decisionmaking in Markets and Firms: A Behavioral Perspective. D. North Holland. “The Welfare Effects of Lotto: Evidence from the UK. “Online Investors: Do the Slow Die First?” Review of Financial Studies 15 (No.” in R. 35-50. Forthcoming.. Y. Pouget. “An Experimental Test of the Impact of Overconfidence and Gender on Trading Activity. Hilton. 2001. Frankfurt am Main. 2004.” Journal of Public Economics 72. 1993. Huberman. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Dorn. 287-312. 1839-1886. M. 455-487. Daniel. K.. “The Utility of Gambling. J. Dorn. and P. 17-27..” Journal of Finance 53 (No. Selling Hope: State Lotteries in America.” Journal of Finance 42 (No. Cambridge University Press. Kumar. Vol. 1982. D.C.” Financial Analysts Journal 14-21. 437-481. T. MA. Volatility.” Journal of Economic Theory 26 (No. Deaves. S. and S. Stein. K. and L. Lüders. M.A. Subrahmanyam. Odean. . 1163-1182.. Cambridge. Barber. D. “Who Gambles in the Stock Market?” Journal of Finance. Learning to Be Overconfident. Cook. J. and T. Glaser. 2008. “Judgmental Overconfidence. “Lottery Players/Stock Traders. 4). 1994. Odean. Finance. 1995. H. R. A.” Quarterly Journal of Economics 116 (No. Conlisk. “Turnover and Volatility. Handbooks in Operations Research and Management Science. 2008. “Volume. 529-543. 1985. Jarrow. Nadler. and T. Overconfidence. “Noise. “Boys Will Be Boys: Gender.. Review of Finance 9 (No. 255-275. J. DeBondt. 1998. Trade and Common Knowledge. 205-225. F. Odean. 2005. 9. “Overconfidence and Trading Volume.” Journal of Gambling Studies 13 (No. Walker. Self-monitoring and Trading Performance in an Experimental Financial Market. D. 2002. H. Deutsches Aktieninstitut.” Journal of Finance 55 (No. and N. and Trading Activity. “Bettors Love Skewness.” Journal of Finance. 1998. Tamarkin. “Gambling Market and Individual Patterns of Gambling in Germany. Odean. Amsterdam. Black. M. Luo. 2000. 483-510. 6). 99120. “Forecasting Crashes: Trading Volume. “Investor Psychology and Security Market Under.. 1986. W. 1998. Hübl.” Journal of Risk and Uncertainty 6. and R.F. and T. Weber. Mazurier. 2000. 1).” Journal of Finance 41 (No.. 2007. 2). Biais. Behavioral Expressions and Biosocial Bases of Sensation Seeking. J. chapter 13. 1987. and J. Milgrom.

J. They cover the months of January and February. 26. 2001 as the pre-decimal period.22 between the first quarter of 1998 and the last quarter of 2000. Asymmetric information can have an important impact on microstructure. and Van Ness (2005) examine the effect of the same event on adverse selection costs. 1999-January 31.22% of the public information on Boeing stock came from transactions on its shares during that period. on average. Zhao and Chung (2006) analyze the effect of NYSE decimal pricing on the probability of informed trading. 2001. The evidence suggests that trade informativeness was related to institutional buying and both uninformed and informed trading.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. 2001 as the post-decimal period. 2000-January 28. Collver (2007) uses August 1. The results indicate a positive relation between institutional buying and trade informativeness before Regulation Fair Disclosure and decimal pricing. the evidence is consistent with both a rise in uninformed trading and a fall in informed trading. Van Ness. Trade informativeness is defined as the percentage of efficient price variance that is attributable to trades (Hasbrouck (1991)). 1 51 . Therefore. While the decrease in informed trading may be a continuation of the decimalization effect. measure of the degree of asymmetric information (relative to total information) in the market for the security under study. They define November 1. Chakravarty. implementing trade informativeness. Rakotomavo is an Assistant Professor of International Business Administration at the American University of Paris in Paris. Trade informativeness is defined as the percentage of efficient price variance that is attributable to trades. They conclude that the percentage adverse selection cost has increased and the dollar adverse selection cost has decreased. France. 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. and June 1. 2001-August 31. and focusing on a longer term (1998-2004) that includes the enactment of the Sarbanes-Oxley Act (SOA). Logistical support was provided by the Andrew Batinovich Trading Room and Research Center. For example. This paper complements the above studies by using a different information asymmetry metric. While it is well known that some traders have superior information. this percentage for Boeing was. about 26. By decomposing the variance of changes in the efficient price into its trade-correlated and uncorrelated components. After these events. For example. 1 studies of the dynamics of asymmetric information among investors have focused on short time periods. 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. Hasbrouck (1991) proposed the ratio. as illustrated by Whitcomb (2003). of the trade-correlated component to the total variance as a Michel T. I thank an anonymous referee and Betty Simkins (Editor) for their constructive comments. Similar results are found for the period following the enactment of the Sarbanes-Oxley Act (SOA). trade informativeness. They find that the post-decimal probability of informed trading is greater than its pre-decimal equivalent.

then: Hypothesis 5: An increase in price. p. Jiang. over time.”4 Hence: I. for example. if informed trading affects the time variation of trade informativeness. over time. ceteris paribus. Therefore. over time. and Titman (2006) report a positive correlation between changes in institutional holdings and contemporaneous stock returns. This paper assumes that any residual institutional buying of a stock. holdings levels would capture the intensity of any lagged herd net buying. then: Hypothesis 6: An increase in the percent price range. and Titman (2006) find that changes in institutional holdings. ceteris paribus. then: Hypothesis 3: An increase in trade size. Chiyachantana. Therefore. implies an increase in the informativeness of trades. and Wahal (2002) for a review of the theory and evidence on institutional trading. They also show that institutional investors use their information advantage to buy before dividend increases. implies an increase in the informativeness of trades. Hence. and 2. 857.3 This implies that institutions have decreased the size and increased the frequency of their trades. are correlated with contemporaneous stock returns because of information effects. 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. ceteris paribus. The basis of this assumption is the herding behavior of institutions evidenced. 3 4 See Coughenour and Deli (2002) p. Therefore. ceteris paribus. over time. and went further down to 3.77% after Regulation Fair Disclosure and decimal pricing.) analyzing the temporal variation of trade informativeness in terms of microstructural variables. Similarly. implies an increase in the informativeness of trades. Hypothesis 7: An increase in depth. as previously mentioned. Jain. if institutional buying affects the time variation of trade informativeness and institutions have an information advantage.52 Fair Disclosure (RFD) and the switch to decimal pricing. a greater market depth may increase profit. For Boeing. implies an increase in the informativeness of trades. implies an increase in the informativeness of trades. Starks. it may attract more informed trading). which is not captured by net institutional demand. both Nissim and Ziv (2001) and Garrett and Priestley (2000) find that only dividend increases are directly correlated with earnings. implies an increase in the informativeness of trades. Interestingly. over time. This conclusion is consistent with holdings changes being a measure of informed buying intensity. over time. The main results of this paper are illustrated with Boeing data in Section II. Hypotheses Amihud and Li (2006) find that the abnormal returns at dividend change announcements is a decreasing function of institutional holdings. the average quarterly trade informativeness of 26.22%.878) by institutions over time. . Johnson. is correlated with the level of institutional holdings. over time. Hypothesis 2: An increase in the change of institutional holdings. thus enabling a distinction between the various causes of such variation. Sias. ceteris paribus. ceteris paribus. 2 see Conrad. which represent net institutional demand. Therefore. 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. Starks. the percent of dollar volume executed at the NYSE could reflect the degree of informed trading.66% after SOA. ceteris paribus. in Nofsinger and Sias (1999). 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. Section III concludes the paper. Both Nofsinger and Sias (1999) and Sias. this paper hypothesizes that institutional buying may increase trade informativeness because of institutions’ information advantage.2 Collver assumes that the reduction is caused by less informed trading. 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. The Kyle model (1985) suggests that a greater volatility may increase the profit of informed traders (therefore. implies a decrease in the informativeness of trades. Hypothesis 4: An increase in trade frequency. This paper extends his finding by 1. If institutional buying affects the time variation of trade informativeness. Section II discusses the data and results. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Hypothesis 1: An increase in the level of institutional holdings.) considering a longer term that includes the post-SOA period. Section I develops the paper’s hypotheses on the temporal covariation between the informativeness of trades and some microstructure variables. and Wood provide international “evidence of an increased use of order-breaking strategy” (2004. decreased to 16.

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. on Boeing stock came from transactions announcements. and are computed for each of the 82 A. (4) ask) data. this price variance. will be investigated shortly. Three trade attributes are considered: the sign of the trade.t firms and each of the 28 quarters between 1998 and 2004. Transactions occurring within 5 seconds of each other without any intervening quote are aggregated. 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 . resulting in 2. trade size. These. A quote posted within less than 5 seconds prior to a trade is resequenced.5 Two notable patterns are the declining time trend for trade informativeness and the clear break in the depth data after 2001 Q1. and t is the time of a transaction or a change of quote. More recently. bid and vw = vw.22 between the first quarter of 1998 the variation of trade 82 firms and each of the 28 and the last quarter of 2000.RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS 53 . and other patterns.t-10 (2) II.var (v2.11 fi). The data are from the NYSE TAQ intraday trade and quote database. financial institutions.t) . A moving average representation of the VAR in equation (1) is computed as follows: rt = v1.t-1 + … + e11 v1. To aggregate values and test hypotheses.edu/~mrakotomavo/JAFdata.t-1 + … + f11 v2. The data are from the Thomson Financial base. C.aup. Therefore. Trade Informativeness In Hasbrouck (1991). implies an increase in the informativeness of trades.296 quarterly observations of 82 and randomly chosen firms for which 1998-2004 NYSE TAQ intraday trade (time.t + e2 v1. Al-Suhaibani B. trade frequency.296 about 26.11 ei)2.t). the percent of trades executed at NYSE.x + (1+i=2. The change in institutional holdings in quarter q for each stock is the difference in institutional holdings between quarter q and quarter q-1. The return is set to zero if no quote revision follows a trade within 5 seconds. (i=1.t-10 + f1 v2. Trade (1996) used a random informativeness is estimated percentage for Boeing was. volume. For a is divided by the full efficient comparison. Price. For example. + f2 v2. as well as 1998-2004 Thomson Therefore. where: vw. Other Variables on its shares during that period. (3) Our sample includes 2. the sample 5 These values are available at http://ac. A transaction that has a price above the prevailing quote midpoint is assigned a positive sign. on average. . and depth are estimated daily for each stock before the quartely averages (of the daily values) are computed. over time.x/vw.htm.x = (i=1. price range (as a percentage of the minimum price of the day). Sample Description and Preliminary Results Figure 1 shows the evolution of the median values of the previously mentioned variables from 1998 to 2004. 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. informativeness quarters between 1998 and around earnings and 2004. (1) where rt is the mid-quote return (logarithm differentials). Brooks is attributable to trades. the signed trade size.11 fi T) . size and price) and quote (time. trade informativeness (TINFO) is the ratio vw. Hypothesis 8: An increase in the percentage of NYSE executions. ceteris paribus. the opposite holds for a negative sign.22% of the public information d i v i d e n d values. This allows a comparison with the daily averages reported in the literature.var (v1. and the signed squared trade size. sample of 90 dividendfrom 1998-2004 NYSE TAQ paying stocks to study intraday data for each of the 26. xt is the column vector of trade attributes. Data and Results If w represents the innovation in the efficient price which is assumed to evolve as a random walk.

00E+07 2.56 -.40E+07 1.3 . D IHLDG is change in institutional holdings.2 .20E+07 2000 2001 2002 2003 2004 1.80E+07 2.40E+07 2.84 .83 . Sample Statistics Between 1998 And 2004 TINFO is trade informativeness.52 1998 1999 2000 2001 2002 2003 2004 1998 1999 2000 2001 2002 2003 2004 -.04 1200 1000 .00 .60E+07 1.72 . PCNTNYSEEXEC is the percent of trades executed at NYSE.4 .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. TRADEFREQ is daily trade frequency.08 1600 Median of IHLDG Median of DIHLDG Median of TRADESIZE .4 4.4 2. IHLDG is institutional holdings.02 .54 Figure 1.88 .89 .6 3.8 4.60E+07 Median of DEPTH Median of VOLUME .5 .20E+07 2.8 2.68 .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 4.1 .87 .85 .60 -.0 3.86 .64 .02 1400 .06 400 1998 1999 2000 2001 2002 2003 600 800 .2 2.06 . Median of TINFO .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 .04 .

This Coughenour and Deli (2002) for their September-November value is comparable with Hasbrouck (1991)’s average 1997 sample. The P1 median depth of 23.87% in P3. a reduction in dollar adverse selection after decimalization.78% that the quarterly variation of the before SOA. sample. The same time variations are observed within each Chi-square. RFD. and Van Ness (2005) who find in trade frequency after SOA could not be found. 2002.51 shares per day is consistent level.15% to 15. Regulation Fair Disclosure (RFD) became effective on October 23. 2001. its time trend has changed from is not offset by the observed decrease in the frequency of insignificant in P1 to negative in P2. 2000. Van Ness. For comparison. Grinstein and SOA. No other study seems to be available on this result. has not moved over time within P1. index running from 1 to The evidence in Panels 28) for each variable. and informed trading. median values and Panel these events. Similarly. Therefore. The results sample.596 group means reported in The median trade informativeness is 33. The same table provides P2. but culminating at 57. are consistent with this evidence.30 in Table II of the median trade informativeness drops from 15. uninformed The Jarque-Bera test beginning of P1) for their indicates a significant trading.07% in P2. the P1 and P2 throughout the paper. 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. pricing and RFD. or lower significance P1 median trade size of 1296. 54. period. RFD and SOA.3% for a sample of 177 firms on the NYSE transactions per day is consistent with the 282 average for 1989 Q1.35% average that Sias. Institutional holdings have increased from a median of 1. and after periods is tested. the evidence is consistent with Starks. for 1991-1996. 58.71% in P2 . although Jain and Rezaee (2008) find an improvement in market liquidity after SOA. after both RFD and decimalization.84 round lots is smaller than The same pattern is observed in P3. shown in decrease in trade informativeness after both the Figure 1. informativeness has decreased from 33. This indicate a positive relation between seems to have stayed leads to the use of institutional buying and trade constant over time. These results are medium and large trades. 2001-January 26.71% in 2001 pre-decimalization sample. using the Wilcoxon/Mann-Whitney. Furthermore. before decimal pricing. The Sarbanes-Oxley Act (SOA) was enacted on July 30.500 and 1. after decimalization. Van Ness. after decimal holdings for their sample. The statistics are significant at the 10%.RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS 55 is divided into subperiods. The median departure from normality change in these holdings of the panel data. However. The NYSE’s decimalization was fully implemented on January 29. another implementation of RFD and decimal pricing. They also agree study documenting a decrease in trade size and an increase with Chakravarty. However. 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 . Therefore. Panel B suggests that trade informativeness (116+153+13) reported in Table 4 of Chakravarty.) 2002 Q3-2004 Q4 and Li (2006) have the (P3) after decimal stocks between 1998 and 2004 was related median value climbing at pricing. before informativeness of trades for NYSE-listed P1. and Van Ness (2005) for their January 1.32 and 38.) 2001 Q1-2002 Q2 report increasing median This paper presents evidence suggesting (P2).73% in P1 to 68.38% Panel A of Table I Disclosure and decimal pricing. and the difference is evidence of an increase in the frequency of small trades that significant. Amihud 3. Kruskal-Wallis. most notably. and Van der Waerden statistics.47% in 1998 (the to institutional buying. informativeness before Regulation Fair values of . The P1 median trade frequency of 294. Median SOA. 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.) 1998 Q1-2000 Q4 (P1).19 estimate of 34. After are comparable with the contains the periodic . three periods are considered: to 1. after SOA is enacted: the subsample averages of 36. the changes consistent with Collver (2007)’s finding of a significant in trade size and frequency between P1 and P2.37% and . SOA may have lowered trade informativeness. and the time trend is still negative in P3. in nonparametric tests particular. with the 1. and Michaely (2005) 2.15% for P1.

30 0.06 0.4995*** -0. for each stock. Median Chi-square.3345*** -0.5219*** JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 ***Significant at the 0.3512*** 0. The time index equals 1 at the start of each period.3766*** -0.0948*** 0.25 3.7210*** 0.0037 0.3377*** 0.2007*** -0.95 19.1282*** 0.2097*** 0. The sample contains 2.12 1263. postSarbannes/Oxley period.2219*** -0. Descriptive Statistics This table shows the descriptive statistics on the data used in the paper.3891*** 0. The hypothesis of equality is rejected when at least 3 out of the 4 statistics are significant at the 10% or lower level.56 18. and depth are estimated daily. trade size.2736*** -0. The ranking of values is repeated over all firms.25 26.4906*** -0. **Significant at the 0.1160*** 0.8446 1296. volume.0053 0.0038 643. Rank Correlation with the Time Index by Periods Change in Institutional Holdings -0.2514*** -0.0126 0. The data are from the NYSE TAQ intraday trade and quote database. before the quartely averages are computed. is the percentage of the efficient-price variance attributable to trade innovations.84 27. price range (as a percentage of the minimum price of the day).8720 0.0191 -0. between 1998 and 2004. Kruskal-Wallis.06 11.0715 1998 Q1-2000 Q4 2001 Q1-2002 Q2 2002 Q3-2004 Q4 All periods -0.3843*** -0.0481 0.2581*** 0.1864*** -0.2327*** -0.0651 0.6807 0.51 294.05 level. the percent of trades executed at NYSE.19 23.0908*** -0. It is computed by following Hasbrouck (1991) and estimated for every quarter between 1998 and 2004 from NYSE TAQ intraday trade and quote data. 2002 Q3-2004 Q4 (P3) is a post-decimal pricing/Regulation Fair Disclosure.7862*** Trade Size Trade Frequency Depth Volume NYSE Executions Price Range Period Trade Informativeness Institutional Holdings 0.2249*** -0.3338*** 0. Price.3623*** 0.5713*** -0. trade frequency. Panel A uses the Wilcoxon/Mann-Whitney.296 quarterly observations.56 Table I.5873 3.0187 P1=P2 P2=P3 P1=P3 No No No Yes No No No Yes No No No No Panel B. of 82 stocks listed on the NYSE.5944*** 0.01 level.01 2.79 8.52 673. 1998 Q1-2000 Q4 (P1) is a pre-decimal pricing/Regulation Fair Disclosure.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. 2001 Q1-2002 Q2 (P2) is a post-decimal pricing/Regulation Fair Disclosure. The change in institutional holdings in quarter q for each stock is the difference in institutional holdings between quarter q and quarter q-1. and Van der Waerden tests of equality of median values across different periods.14 Trade Size Trade Frequency Depth (Round Lots) Volume ($Millions) NYSE Executions Price Range (%) Period Trade Informativeness 1998 Q1-2000 Q4 0.3315 0.74 2001 Q1-2002 Q2 0.84 17. pre-Sarbannes/Oxley period. Panel A.1720*** 0. pre-Sarbannes/Oxley period. Trade informativeness. Median Values by Period Institutional Holdings Change in Institutional Holdings 0.0608 0.3765*** -0.8653 33.0373 0. 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.87 935. .0783** -0.0869*** 0.2714*** 0.6149 0.

For P1. The P2 variable has a negative and significant coefficient. Instead. The previous results are further detailed by restricting the observations to each of the three periods (and removing the dummy variables). This supports the hypothesis that institutional buying leads to greater informativeness of trades. is computed for each stock. of trade informativeness is used in the multivariate analyses that follow. There seems to be a decrease also after SOA. followed by a decrease in P2. not the value. which predicted a positive relation. All Jarque-Bera tests on residuals cannot reject the hypothesis of normality. • institutional holdings. range. Van Ness. The evidence suggests that volume has not changed after RFD/ decimalization. it is consistent with a rise in uninformed trading between 1998 and 2004. The results are shown in the first column of Table II. Ordered probit models are used throughout this section. • range. which goes from 1 to 28. This figure is comparable with 84. However. • log(trade frequency). The Kyle (1985) finding that volatility is positively correlated with informed trading profit is strengthened by the positive coefficient for price range.01% and 85. for this sample. The median price level has gone down from P1 to P2. For P2. the negative coefficient for trade frequency. • log(price). Van Ness.09 and $31. • log(trade size). Therefore. By hypotheses 1 and 2. This confirms the univariate result showing a decrease in trade informativeness after RFD/decimalization. The price range has increased over time within P1. The coefficients for price level. the rise in trade frequency (see Panel B. Trade informativeness rank.84 is similar to the subgroup averages of $32. with an increase at a small rate. This percentage seems to have not changed after RFD/decimalization. again confirming the univariate result. this suggests that the time variation of trade informativeness before RFD/ decimalization has been related to institutional buying.RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS 57 Coughenour and Deli (2002). This seems to indicate a drop in trade informativeness after the enactment of the Sarbanes-Oxley Act. • log(depth). these results point to a decline in informed trading. The use of log mirrors Coughenour and Deli (2002).05% subgroup averages for the 806 stocks sampled by Coughenour and Deli (2002). and Van Ness (2005). It also agrees with Collver (2007)’s finding of a significant decrease in trade informativeness and Chakravarty. and Van Ness (2005)’s finding of a reduction in dollar adverse selection after decimalization. Multivariate Results Because of departure from normality.58 million found in Table I of Coughenour and Deli (2002). pre-SOA) dummy variable.37 for the 806 stocks in Coughenour and Deli (2002). The P1 volume of $17. Change in institutional holdings has a positive and significant coefficient. Table I) suggest that a rise in uninformed trading may have taken place after RFD/ decimalization.66% reported in Table II of Coughenour and Deli (2002) for 1997.56 in P2 is consistent with Bessembinder (2003)’s finding of a depth reduction after decimalization. • change in institutional holdings. Table I). The decrease in range shown in Panels A and B. The coefficient for trade frequency is negative and significant. No other coefficient is statistically significant. The P1 median price of $33. The P3 variable has also a negative and significant coefficient. the coefficients for both level and change in institutional holdings 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. D. The first model uses all observations and the following explanatory variables: • a period 2 (post-RFD/decimalization.06 million per day is higher than the group means of $11.53%.99 million and $8. and depth are positive and significant. and the decrease in trade informativeness (see Panel B. between P1 and P2. P3 shows a reversal of that trend. the significant drop to a median of 11. . The median percent of trades executed at the NYSE in P1 is 86. the P1 median range of 3. but has increased after SOA. 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. This result does not support the institutional order-breaking hypothesis. • Percent of NYSE trade executions.25% is consistent with the 2. The evidence for P2 is consistent with Chakravarty. but it appears to have decreased after SOA. The evidence in Panel A points to an increase of depth within P1. is in step with the decrease in return volatility documented by Bessembinder (2003) after decimalization. The results are shown in the last three columns of Table II. • a period 3 (post-RFD/decimalization/SOA) dummy variable. the rank. 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. but does not seem to have changed from P2 to P3.

0855 (0. .9125) -0. and depth are estimated daily.0219 (0.2684) 0.17*** (-14. the percent of trades executed at NYSE. is the percentage of the efficient-price variance attributable to trade innovations.3256) 0. for each stock.3328*** (2.7113*** (-3.0289 (-1. Price. 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.07) 0.0410* (1. between 1998 and 2004.93) 0. 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.0285 (0.79) 0.05 level.0081 (0.0574 (1. trade frequency.9850* (1.0242** (2.21*** (-19. The data are from the NYSE TAQ intraday trade and quote database.8107 (1.39) 0. The ranking of values is repeated over all firms.10 level. Price.22) -2.58) Change in Institutional Holdings Ln(trade size) Ln(trade frequency) Ln(price) Range (%) NYSE Executions (%) Ln(depth) ***Significant at the 0.0588*** (3.4141 (0.0711** (-2.0553) 0.38) -0.296 quarterly observations.0511 (-0.0373 (0.0145 (0.38) -0.12) 0. *Significant at the 0. trade size. The data are from the NYSE TAQ intraday trade and quote database. 2002 Q32004 Q4 (Period 3) is a post-decimal pricing/Regulation Fair Disclosure.52) -0.4749 (1.4406 (-1.22) Trade Informativeness Rank (Period 1) Trade Informativeness Rank (Period 2) Trade Informativeness Rank (Period 3) 0.4941) -0.0103 (-0.0580 (1.59) -0. **Significant at the 0. before the quartely averages are computed. 2001 Q1-2002 Q2 (Period 2) is a post-decimal pricing/Regulation Fair Disclosure.04) 0.7020) -0.47) -0. of 82 stocks listed on the NYSE.10) 0. z-statistics are in parentheses. Trade informativeness (TINFO).1207 (-1.1686** (-2.2710*** (-3. pre-Sarbannes/ Oxley period.1171 (1.02) 0.58 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Table II. trade frequency. 1998 Q1-2000 Q4 (Period 1) is a pre-decimal pricing/Regulation Fair Disclosure. Trade Informativeness Rank (All Periods) Period 2 Dummy Variable Period 3 Dummy Variable Institutional Holdings -1. the percent of trades executed at NYSE.5248*** (2. The sample contains 2. post-Sarbannes/Oxley period. volume.3214** (2.73) 0.0414 (-0.1761) -0.76) 0. trade size. and depth are estimated daily. before the quartely averages are computed.01 level.3259) 0.6160) -0.73) -0.21) 0. for each stock.9124) 0.0335 (-0. pre-Sarbannes/Oxley period. It is computed by following Hasbrouck (1991) and estimated for every quarter between 1998 and 2004 from NYSE TAQ intraday trade and quote data.4097) 0. price range (as a percentage of the minimum price of the day).8179** (2.91) 0.0166 (-0.94) -0. price range.5340) 0. The change in institutional holdings in quarter q for each stock is the difference in institutional holdings between quarter q and quarter q-1.

26 (while the evidence indicates that trade size did not affect trade informativeness). RFD and SOA. Meanwhile. .22% in 1998 Q1-2000 Q4 (P1). is estimated for each of the three periods. thus illustrating the effect of institutional buying during this period. not prove. the evidence indicates that the level of institutional holdings affected trade informativeness negatively. has decreased over time. before decimal pricing. As previously mentioned. After SOA. However. Boeing’s trade informativeness went from 23. Regulation Fair Disclosure (RFD). like that of the average stock in this study. which may explain the reduction in adverse selection cost. The results suggest that trade informativeness was not affected by lags of institutional buying in the first two periods. 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. this result is consistent with a decrease in informed trading after SOA. 2-quarter) lagged institutional buying. which assume that institutional buying (including lagged herd buying captured by the level of institutional holdings) drives the time variation of trade informativeness. the negative coefficients are consistent with 1. in P3. the level of net institutional buying of Boeing shares per quarter went from an average of 0.77% in 2001 Q1-2002 Q2 (P2). to 15. but before SOA.32 round lots in P1). their level is a proxy for the amount of trades that are related to such past institutional actions.31% to 29.14%. For P3. Instead. and the strongest evidence of a decrease in adverse selection cost for trades of medium size. a probit model. and 2. after decimal pricing and RFD. suggesting that the hypothesized link between informed trading and trade informativeness exists. at a weaker significance level. since both institutional holdings and price increased. and 1-quarter (and. At the same time.33 round lots in the first half of P2. with a lag (the evidence in the paper points to a lag of one quarter).RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS 59 who find a reduction in dollar adverse selection cost after decimalization on the NYSE.15 to 3. To investigate these hypotheses. Since both range and trade informativeness decreased during this period (see Panel B. in P2. smaller decreases in the frequency of medium and large trades (which they interpret to mean less institutional trading).68% to 11. during that period. In P1. After decimal pricing and RFD. Table I). while trade informativeness decreased.79% of shares outstanding in the first half of the period to 1.) a relation between uninformed trading and lags of institutional buying after SOA. They suggest that institutions trade less because of lower liquidity supply (as evidenced by the smaller depths and smaller limit-order sizes).77%. trade informativeness for Boeing.55% to 2.25%.86%. This result is inconsistent with hypotheses 1 (positive institutional holdings coefficient) and 5 (positive price coefficient). as quoted depth for Boeing stock went from an average of 17.23 round lots in the second half (and the average was 57.471. 16. the trading related to lagged institutional buying previously described was mostly uninformed. the main results of the paper. the quarterly informativeness of trades for Boeing’s stock averaged 26. there was a negative association between trade informativeness. the average number of daily trades increased from 2. Since Boeing’s informativeness of trades dropped from 4. Only the significant lags are reported in Table III.66% in 2002 Q3-2004 Q4 (P3) after decimal pricing.366.) a rise in uninformed trading. the coefficient for price range is positive and significant. after SOA. and the Sarbanes-Oxley Act (SOA). where various lags of change in institutional holdings are included as explanatory variables. this level went from 61. Concurrently. 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. thus reducing the informed investor’s profit potential.17% in the second half. Since institutional holdings reflect past institutional buying. Therefore. The Case of Boeing The following discussion is only meant to illustrate. Both coefficients for price and institutional holdings are negative and significant. an increase in the frequency of small trades (which they interpret to mean a greater participation by retail customers).59% in the second half. informed trading became relatively less attractive as the daily price range narrowed from 3.43% (of the minimum price of the day) to 2. As Boeing’s trade informativeness went from 21.65% in the first half of P3 to 63. For Boeing. and 3. informed trading of Boeing shares became relatively less attractive.

0627 (-1.7942 (-0.2434) 0. 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. z-statistics are in parentheses.75) 0.0083 (0. *Significant at the 0.6806) Trade Informativeness Rank (Period 2) -0.0173 (-0.3182*** (2.73) 0.6821 (-1.8589) -0.25) -0.70) -0. After these events. The data definitions in Table II apply here.2122) -0.0846 (1. **Significant at the 0.35* (-1. Trade Informativeness Rank (Period 1) Lag 1 of the Change in Institutional Holdings Lag 2 of the Change in Institutional Holdings Ln(trade size) -0.3449*** (2.0820 (-1. and informed trading.03) 0.9743 (-1.1681* (1.6939) 0.05 level.0128 (0.3802) 0. Similar evidence is found for the period following the enactment of the Sarbanes-Oxley Act (SOA).10 level. the evidence is consistent with both a rise in uninformed trading and a fall in informed trading.0555 (0.20) 0.3551) -0.79) 0.1525 (0.1129 (0.0386* (1. 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.4435 (0.0118 (-0.2713*** (-3. there is evidence pointing to a relation between uninformed trading and institutional buying in the previous quarter.04) 0. .81) 0. III.4622) 0.91*** (-2. uninformed trading.60 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Table III.09) Ln(trade frequency) Ln(price) Range (%) NYSE Executions (%) Ln(depth) ***Significant at the 0. a phenomenon that does not seem to be present before SOA. The decrease in informed trading may be a continuation of the decimalization effect.13) -0. The results indicate a positive relation between institutional buying and trade informativeness before Regulation Fair Disclosure and decimal pricing.81) -0.72) Trade Informativeness Rank (Period 3) -1.1793** (-2.07) -0. However.1431) 0.01 level.0236 (-1.59) -1.0583*** (3.

2000.K. and M.C. and Z. Jain. J. C. Chakravarty. J.” Contemporary Accounting Research 23 (No. 2007. Nofsinger. “The Trading of Institutional Investors: Theory and Evidence. “Insider Trading Before New Issue Announcements..M. Titman. Jorgensen. 1). and A. N. and M.” Journal of Finance 46 (No. 2004.” Journal of Financial and Quantitative Analysis 35 (No. 3).Jiang. 869-898.. and R. F.. Sias.” Review of Financial Studies 4 (No. and R. 753-766.. 1999. C. and Z. R D. Lee. and D.3). 2008. C. R.” Journal of Applied Finance 13 (No. 2). Y. 5. Karpoff. “The Information Content of Orders on the Saudi Stock Market. 2006. 1991. “Inferring Trade Direction from Intradaily Data. J. 2).W.S. A. 2).” Journal of Applied Finance 14 (No.H. “Changes in Institutional Ownership and Stock Returns: Assessment and Methodology.N. and S. “Liquidity Provision and the Organizational Form of NYSE Specialist Firms. “Continuous Auctions and Insider Trading. Wingender.. Amihud. 1).” Financial Review 43 (No. 2111-2133.2). 2003. “Dividend Changes and Future Profitability. S.” Journal of Financial and Quantitative Analysis 41(No. Priestley. Van Ness. “Bidder Incentives for Informed Trading Before Hostile Tender Offer Announcements. 3). 841-869. D.” Journal of Finance 54 (No. 1985. Lee.” Journal of Financial Research 23 (No. “The Declining Information Content of Dividend Announcements and the Effects of Institutional Holdings. “The Effect of Decimalization on Trade Size and Adverse Selection Costs. 2003. 747-777. and K. 1315-1335.” Journal of Applied Finance 12 (No. 1063-1081. 1996.R.. 77-84.6). 4). “Dividend Behavior and Dividend Signaling. Hasbrouck.” Journal of Business Finance and Accounting 33 (No.K. Ziv. 2006. “Trade Execution Costs and Market Quality after Decimalization. “Applied Market Microstructure. 361-382..” Journal of Finance 59 (No. 6). Shranz. D. Li. M. Zhao.4). 2004. and R. Kyle. . Van Ness. Kim. J. L.” Financial Management 20 (No. C. M.” Journal of Finance 57 (No..” Journal of Finance 56 (No. 2005. 5. 637-660. “The Summary Informativeness of Stock Trades: an Econometric Analysis. Brooks. D.T. M. 2006. “A Survey on the Dissemination of Earnings Information by Large Firms. Wahal. Golbe. 733-746. Ready.K.F. Jr. 7-14. B.K. 6). 1991. 2000. Bessembinder. D.J. Wood. 3).RAKOTOMAVO — THE LONG-TERM VARIATION OF TRADE INFORMATIVENESS 61 References Al-Suhaibani. P. 2006. and L. 57-68. 145-156. 6). A.W. Coughenour. 2869-2910. P.. Rezaee.” Journal of Financial and Quantitative Analysis 38(No .R. Whitcomb. Chung. “Is There Less Informed Trading After Regulation Fair Disclosure?” Journal of Corporate Finance 13 (No.” Journal of Business 79 (No. Collver. 2002. and D. L. 6). 359-378. 6). 2263-2295. 1991.” Journal of Business Finance and Accounting 23 (No. “Changes in Asymmetric Information at Earnings and Dividend Announcements. X. Garrett. “The Sarbanes-Oxley Act of 2002 and Market Liquidity. 1). 18-26. Starks. Chiyachantana.” Econometrica 53 (No. and K. S.C..” Financial Management 23 (No. and J.A. P. J. K. Conrad. X. Jain. 629-654. R.. 2002. and R. “Herding and Feedback Trading by Institutional and Individual Investors. Rezaee. J. 2001. Nissim. 2). Jain.” Journal of Business Finance and Accounting 32 (No. 1994. 2). 3). Johnson and S. “Decimal Pricing and InformationBased Trading: Tick Size and Informational Efficiency of Asset Price. M. 270-281. “International Evidence on Institutional Trading Behavior and Price Impact. 77-80.. 2). I. Deli. H. 571 605. “The Sarbanes-Oxley Act of 2002 and Capital-Market Behavior: Early Evidence. Kryzanowski. 173-189. Sias.

Recently.” This paper explains why such critiques of shareholder theory are misguided yet understandable. Danielson. incentives to increase a firm’s current stock price can distort operating and investment decisions. 1932. PA. stock options) and informal (e. exhort managers to maximize the firm’s current stock price (Keown. proponents of shareholder theory have helped to create this confusion by exhorting managers to maximize the firm’s current stock price. . and Petty. Myers. When wealth maximization is properly defined as a long-term goal. Lasher 2008. Shaffer is an Associate Professor of Finance at Villanova University in Villanova. 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). Martin. and encouraging short-term managerial thinking. Westerfield. Fuller and Jensen (2002) criticize mangers for focusing undue attention on whether a firm meets analyst earnings forecasts each quarter. the shareholder model has been criticized for encouraging short-term managerial thinking and condoning unethical behavior. PA. Danielson and Heck gratefully acknowledge financial support from the Pedro Arupe Center for Business Ethics at Saint Joseph’s University. Although these critics are misguided. 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. However. Brealey. Sundaram and Inkpen. Smith (2003) notes that critics believe shareholder theory is “.. The criticisms are understandable because many proponents of shareholder theory. Wicks. 2002. 2007. 2008. it is not as narrowly focused as critics believe. 2004a). to avoid stock price declines. Because a firm’s stock price can be manipulated in the shortterm. involves using the prima facie rights claims of one group— shareholders—to excuse violating the rights of others. in a stylized version of the model. Jean L. 1962). Heck. wealth maximization has been criticized by a growing array of opponents for condoning the exploitation of employees. Melicher and Norton. . This notion underlies the formal (e. 1 Although incentive stock options typically vest over several years and can have long maturities. customers. Friedman. 2007). Danielson and Press (2006) argue that these incentives can create agency costs whenever the stock price falls below the option exercise price.g.2 By For example. David R. Danielson is an Associate Professor of Finance at Saint Joseph’s University in Philadelphia. Wicks. Shareholder theory defines the primary duty of a firm’s managers as the maximization of shareholder wealth (Berle and Means. Heck is an Associate Professor of Finance at Saint Joseph’s University in Philadelphia. and Marcus. . and other stakeholders. Freeman. geared toward short-term profit maximization at the expense of the long run. Ross. and David R. and Parmar (2004) assert that shareholder theory “.g. pressure from the investment community and corporate boards) incentives that reward managers if a firm’s stock price continually increases. but also key stakeholders including employees and customers. is a fundamental building block of corporate financial theory. Shaffer Shareholder wealth maximization is accepted by most financial economists as the appropriate objective for financial decision-making.. and Jordan. Jean L.”1 Freeman.62 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Shareholder Theory – How Opponents and Proponents Both Get It Wrong Morris G. PA. 2008. and Parmar (2004) criticize managers for pursuing policies designed to continually increase a firm’s stock price. 2 62 . . The main prescription of shareholder theory—invest in all positive net present value projects—benefits not only shareholders. The theory enjoys widespread support in the academic finance community and Morris G.

proponents of shareholder theory open up the model to criticism. if managers (who will typically know that business conditions have changed before the rest of the market) are incentivized to increase the stock price.g. if the firm is expected to pay a $1 dividend next year.04)). unless the interests of future stakeholders are explicitly considered. when viewed from a long-term perspective.” Clearly. the stakeholder model can lead to the same type of short-term thinking that shareholder theory has been accused of encouraging. Thus. Jensen (2005) and Danielson and Press (2006) argue that efforts to further inflate (or to maintain) the stock price may destroy long-term value. R&D. Thus. the firm should continue to invest in all positive NPV projects (which are now less valuable than the market originally expected). However. maximizing and increasing shareholder wealth are two very different objectives.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. a firm’s maximum possible value can decrease. quality control. evidence in Summers (1986) and Cornell (2001) suggest that such deviations can persist for prolonged periods. V0 = ∑ ∞ n =1 (1 + r )n CFn . Indeed. 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. shareholder wealth maximization is not a short-term goal. and the required return on equity is r. and the dividend is expected to grow at a 4% rate per year (forever). AND SHAFFER — SHAREHOLDER THEORY 63 Although shareholder theory directs managers to maximize shareholder wealth. customers. trying to fool the markets through accounting manipulations. (1) To maximize the value of Equation (1). Should Firms Maximize the Current Stock Price? In the shareholder model. In addition. reducing discretionary spending (e. 2003). If the business conditions facing a firm change unfavorably (through perhaps no fault of management). I. managers should invest in all positive net present value (NPV) projects (Brealey and Myers. the goal of the firm is to maximize the present value of future cash flows. As the business conditions facing a firm change.DANIELSON. and they did not pay off for Enron’s shareholders.6% of the value to be realized in years 11 through infinity. If the cash flow a firm is expected to pay shareholders (in the form of dividends or stock repurchases) in year n is CFn. In this example. the intrinsic (per share) value of the firm’s equity today (V0) is defined by Equation (1). At Enron’s peak market value of $70 billion. advertising. focusing on the current stock price.. the shareholder model. Jensen estimates the company was only worth $30 billion. However. A large portion of shareholder wealth is often tied to cash flows to be received in the distant future. the stock price today is $25 if the required return is 8% (= $1/(0. Thus. incentive stock options will provide a positive payoff to managers only if the firm’s stock price increases from the grant date level. For example. 2005). For example. Clearly. 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.08 – 0. maintenance. This is not an unusual or unlikely occurrence. However.86/25). accounting manipulation. etc. stakeholder theory is not superior to shareholder theory from an ethical perspective. . He notes that Enron’s managers tried to justify the excess valuation of $40 billion by “. .4% of the stock price (= $7. and the stock price will eventually decrease to the new intrinsic value. and other stakeholders when making business decisions (Freeman. or adopting fraudulent business practices. P 0 will exceed V 0 and the stock will be (temporarily) overvalued. These actions could include delaying new investments (even if the NPV is positive). 3 4 . these efforts were not designed with the long-term interests of the firm in mind. If business conditions change unfavorably. the case of Enron does not provide evidence against shareholder theory. which can be manipulated in the short-term by unscrupulous managers. Opponents of shareholder theory often recommend that firms balance the interests of shareholders against those of employees. leaving 68. managers face formal and informal incentives to increase the firm’s current stock price. This is the stylized form of the shareholder model. HECK. a firm’s stock price can diverge from its intrinsic value because information is not instantaneously and continuously communicated to the market. provides a better framework than stakeholder theory in which to protect the interests of both current and future stakeholders. dividends during the next 10 years only account for 31.). 1984). 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). Jensen (2005) uses Enron to illustrate the agency costs of overvalued equity. But this experience does Although deviations between P0 and V0 can arise in the short-term even in efficient markets. hiding debt through off-balance sheet partnerships. and over hyped new ventures such as their broadband futures effort.4 To implement the shareholder model correctly. some managers face pressure from corporate boards and the investment community to continually increase firm value (Jensen.

10%. costs. II.” (Kaler. the only yield. but it did not surplus to employees (by paying wages in excess of the create an economic surplus. and a basically capitalist return toward the required structure . it would not be ethical reducing prices). 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%. 1995. the auto industry.10)). $150 in the good outcome. the expected payoff at t = 1 was $110. If so. Similarly. of course. 2004. project had a net present value of 0 (= 110/1. hurting shareholders. stakeholders. this Along these lines. 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. future customers would not benefit from potential life-saving products that might otherwise have been developed. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 example.If the entrepreneur had known at t = 0 that the project would run.10 – 100). 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. and the employees. and other stakeholders. for expense of future ones. ($160 – $110) between shareholders and other stakeholders. . Jorg. proposals to split the realized economic surplus among half of the twentieth century. argue that the shareholders’ Loderer.entrepreneur would not have made the $100 investment. t = 1. 2006). be managed as if it can continue to serve the interests of stakeholders through time. 5 This statement is reprinted in the appendix to Marcoux (2000). . However. Thus. Freeman (1994) recommends that a corporation “. future stakeholders) in the long run. and society in the long. Does Stakeholder Theory Promote a The following example illustrates the potential problem. to distribute the benefits of their activities as equitably as Most investments offer risky outcomes. US employees in the steel industry.” Similarly.”5 To do this. an equitable goals of stakeholder theory distribution of the economic short-term. to customers (by reducing prices). society in the long-run. this policy would reduce both the funds available to invest in research and development and the incentive for firms to do so. reducing job reducing future investment. The entrepreneur invested $100 in the firm at t = 0. these stakeholders will benefit in the short. the current customers of pharmaceutical companies would Stakeholder theory. justice within the confines of reducing the investment shareholders. Assume that at time t = 0. term from lucrative union contracts negotiated in the latter Thus. However. employees’ marginal productivity) or firms (Donaldson and stakeholder advocates might Preston. 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. But these contracts ultimately various stakeholder constituencies have the potential for contributed to financial difficulties at the firms. these policies could stifle future innovation. From these stakeholders will benefit in the and Kaler. in light of their respective entrepreneur did not know with certainty that the project’s t contributions. employees’ marginal productivity) or to customers (by Once the future outcome is revealed. stakeholder theory return is 10%. and all drugs be managed in the interests of current stakeholders at the were then sold at a price equal to production costs plus. . return of 10%. the investment economic surplus (i. On If a firm is forced to allocate a portion of its economic an ex-ante basis the project was acceptable.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. these policies is to promote “an surplus might be to increase could stifle future innovation. 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. . . stakeholders. However. and Roth. 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. 2006). Instead. hurting enhancement of distributive wages or decrease prices. customers. Long-Term Focus? Assume that a firm operates in a simple one-period world. and risks. . does not advocate that firms benefit greatly if patent laws were revoked.. it is likely that the possible among stakeholders. harming society (and potential security and compensation for today’s employees. and the airline industry benefited in the short.e. for example. But. One of the their perspective. profits are excessive.to change the rules of the game and split the excess return term. For example.

proponents of shareholder theory point out that policies adopted by Enron. however. AND SHAFFER — SHAREHOLDER THEORY 65 Parmar (2004). Wicks. Worldcom. providing benefits to stakeholders both now and in the future. And. shareholder theory faces increased scrutiny and criticism. In addition. DesJardins and McCall (2005) argue that a corporation should be managed as a social institution. and Wood (1997). IV. many leading finance texts equate shareholder theory with the maximization of a firm’s current stock price. Worldcom. Jensen. the interests of future stakeholders can only be satisfied if the firm remains financially strong. and executive compensation (e. “It is hard to imagine how anyone can look at the recent wave of business scandals. shareholder wealth maximization (when defined properly as a function of all future cash flows) is inherently a long-term goal. Thus. Agle. HECK. In particular. because a firm must continue creating value for employees and customers to generate future cash flows. the question of how a manager might balance the interests of current and future stakeholders has received very little attention in the stakeholder literature. the goal of financial managers should be to invest in all positive net present value projects. Indeed. and argue that this philosophy is a good idea. Because an investor must find a person who believes the firm will produce sufficient cash flows to justify the prevailing market price. The identity of shareholders will also change over time. The Shareholder Model and Long-Term Stakeholder Interests One drawback of stakeholder theory is that the identity of the individual stakeholders is constantly changing. Despite its current popularity. 2000. all of which are oriented toward ever increasing shareholder value at the expense of other stakeholders. Before dismissing critics of shareholder theory outright. Indeed. 2002). but there is a key difference. 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. who argue that managers should consider the urgency of various stakeholder claims when making decisions. 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. One notable exception is Mitchell. III.DANIELSON. Thus. and . stakeholder theory provides little guidance about how to balance the often competing interests of various stakeholder groups (Marcoux.g. Conclusion In the aftermath of financial scandals at Enron. the customer or employee who extracts excess benefits from a firm during the current period is not the same person who loses future benefits.” However. regardless of whether these decisions will cause an immediate increase in the firm’s stock price. it should not be surprising that some critics of the shareholder theory might (incorrectly) view it as being a short-term 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. 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. However. proponents of shareholder theory must recognize that it matters how the theory is defined and implemented. and Global Crossing. To focus managerial attention on this goal. and thus are not evidence against shareholder theory (Sundaram and Inkpen. incentive stock options) frequently rewards managers for increasing the stock price. However. and Global Crossing clearly did not benefit the firms’ shareholders in the long-run.. We disagree. 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. it is important to recognize that supporters of shareholder theory often emphasize the model’s short-term implications when defining the theory. 2004b). As stated by Freeman. the maximization of a firm’s long-term cash flow stream should not harm the firm’s stakeholders. with those who would use the deficiencies of the stylized model as a reason to abandon shareholder theory in favor of stakeholder theory.

and J. Smith. Practical Financial Management. New York. “Stakeholder Theory and ‘The Corporate Objective Revisited’: A Reply. DesJardins. University of Chicago Press. New York.4). Friedman. 4 th edition. 2004. “Stakeholder Theory and ‘The Corporate Objective Revisited’..J.R.66 References Berle. A. Fundamentals of Corporate Finance. J. 1999. E. Brealey.. R. Inkpen. and B. “Just Say No to Wall Street: Putting A Stop to the Earnings Game.J.. 357-378. Fuller..E. 235-256. R. Hoboken.K. Parmar.. A. 1-16... 2000. Macmillan. “The Corporate Objective Revisited.M. Capitalism and Freedom.1). S. 1994.. 853-886. and A.J. 85– 90. Wicks.R. 2005. P. “Shareholder Value Maximization: What Managers Say and What they Do. Melicher. 1997.2).” Advances in Financial Education 4 (No. 1986.J. 247-328. 2008. “Does the Stock Market Rationally Reflect Fundamental Values?” Journal of Finance 41 (No. John Wiley and Sons. 2006. Roth.” Organization Science 15 (No. Keown. and D. C. S.J. “Is the Response of Analysts to Information Consistent with Fundamental Valuation? The Case of Intel. Myers. and B.” Financial Management 34 (No.J. 2005. 1932. Strategic Management: A Stakeholder Approach. Stakeholders Debate. and L. Wadworth. and L. Freeman.1). W. 2002. 2004a. Myers. Danielson. Pitman..” Journal of Business Ethics 69 (No. and A. A. . McCall.C. “Balancing Act.C.W.2). Inkpen. Donaldson. Loderer. and M. DesJardins and J. Freeman. “Evaluating Stakeholder Theory. Summers. Petty.” Academy of Management Review 22 (No.4). Belmont. Stout.A. 1984.. Preston. New York.. R. Jordan. New York. A. Marcoux.. Stakeholder Theory. Press.R..3). Martin. and J. Norton. Means. 364–369. R.A.J. Boston.1). B.” MIT Sloan Management Review 44 (No. edited by J. Jensen. R.R. 591-601. 113-136. Introduction to Finance. CA. C. 5th edition.. and E.” Academy of Management Review 20 (No. Jensen. Agle. M.” Journal of Applied Corporate Finance 14 (No. M. Jorg.E.M. Chicago. 409–421.” Organization Science 15 (No.” DBW Die Betriebswirtschaft 64 (No. J.E. and Implications..A. A. T. R. 2004b.C. 65–91. Principles of Corporate Finance. Sundaram. 249–268. R. A.4).” Financial Management 30 (No.D. 1962.. M. and G.W. 41-46. DesJardins and J. Thomson South-Western. Cornell. J. 350–363. Fundamentals of Corporate Finance. R. “A Team Production Theory of Corporate Law. 3). Lasher. 1995. 2008. Edited by J. 2006. 4).. Ross. Upper Saddle River. McGraw-Hill/Irwin.” In Contemporary Issues in Business Ethics. Belmont. Pearson Prentice Hall.. Blair.C. 370–371. Sundaram. and S. Brealey. CA. Mitchell.” Business Ethics Quarterly 12 (No.3). 2007.” In Contemporary Issues in Business Ethics.3).A.2).C. Wadworth. Wood. 2003. 2004.R. Mason. L.H.D.” Organization Science 15 (No. M. Westerfield. H. McGraw-Hill/Irwin. “Toward a Theory of Stakeholder Identification and Influence: Defining the Principle of Who and What Really Counts.W. and the Corporate Objective Function.4). 2003. and E. B. 2001. McCall.A. “Value Maximization. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Jensen.A. and A.. 3). McGraw-Hill/Irwin.” Business Ethics Quarterly 4 (No. M. The Modern Corporation and Private Property. Foundations of Finance. “Agency Costs of Overvalued Equity. McCall. 2007. “The Shareholders vs. “The Stakeholder Theory of the Corporation: Concepts. Evidence. 5-19.” Virginia Law Review 85 (No. J. “Do Stock Options Always Align Manager and Shareholders’ Interests? An Alternative Perspective. Kaler. Marcus. “The Corporation as a Social Institution. 2008. Freeman. and L. 2002. “The Politics of Stakeholder Theory: Some Future Directions..

it is well recognized that play money often leads to excessive risk taking as students try to outperform the market without realistic penalties. or investing in the stock market with play money. managing a bank. 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. employers. These funds allow students to invest real money in the stock and bond markets. he does not specifically address the effectiveness of many of the initiatives mentioned above. 67 .St. Anthony Lerro.2 In the field of finance. stock selection. allowing students to simulate starting a new company.Student Managed Investment Funds: An International Perspective Edward C. Most of these programs supplement the more traditional investment courses. This paper should be of interest to faculty. Deans also started encouraging their faculty to invite more guest speakers from government and industry to address classes on issues of the day. interactive software became more prevalent. 1 2 Even now. Nevertheless. and all of the university faculty participants who generously gave their time to complete this survey. students. Karen Wagster. Louis. and practitioners in the financial community who desire basic knowledge about state-of the-art teaching investments and portfolio management. In the early 1970s. which are offered by every institution with a business college. student-managed investment funds have grown in both size and complexity as universities have tried to mirror real world experiences. Finally. The career success of students coming out of these programs demonstrates the benefits of providing students with as much hands-on experience as possible. To Edward C. Larry Belcher. 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. portfolio composition and market timing.1 Although all of these approaches were a significant improvement over what educational institutions had done historically. The vast majority of SMIFs have close faculty involvement to provide oversight and structure to student activities. there was still a need to offer students even greater realism and more practical experience. Employers were often critical of new graduates who had difficulty stepping into employment without first receiving extensive on-the-job training. with the development of computers. However. 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. MO. Over the last two decades. Lawrence is a Professor of Finance and Department Chair at the University of Missouri . However. Brian Bruce. overcome this obstacle. 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. Louis in St. students are directly managing more than $407 million in assets in 2007. 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. many business colleges began partnerships with major companies to offer students co-op and internship programs while the students were still pursuing their degrees. In aggregate. The author would like to acknowledge the assistance of Kerry Sallee. student managed investment funds (SMIFs) were created to take investment education to the next level. Ken Locke.

In aggregate.4 For US participants. a written survey was electronically sent to all universities in the US and abroad with both known and possible SMIFs. This led to an explosion of programs. The Survey From June 2007 to April 2008. It was assumed that faculty involved with current programs would most likely know of other SMIFs from their professional contacts at conferences. it was common for paper authors to only describe a single fund. 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. Since locating foreign programs would be more challenging. II. In the case of foreign countries. Europe. Hirt (1977). a significant number of contacts were made with finance department chairs and deans at major I. While SMIFs were first started in the US and have been around since 1950. The purpose of this paper is to discuss the evolution of SMIFs and the impact they are having on teaching investments around the globe. Kester (1986). survey participants were also asked to share their knowledge of other programs in their states. Bhattacharya and McClung (1994) and Kahl (1997). those programs were not widely known outside of those campuses. Bear and Boyd (1984). Block and French (1991). Recognizing this problem. Australia and Asia.000 to $62 million. Some funds have outside professionals that serve in an advisory capacity to enrich the experience for students and showcase their programs. sharing technical resources. 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. Furthermore. Not only do new and existing SMIFs learn from the innovations of other successful programs. there was not even a list of universities in the US that had such funds. students are directly managing $407 million in assets. Lawrence (1990) conducted the first survey to profile and discuss the characteristics of almost two dozen established programs. 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. it became an “easy sell” for finance faculty and alumni in North America. it was Some of the earliest case studies included Belt (1975). it was very difficult for finance faculty to start new funds given the lack of operational data and instructional inexperience with such programs. All of these papers served as a major catalyst for the rapid growth in the number and size of SMIFs worldwide. there are 314 funds worldwide ranging in size from $2. professionals should become educational partners by providing guest speakers. As a result. Besides the opportunity to hire highly trained students coming out of these programs. serving on boards. only 12 colleges had them by 1969. Lawrence expanded his study to include 34 programs in order to better describe their operations and funding sources. 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. 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.68 managing the portfolio. Ammermann and Runyon (2003) investigated risk aversion and group dynamics among students making portfolio decisions at California State University in Long Beach. Today. 4 . Johnson. faculty and administrators benefit from considering a broader array of approaches to solving specific constraints faced by a particular school. 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. Markese (1984). Finally. Finally. Tatar (1987). which have spread to other continents including Asia and Europe. providing funding. While the list provided a good starting point for the US. 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. especially in North America. By the early 1990s. Until then. 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. with so many leading business schools embracing the basic concept. Unfortunately. Alexander and Allen (1996) investigated alternative decision making environments in student managed funds. In 1994.3 While these efforts were insightful. Neely and Cooley (2004) reported 134 funds had been established in the US alone. etc. Until 1990. 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. By 2003. participants were solicited for information on existing and potential funds in their own country or neighboring countries. employers and practitioners also need to become more knowledgeable of the various types of SMIFs. Surprisingly. In fact. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 impossible to fully understand the scope of this movement in investment education without a broader database. there has not been a single paper presenting data on funds from outside the United States.

it could easily mentioned screens) in Europe and Asia to make sure there take another 20 years to catch up to North America. West Allis Central (WI). 8 . with the exception of Canada. the country may be a long way from reaching a saturation point. To remain competitive in the marketplace.Madison which has contacted up to 7 times by $62 million being invested in email and/or phone to some form by students. the 1990s was undergraduate and graduate business students. it of Minnesota. These scaled down programs closely mirror those established at the university level in the way they are structured and operate. Gaithersburg High (MD).LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 69 business schools (not included in any of the previously than in the US. 7 The largest education to the next level.000 or $200. funds started with only $100. The other four portfolios are fixed income portfolios. Groton High (MA). 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. While 99% of all current SMIFs are housed III. Appendices A and B provide a complete list of all management by students.6 There In the field of finance. For After 40 years of very slow growth.4 million and $1. Canada actually has a higher concentration of funds within institutions of higher learning The University of Wisconsin . almost all of the SMIFs continued to participation rate. these funds give a successful program. 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. Ariel Community Academy (IL). “One cannot have a top 10 MBA program today without it. Of the 314 funds. 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. With faculty $10 million including 2 funds. 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. with $25 million was sometimes very difficult each in assets. Of course.2 million for nonUS programs. managed investment funds (SMIFs) advisors frequently rotating Ohio State and the University were created to take investment in and out of the programs. Besides the traditional university teaching investments. but as an extracurricular activity. respectively. As reported in Table II. the largest US and non-US programs. With 1. Yet. In addition.5 The current and business. As reported in Table I. However. Only one portfolio is invested in equity securities. 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. One of the more interesting growth patterns for SMIFs is how widely they are being used in a broad range of educational environments.” 5 It has been reported that the following American schools at one time had active funds: Dominican High (WI). For the remaining 90 SMIFs who declined receive additional investment capital from various sources to participate. As one faculty member stated. there are a few exceptions. more like investment clubs portfolio is necessary to have with little or no faculty involvement. The Growth and Size of Programs within business schools. Tufts University. 6 The average size for US funds was $1. and Burnsville High (MN). Wisconsin Lutheran (WI). there are the turning point. few money in the stock and bond survey.000 in initial seed 224 programs returned the completed survey for a 71% capital. site and media sources. which provide a different set of learning experiences depending on the objectives of the fund. 7 A university having a student investment fund has become the gold standard for investment programs at all levels.680 business schools in the US according to the AACSB. was as much international exposure as possible. Each program was fund is at the University of Wisconsin .Madison actually has five distinct funds. student encourage their participation are 8 SMIFs with more than in the survey. a small faculty members would argue number of the funds operate that a multi-million dollar markets. without a business college. In other parts of the world. the number of SMIFs example. 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. given the The size of SMIFs today has expanded beyond what many obvious language barriers. The first non-US fund was established at the University of British Columbia in 1987. SMIFs are just in the early stages of development in the rest of the world. Jenks High (OK). The largest portion of the fund is private money managed for clients based on set investment criteria. 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.

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.US Universities Country US US US US US US US US US US Total Assets $62.3 Million . Of New Brunswick Concordia University University of Alberta Country Canada Canada Canada Canada Canada Canada Canada Total Assets $10. of British Columbia Queens University Univ.5 Million Rank 1 2 3 4 5 6 7 Institution Simon Fraser University HEC Montreal Univ.2 Million $6.5 Million $12.0 Million $3.70 Table I.0 Million $18.0 Million $25. The Largest Funds (in US Dollars) Panel A.2 Million $17.2 Million $1.5 Million $6. Non.0 Million $13.0 Million $2.0 Million $9.5 Million $3.8 Million $3. 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.8 Million $25.4 Million $1.

Winthrop University. and the University of Louisiana. Depending on the quality of the student research and group presentation skills. this was unexpected. In addition. a former US Treasury Secretary. which are solely focused on generating the highest risk-adjusted rates of return possible. With many schools emphasizing research productivity for promotion and raises. partially due to the greater amount of time it takes to stay abreast of the financial markets on a daily basis. 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. the average program in the US during the 2000s was started with approximately $414. the Tennessee Valley Authority reported that over a three year period. the limited anecdotal evidence suggests students generally do as well and sometimes better than investment professionals or the market as a whole. 10 Faculty advisors generally recognize that some of the best learning experiences come from failures. The balance of programs was a combination of capital sources.LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 71 unlimited number of reasons that could not have been accurately forecasted a year or more in advance. Southern Illinois University . unlike practitioners in real life. 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). As any experienced investor knows. Professionals can also trade on margin or use derivatives to enhance returns which are not widely available techniques for the majority of SMIFs. faculty salary) from the universities and rarely pay all expenses related to the fund. The most common form of organizational structure is having the SMIF be part of the university endowment. Offsetting some of these advantages. there is always an element of luck and incomplete data behind any decision. a very carefully analyzed opportunity with great potential can fail for an almost V. 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. the University of Missouri . SMIF returns are secondary in nature to the educational mission. However. Twenty-eight percent of the funds got all of the money from the university’s own endowment. students lack a strong incentive system of monetary rewards for beating benchmarks or penalties for poor performance (being fired). computers. Other finance faculty were unwilling to take over the fund. not successes per se. Another 14% are set These universities include the University of Central Florida. whereas most SMIFs get subsidized resources (e.St.Kansas City. IV. it is most onerous for tenure track faculty to lead the fund activities for more than a few years. it takes several economic cycles to really evaluate the success of any investment strategy. Thus. Of course. Funding Sources and Organizational Structure The majority of older SMIFs received ear-marked money from alumni and other private donors to establish the funds. About 62% of all funds are structured this way. most SMIFs are structured to rely on group or committee decisions rather than those of a single portfolio manager.9 Given how difficult it is for many schools to establish SMIFs and their popularity with students and employers. O’Neill has been aggressively promoting greater financial skills at the primary and secondary education level. Student portfolios often have constraints that most professionals do not have. the most common reason the program became inactive was the loss of the key faculty member (often due to retirement) who advised the students. For universities wishing to establish a new fund.g. The fund at the University of Missouri . the 19 universities participating in its program in 2002 outperformed the S&P 500 benchmark by 5.3% (Mansfield. facilities. provided they are not trading on unfounded rumors and there really are fundamental market changes taking place. 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. some of the stimulus is being provided by Paul O’Neill. that used to be taught only at the college level are now being introduced in high schools. This factor alone should favor practitioner performance. the University of Florida.000 in initial capital. Obviously. decisions are not always based entirely on objective analysis. In personal discussions with faculty and administrators at these schools. 2002). including many small donors and corporate donations. For example. professionally managed funds must absorb all of their own operating expenses. Unlike professionally managed funds. Sometimes there are more than 30 students involved with various levels of skill and knowledge. Professional managers can react almost instantaneously to rapidly changing market conditions without the need to assemble the group for a vote. For example. 9 10 Although there has been no systematic data collected on SMIF performance. Another 23% of schools had only a single large donor.Edwardsville. According to McInerny (2003). 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. All have an equal vote in the ultimate decision. many schools have dealt with this issue by hiring nontenure track faculty or adjuncts to run the SMIFs. .

000 students participate in SMIFs in the US each year. Approximately 71% of the programs in the US (45% of foreign programs) are structured as part of a formal class.g. most SMIFs carefully control the level of student participation.91% 15.12 The number of credit hours a student can earn ranges from 1 to 12. like a nonprofit foundation or trust to provide more autonomy from the university. The SMIFs that are not part of a formal class allow students to participate as an extra curricular activity. partnership tax returns with K-1 forms) due to the taxable nature of the investments. 44% allow students to earn 6 or more semester credit hours over 2 or more semesters. For approximately 90% of SMIFs. University of Houston. can join the group. Actual 5 Year Historical Annual Returns (Including Dividends) Year 2003 2004 2005 2006 2007 Actual fund performance 36. students performed all of the usual research on securities but they had to make formal presentations to a professional board. students were responsible for making all investment decisions. which actually made the final investment selections.25% S&P 500 28. LLCs or partnerships) where students are managing the portfolio for private companies or other investors. Several innovative companies in the US have long provided money to support financial education at institutions in the markets they serve.88% 4.68% 10.72 Table III. Thus. In many cases. with another 500 students being trained at foreign universities.g.11 It should be noted that this is a more complex structure in the US. the average fund in the US had only 29 student managers per year (23 students for foreign funds).93% 12. At the London Business School. and University of Texas are all managing some private investor money.76% 6. formal classes often restrict the quality of students usually through an application process (59% of schools have a formal application process to screen students). VI. 12 . While Panel B. only a few informal clubs not sponsored by the school exist. It is also becoming more popular for programs to establish profit making companies (e. In case of a falling stock market. This allows students to interact with professionals and showcases the program to the local community. Of those providing credit. universities have legal barriers for incorporating SMIFs into the curriculum. Profile of the University of Missouri . advisory boards or a faculty member also shared in the decision making. In France. This less structured format permits greater inclusion since almost any student enrolled in the university. which sponsors 25 universities in its service area. the company absorbs all losses and fully replenishes the money the following year. Two brokerage firms followed the TVA’s lead with D. In contrast. At least ten of the largest funds.80% 5.67% 5. Student Participation Just over 5. fixed income and options Bottom-up approach Yes Scholarships JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Company sponsoring 20 schools and Stern Agee Group. which are advised by outside professionals. In the other 10% of programs. All of these boards have outside investment professionals and alumni serving as a valuable resource in a counseling capacity. Unlike many other university programs. Only 22% of programs limit the student learning experience to a single semester. which requires government reporting (e. a large electric utility company. students make formal presentations to the boards to sharpen their presentation and analytical skills.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.000 and $50. Inc. 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. Pennsylvania State University. The basic model at these programs is for the company to provide all funding ($400. supporting 5 universities.000 per year. University of Minnesota. Although a few schools allowed more than 100 students to manage the portfolios each year. The largest is the Tennessee Valley Authority (TVA).St. A. The faculty advisor also reports the additional burden from accounting and tax costs for the LLC run between $25. About 58% of universities have an advisory board associated with their programs. Penn State had 68 private investors and was planning to expand the number to 99. Davidson & 11 At the time of this survey.49% up as a separate entity. including the University of Wisconsin.000 each for the TVA) with the company and universities sharing the profits.78% 17. regardless of major field or prior course work. Another 39% of schools limit students to a maximum of 3 credit hours. Louis Fund Panel A.

research grants. 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. the higher programs have become inactive because key faculty members authority vetoed less than 4% of all student decisions. 64% courses take significantly more time. and be guest speakers. VIII. could be in the form of added salary.5 credit hour course rather than a 3 hour course. In the previous reduced service load. profile programs. The most of students. Vanderbilt University and the Georgia Institute of Technology. The an advisory board to veto student recommendations if an balance of the universities provided additional compensation investment is deemed inappropriate for the portfolio. Given the nature of SMIFs. Faculty and Professional Involvement growing number of schools take this approach. it is a formal class. Dartmouth University. or a However. Princeton University. Even when the program Of the universities requiring students to participate through is being taught by a full-time faculty member. In some cases. 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. Frequent contact between the various parties teaching a 4. For 70% of the on short-term economic SMIFs. This feel the compensation levels are not commensurate with the result indicates that students time involved. when the financial markets are open. A small but VII. in the form of research funding. the students invest their own money. Harvard University. Investment Activity A number of universities have real money funds that operate more as an extracurricular activity with no direct faculty involvement. Auburn University. many of the programs have 10% permit only graduate students and 48% allow both levels local investment professionals closely 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. allowed to erode based solely management by students. classes. This may partially explain why some five years for those programs with a veto power. This time with the remaining funds using a combination of students. 13 It is interesting that 28% of schools with a SMIF have more than one fund. faculty and 50% higher than a traditional class. One possible solution are determined by a simple majority vote by the students. a course faculty advisors and/or boards to reach a consensus. 58% of these professors believe these classes commonplace to have professionals serve on advisory boards take substantially more faculty time than a regular course. or the equivalent to professionals. There manage the portfolios. involved with SMIFs at all levels. investment decisions savings. It would be Depending on the specific a shame if much of the progress are 78 universities worldwide with portfolio and its goals.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. benefit ratio of SMIFs and expanded beyond what many To facilitate investment make a conscious effort to decisions. etc. Schools that operate this way include the University of Edinburgh. etc. Only 9% of faculty thought it knowledge and skills. drawing on the general finance community to provide the course instructor. 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. Of the release. the adjunct faculty member is retired With a small number of exceptions. actually took less time than their normal instructional duties. 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. direct role for outside professionals is to serve as an adjunct faculty member and run the actual program. Many of these funds have different investment objectives and are specifically designed to give students a . supplemental pay. faculty are closely and thus has time available to staff day sections of the class. In most cases. 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. 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 is a power that is rarely used. the California Institute of Technology. 42% of programs allow only undergraduate students. a graduate assistant. industries.13 Because so many of the It works less effectively for active professionals who may programs are relatively new. There is no question that professional The average assessment was that the instructional load was involvement enriches the experience of students.

Thus the Canadians adopted SMIF organizational structures and operating procedures modeled after those already successfully employed in the US for several decades. real estate investment trusts (REITs) were the most popular followed at a distance by limited partnerships. Twice as many programs thought asset allocation was a very important consideration compared to those who deemed it not very important (38% v. averaging only $142. and the remainder in other securities. Canadian and American universities share a similar educational environment. In terms of investment policies. The balance of the SMIFs used a combination of these strategies and others (e. the average portfolio contained about 30 individual securities with three SMIFs exceeding 75 different issues. some schools. Canadian SMIFs averaged 70% of their money invested in common stock. stricter university controls may eventually develop. 81%). the most common number was to have three distinct portfolios. . Hebrew University had the most balance with 25% in common stock. European and Asian schools were far more likely (by a ratio of 2 to 1) to have extra-curricular programs rather than formal classes. IX. 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%). For actual trading activities. The funds are large enough to negotiate some very favorable rates with full service brokerage firms. With maturity and more money at risk. As to investment styles. 65% of funds believed market timing was not important. only 10% of the programs characterized their investment style as focusing on growth stocks. regardless of where the companies were domiciled. Of course. full service brokerage firms were most often used by SMIFs. Unlike the US. 18%). When it comes to investing. Only 10% of non-US SMIFs had this same freedom.74 broader investment experience than a single fund could provide. 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. It also makes sense to have more than one SMIF where there are multiple bodies of students including undergraduate/graduate. etc. only 14 Myers (2004). price momentum and contrarian) in making stock selections. 25% in Treasury bonds. For SMIFs operating a single fund. none of the European or Asian programs had a taxable structure similar to a LLC or partnership. In the first reported case of an actual hedge fund on a university campus.g. most American SMIFs focus on investing in common stock with an average portfolio in 2007 containing 82% of their money invested in these securities. are very restrictive and require that all investments must be in domestic markets. Foreign programs generally leaned more toward balanced portfolios with greater allocations of fixed income securities. It is interesting that 19% of US SMIFs are not prohibited from becoming hedge funds to increase returns by taking on more risk. such as the University of Toledo. 20% in preferred stock. day/evening classes. With the growing size of the average portfolio. Approximately 80% of the programs have clear guidelines that specifically require the funds be diversified. The $3 million fund uses investors’ money (alumni and friends) which allow students to manage the money for the experience without any fees. 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. Most SMIFs focused on traditional securities with common stock dominating portfolios. Comparisons of US and Foreign Funds With the widely divergent political and economic climates abroad. Counting a mutual fund as a single holding.000. Another 23% considered themselves to be JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 more value investors. almost the same percentage of foreign SMIFs required diversification in their written policies as in the US (79% v. On the lower end of the scale. Bank trust companies were used only about half as much as either type of brokerage firm. Roger Williams University limits domestic investments to a maximum of 20% of the portfolio with the other 80% being comprised of international securities. Corporate and Treasury bonds were also widely used to balance the portfolios. Individual security selection was rated very important by 58% of SMIFs. As to types of security investments. For alternative investments. Cornell University changed its SMIF’s investment strategy from a indexed styled fund to a “marketneutral” hedge fund in 2002. In addition. followed at a distance by the buy-and-hold strategy reported by 11% of respondents. For the most part. 15% in corporate bonds. Diversification is a principle widely emphasized by most SMIFs. it is not surprising to find 92% of universities have formal written investment guidelines. Outside of North America.14 The stated goal was to produce positive returns regardless of which direction the market was moving. while European and Asian funds were much lower at 59%. For schools with more than one fund. On the other extreme. A close second was the top-down approach used by 27% of funds. In contrast. most likely reflecting their long-term investment horizons. the funds outside of North America are much younger and smaller. however. 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. followed closely by discount brokerage firms.

field trips. Almost a third of faculty believed having a real money fund provided synergy and significantly improved the quality of the overall finance program.LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 75 hundred thousand dollars in cash flow while providing students with a valuable learning experience. For 2006. since SMIFs do not normally charge management fees. B. X. Conversations with faculty also indicated greater job opportunities for students participating in SMIFs. and other university programs. 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. Some opponents argue that it is pushing a political agenda. Texas Christian University. Establishing socially responsible funds can be highly controversial in academia. In 1997. noting their endowment already had substantial stock investments in socially responsible companies and industries and thus it was not needed. 2002). and mutual funds have been aggressively recruiting students who have these experiences to draw on. alcohol. . the predominant investment style was a “blend” rather than a single focus. Students will hold long positions in firms that are socially responsible and take short positions in firms with poor social records. which creates new opportunities for guest speakers. The director of its program maintains one does not have to sacrifice financial returns for a good record of social responsibility (Alsop. Eighty-one percent of all program directors cited better trained students as a major benefit of having a SMIF. which include tobacco. Regardless of where the fund was domiciled. students also benefit in many programs by going on field trips to Wall Street and other financial markets. student recruitment. University of Michigan. Of course. the limited evidence shows SMIFs have performed as well and sometimes better than funds managed by professional investment advisors (Mansfield. the programs provide badly needed financial support for student scholarships. Finally. As anyone involved in a SMIF can attest. Roger Williams University has taken it to the next level by taking students abroad to the London and Frankfurt stock exchanges. and the University of Missouri . visits to financial markets. Nevertheless. Very few university programs have such a high benefit/cost ratio. However. or a combination of the two. an even larger number of SMIFs reported making no cash distributions in the previous year. or an average of $29. These schools believe this development has raised the bar in attracting top students and community financial support. This activity not only showcases the students and the finance discipline but also the business school and the university. Of course. it was not all that unusual for the larger SMIFs to spin-off several XI. Many of the universities with SMIFs have invested up to $1 million to fully furnish and equip trading rooms. Historically.2 million as part of its MBA program. Michigan State University. The University of California at Berkeley started a new social responsibility fund in February of 2008 with $1. Recent Developments A. Stetson University. having a real money portfolio generates a substantial amount of media attention. internships. sixty-six of the American funds made cash distributions totaling more than $1. Many employers. which would not have been possible without a SMIF to generate the student interest and fund the activity. 2007). The expanded programs include Pennsylvania State University. operating trading rooms. Bluffton University in Ohio has had an investment policy since 1956 of avoiding “sin stocks”. Rice University. and defense companies. Trading Rooms A growing contingent of programs are operating trading rooms to add even more realism to student learning.9 million to support academic programs. Social Responsibility Funds Social responsibility funds are becoming more popular in academia and with investors in general. The equity strategies most used in the US and abroad was a bottom-up approach. Iowa State University. there are other universities with trading rooms that do not have SMIFs and simply simulate trading activities. closely followed by the top-down approach. Alumni in particular are highly supportive of SMIFs. bank trust companies. Stanford University rejected a student proposal for such a fund. These are highly competitive jobs that can be difficult for a new college graduate to obtain without sufficient experience. this saving alone favors student-managed funds even without the educational benefits. etc.381 per school. 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. especially since almost all universities have endowment funds that must be managed by someone. Villanova University follows a similar investment guideline with two of its funds. Many of these were still relatively new and therefore were still in the capital building years. 10% of American funds with a single fund considered themselves “growth” investors in 2007 compared to 20% of the foreign funds.Columbia. including private equity and hedge funds.

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.5% of the assets. including Columbia permits students from more than 15 other universities to University. or the general community at large. Science and Engineering produce a continuing stream of and 1. the venture capital structure limits the number funds offer an excellent vehicle of investors. Several universities. Others. Minnesota. an amendment for establishing a student investment fund to encourage For further information. are starting micro finance funds to make small participate in the activities. Cornell University. Yale University. 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. conventional SMIFs in new dimensions.76 C. which also investing. the Colleges of Medicine. see Rombel (2007). In evolve in new directions. Daily Herald (2006) and the Business Wire (2006). Along with business schools. Doing so may speed technology transfer from universities and fulfill one Several of the largest funds (including the universities of mission of higher education. One of the most exciting financial performance record. Given the success of SMIFs. so they must natural that the programs would for the College of Business to make large contributions. 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.5% university community. 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. The University from the research strengths of universities while enhancing of Michigan created the first student-managed venture capital the teaching mission. to investors similar to a investment opportunities. 15 student entrepreneurship. 16 partnership distribution. All of these innovative programs Texas. Micro Finance Funds of Utah. fund in the US in 1997 with about $3 million in capital. Wisconsin. recent graduates. 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. investor money). it will provide another venue University of Queensland and the University of Melbourne (both in Australia). have similar funds run by professionals. 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. Houston. Venture Capital Funds patents. By 2007. Yale Bulletin (2001). Although not a student-managed entrepreneurial investments in third world countries. This Given the success of SMIFs.15 The University of Utah sponsored accomplish more than simply teaching students the basics of the largest venture capital fund with $18 million. including the Columbia’s program is a success. 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. and Pennsylvania expand the practical training offered to finance students by State) manage investor money in one or more of their funds. An fund. innovative research and technology but have great difficulty proving the commercial viability of their inventions and D. it is only Finally. If assisted by students and faculty. Using Investors’ Money JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 faculty. For example. 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. it is only natural that offers the best opportunity in years to capitalize and profit the programs would evolve in new directions. publicly-traded capital funds managed by students. the students and university sometimes get a management fee of between . the University of Wyoming. provide value added support to exchange for managing the other units within the money. 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. Chapter 23A of the budget) . the University E. A student managed venture capital fund students. the University of North Dakota.

Although the university experience is not as intense as in a professional job. establishing hedge funds. In the process. or venture capital funds and micro lending initiatives. Students graduating today from universities with SMIFs already have at least 1 or 2 semesters of actual trading and research experience. The benefits of providing students with greater practical experience and technical skills in finance are widely recognized in the job market. SMIFs are evolving in exciting new directions. 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. it still provides a solid foundation for the knowledge needed in portfolio management. These programs have expanded to 314 worldwide today from only a few funds in the 1950s.LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 77 business schools are becoming even more relevant by addressing important issues in both the financial markets and society. for students to learn about business while benefitting social welfare initiatives around the world. California State University Fresno California State University .300 175 577 100 6.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.000 400 1. student-managed investment funds have revolutionized the way in which investment education is taught in universities.866 425 100 750 1.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. 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 . Conclusion Over the past 50 years. XII. These include managing money for private clients.000 490 453 90 100 . While it is much too early to evaluate the success of these new programs.500 400 555 130 174 149 360 25 265 13 1.

Listing of All US Funds (Continued) University Name California State University . 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 2.300 24 190 77 13.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.500 200 810 368 200 122 123 na 343 7 59 458 383 740 437 500 223 195 .Northridge California State University .500 55 400 341 239 250 162 370 136 50 130 1.000 100 800 300 64 50 52 75 120 469 400 400 381 300 100 170 590 1.500 2.78 Appendix A. 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.200 300 220 204 126 25 1.

013 2.200 1.000 500 1.375 281 128 25.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 1.000 60 20 280 92 5.LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 79 Appendix A.810 2.200 75 27 21 375 4.442 440 250 5.300 325 275 200 400 385 50 50 1. 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 .000 251 10 400 495 900 500 122 .Bozeman* Montana State University .100 100 190 455 360 450 430 172 1.001 135 110 50 997 230 210 70 2.

500 2.000 2.177 400 390 536 .700 117 916 1. 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. 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.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 .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.Huntsville University of Alabama .000 350 300 58 2. Cloud New York Philadelphia St.80 Appendix A.100 1.100 400 500 250 1.Boulder University of Colorado . Bonaventure St.500 59 349 12 180 18 3.000 388 500 100 350 200 50 81 835 360 6. Worth Lubbock Ft.Tuscaloosa* University of Alaska University of Arizona University of Arkansas-Fayetteville University of California .300 800 550 101 9.Birmingham * University of Alabama .605 45 115 2.Berkeley* University of Chicago * University of Cincinnati * University of Colorado .340 1.000 120 1.200 776 405 1.059 2.419 400 100 428 385 50 550 930 12.Los Angeles University of California .300 6. 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 .

523 400 50 1.000 440 335 1.Wilmington * University of North Carolina .000 1.300 1.350 475 3.Knoxville University of Tennessee . 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. 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 .000 1. Thomas University of Tampa University of Tennessee .McIntire School University of Virginia .000 505 900 700 351 65 151 325 200 520 2. 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.173 500 6.LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 81 Appendix A.100 1.000 510 17.Chapel Hill University of North Carolina .600 308 3.253 1.000 235 676 772 277 1. Information collected from media and institution’s web site.Minneapolis University of Minnesota .400 400 1.700 50 * Did not respond to survey.000 155 460 1.Lincoln University of Nebraska .Platteville* University of Wyoming Utah State University * City Lawrence Lexington Louisville Orono College Park Memphis Ann Arbor Minneapolis Duluth University Columbia St.000 85 135 190 1.700 25.Darden Graduate * University of Washington * University of Wisconsin-Eau Claire University of Wisconsin-Madison University of Wisconsin-Whitewater University of Wisconsin. .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.Martin University of Tennessee. Louis University of Montana * University of Nebraska .100 200 115 5.Omaha* University of Nevada * University of New Hampshire * University of New Mexico University of North Alabama * University of North Carolina .Duluth University of Mississippi University of Missouri-Columbia University of Missouri-St.354 125 50 1.Chattanooga University of Texas University of the Pacific University of Toledo * University of Tulsa University of Utah University of Virginia .200 50 250 62.400 107 50 2.424 1.577 18.Oshkosh University of Wisconsin .

“Students Press for Socially Responsible Endowment Fund. 2006. Runyon. Belt. Ammermann. Listing of All Non-US Funds University Name Birla Institute of Tech. “Talking b-School: Haas Takes New Tack on Investing.” Journal of Financial Education 13 (Fall). B. and R. “An Applied Course in Investment Analysis and Portfolio Management. M. Francis Xavier University St.” Stanford University News Service (February 5). B8. 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.82 Appendix B. Webster. 68-71. 2000.” Daily Herald (April 12). . 2007.” Journal of Financial Education 4 (Fall). “A Securities Portfolio Managed by Graduate Students. R. Ary. Information collected from media and institution’s web site. Bear.” The Wall Street Journal Online (September 14). P. 2003. L. “MBA Students to Run Socially Responsible Fund. 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. Alsop. 1).810 300 70 15 10 3..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. 1. R.” Business Wire (June 19). J. References 1997.” The Wall Street Journal (September 18. 2007. A. R. T. “Risk Aversion and Group Dynamics in the Management of Student Managed Investment Fund”..” Yale Bulletin 29.292 3. 1 (February 16).. and G.000 9. “Local News. & 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. “A Survey of University Student Investment Funds. 2007). “Record $18 million Closing for Largest Student-Run Venture Capital Fund. E. 9-18. 1984. 7781. Boyd. 1975. 2006.378 580 3. and L. Journal of the Academy of Business and Economics 1 (No.514 361 na 53 11 2. Alsop. 1). 2001.” Midwest Review of Finance and Insurance 14 (No.983 2 184 1. “Yale SOM Launches Student-Managed Venture Capital Fund.

L. and J. 1997. C. 1). and D. “The Student-Managed Investment Fund: A Special Opportunity in Learning. 1987. 1). S. 1984..” Journal of Financial Education 6 (Fall). “Learning Portfolio Management by Experience: University Student Investment Funds. 165-173. “Student-Run Venture Capital Firm at Cornell Invests in Local Tech Firm. Pfeffer. F. 2002. R. J. “The Student-Managed Investment Fund at the High School Level. Dolan. D.. J.. 1990. Rombel. 2006.. “Cameron University’s Unique Student-Managed Investment Portfolios. 55-59. 47-53. 3).. R. “TVA’s Student Investors Outperform Market.” Financial Management Collection 1 (No. McInerny. 65-67. Kester.. 55-60. C. 40-45. Alexander and G.” Advances in Financial Education 2 (Spring). W. 1-9. 1994.” Financial Practice and Education 6 (No. 12). 1). 2002. “Extending Investments Beyond the Classroom Through Investment Clubs. W. M. P. “What’s Right and Still Wrong with Business Schools. W. 395-405.. 2004. E. C. “A Survey of Student-Managed Funds: The Way They Work. Markese. 8. 1994. 1). 197200. Myers. “Business Conditions and Economic Analysis: An Experiential Learning Program for Economic Students. “Real Dollar Portfolios Managed by Students .An Evaluation.. 42-48. Allen. 9-10..” Biz Ed 6 (No. “Teaching Securities Analysis with Real Funds. D. D.” Financial Practice and Education 1 (No. K. Hirt. 1). “Student-Managed Investment Funds: A Comparison of Alternative Decision-Making Environments.. Cooley. Tatar.” Financial Practice and Education 4 (No. 5). 2007. T. D. J. J. TVA Managers. “CU Student-Managed Investment Fund Thrives After Change in Strategy. 5 Neely.” Financial Practice and Education 4 (No. 97-101. D. 2007. 1977. W. 4). French.. and J. Kahl. 1991.” presented at the Financial Management Association meeting. Bhattacharya.. “The Challenges and Opportunities of Student-Managed Investment Funds at Metropolitan Universities. “Financial Innovation: The Case of Student Investment Funds at United States Universities. D.” Journal of Economic Education 37 (No.” The Florida Times Union (April 5). P.” Financial Services Review 6 (No. Lawrence. 3). (October 17).” The Clearing House 76 (No. 1996. “Applied Security Analysis and Portfolio Management. 2003. . A.” Business Journal Central New York 21 (No. P. L. G. G. A. 252-254.” Journal of Financial Education 16 (Fall). 57-61.” Journal of Financial Education 13 (Fall).” Financial Review 25 (No.” Cornell Chronicle 35 (No. 1986. L. Johnson. B.LAWRENCE — STUDENT MANAGED INVESTMENT FUNDS 83 Mansfield. and P. McClung. 2004. H. Neely. 1). Lawrence.. 24). Stevens. W.. Block. “A Survey of Student Managed Funds. E.

Institutions and regulators are changing their risk systems and personnel to address this issue. Many small defaults are expected. We tend to underestimate these improbable risks due to behavioral biases. Quadrant B represents many internal operational risks such as check processing errors. The framework supplements current quantitative risk management by improving responses to risk changes over time. Behavioral economics provides insight into risk-assessment errors and possible remedies. Behavioral finance can improve how risk decisions are made. political power and capital flows to successful individuals. No matter how remote. Screening helps identify groups with higher default probabilities. Unfortunately.84 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Behavioral Basis of the Financial Crisis Joseph V. The costs are absorbed and the focus is J. When challenged by events. I. Our existing risk measures account for perhaps 95% of what occurs. The views expressed represent those of the author. Quadrant A events include retail credit products including credit cards. such as misaligned compensation systems. These groups are charged higher rates to offset the risk. A further concern is the likelihood of repeating this experience during the next cycle. however. and applies it to the structure finance market. How this could have occurred given sophisticated tools and massive risk system investments is a concern. Yet despite these advances. Behavioral Finance Framework Risk can be classified along two dimensions. the cause-effect relationship is unclear. The first concerns high-frequency events with relatively clear causeeffect relationships. we try to explain away the events. Behavioral finance recognizes that decision processes influence perception and shape our behavior. The major catastrophic risks lurk in the fat tails of the remaining 5%. and not CapGen Financial. Major strides were made in quantitative risk management during the 1990s. You ignore behavioral risk at your peril. Consequently. Other risks occur infrequently.V. less is known on how these decisions are actually made. The problem. high-impact events cannot be ignored because they can threaten an institution’s existence as was demonstrated in the current market crisis. 84 . Most individuals have a model of how the world works. financial institutions suffered large losses following the collapse of the subprime and structured products markets. is not only with the systems or the quality of the personnel but within the individuals themselves. This article outlines a supplemental behavioral risk framework. The dimensions are reflected in the risk map in Figure 1. it is difficult to determine whether they are truly successful or just lucky. The framework will then be applied to the structured finance crisis. Major risks are frequently ignored due to behavioral biases resulting in incorrect decisions. The second dimension is impact severity. In most institutions. Rizzi Financial institutions suffered large losses following the collapse of the credit markets despite making huge risk management investments. Rizzi is a Senior Investment Strategist at CapGen Financial in New York. however. NY. Risk management should encourage profitable risk taking while discouraging unprofitable and catastrophic risk. These biases are reinforced by organizational obstacles. This article outlines a behavioral risk framework to address judgment bias and develop appropriate responses. Although we know how risk decisions should be made.

Low frequency/high impact events: frequently ignored. 1 Second. Risk estimates become anchored on recent events. Tversky and D. selling winners. “Advances in Prospect Theory: Cumulative Representation of Uncertainty. Overemphasis on recent events can also produce disaster myopia during a bull market. Uncertainty and Profit. Investors incurring such risk can expect mainly small positive events but are subject to a few cases of extreme loss. Consequently. It recognizes models can influence behavior and shape decisions. In this case. and herding. We try to minimize regret by seeking confirming data. there is insufficient data to determine meaningful probability distributions. Sunk-cost bias involves avoiding recognizing a loss despite evidence the loss has already occurred and a further loss is likely. The difficulty stems from two factors. 2 D. Low frequency/low impact events: treated as a cost of business. Houghton Mifflin. Kahmeman. Prospect theory is a key component of Behavioral Economics. Consequently. which will now be investigated. These models build on expected utility theory (EUT). as instruments are priced without regard to the possibility of a crash. and process information. The value function in prospect theory is based on gains or losses relative to a reference point. These beliefbased heuristics incorporate biases or cognitive constraints. Ellsberg. rules of thumb. First.RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS 85 on how risk managers should act. no amount of mathematics can tease out certainty from uncertainty. 1992.5 Risk managers have developed shortcuts. Instead of being expected utility (E(U)) maximizers.” Quarterly Journal of Economics 643. Black 1986. This is reflected in Figure 2. applying its concepts to asset pricing. Regret underlies several biases. however. “The Utility Analysis of Choices Involving Risk.” Journal of Risk and Uncertainty. See F.3 Examples of anomalies include holding losers. This article uses the terms interchangeably. It focuses on responsibility for what we could have done but did not do. 4 5 See E. randomness with unknowable probabilities. Quantitative risk-management models are based on portfolio and option pricing theory and provide a framework 1 This is the Knightian distinction between risk. Type C events include concentrated exposures to high risk borrowers. Regret. 3 A.4 can explain these facts. Figure 1. or heuristics to process market signals. 1921. Information concerns facts affecting fundamental values. Specifically. Risk.” Journal of Finance July. leading to suboptimal results as emotions override self-control. Cyclical risks are low-frequency-high-impact events characterized by their negative skew and “fat-tailed” loss distributions. These facts lead to risk mispricing and the procyclical nature of risk appetite. not predictive. “Noise. Market signals are complex.“Risk. Behavioral finance examines how risk managers gather. Type D events are frequently ignored due to a low frequency. Risk Map Frequency A B I C D Impact A: B: C: D: High frequency/low impact events: reflected in risk pricing. Examples include many of the structured finance products which represented short positions in an option. Friedman and L.2 Empirical support of EUT is mixed with numerous reported anomalies. A. on mitigation and prevention through improved processing and training. and taking comfort that others made the same decision. and perhaps more important. suppressing disconfirming information. They offered long period of steady income punctuated with occasional large losses. Regret Risk is forward looking. usually par or the original purchase price. Sunk costs are the first regret bias considered. 1948. High frequency/high impact events: managed through control. prospect theory. it concentrates on perception and cognitive bias. . Noise is a random blip erroneously interpreted as a signal. Knight 1921. Savage. These well known risks are managed by constant management monitoring and control.” Journal of Political Economy. interpret. which views individuals as expected utility maximizers. Ambiguity and the Savage Axioms. These risks are difficult to understand. An alternative. These biases can corrupt the decision process. M. is backward looking. excess trading. EUT focuses on wealth changes. investors are viewed as expected regret (E(r)) minimizers focusing more on losses than gains. They include both information and noise. Behavioral finance is a subset of Behavioral Economics. regret can inhibit learning from past experiences. randomness with knowable probabilities and uncertainty. the statistics are descriptive. which builds on their earlier work. 1961. infrequency clouds hazard perception.

however. In a crisis. Investors Minimize Expected Regret JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Value function value + . however. the reference is pessimism. Therefore. For example. they acquire underpriced. Minsky financial instability hypothesis. to minimize regret. which inhibits This is consistent with the H. 6 This is magnified by the naïve use of market-based risk-management tools. This illustrates the George Soros reflexivity or feedback principle. Convex slope indicates pain of loss (regret exceeds value of gain . Thus. Positive feedback is self amplifying. Structured finance bankers and quantitative risk managers took credit for results during the boom. risk and regret Utility Losses Gain Reference point - Examples include the reluctance to sell impaired assets at reduced prices. Larger. while negative feedback is self corrective. 8 . risk appetite increases with wealth. Overconfidence Overconfidence occurs when we exaggerate our predictive skills and ignore the impact of chance or outside circumstances. Panic conditions are also based on a combination of regret and herding. better-capitalized financial institutions can absorb more risk than smaller institutions. whereby markets affect psychology and psychology affects markets. reflected in Figure 3. Risk and return are. Their greater risk tolerance lessens their downside sensitivity. which fuels further gains. This further depresses prices leading to continued forced selling and the creation of a negative feedback loop as occurred in the fourth quarter 2008. failing to consider the impact of randomness and mean reversion creating an illusion of control. 7 Studies indicated the underestimate at 15 %-25 %. Risk managers will assume greater risks when they are up in a bull market and lower risk in a bear market. we follow the herd not to be left behind and engage in panic selling. Minsky. higher-yielding. This procyclical phenomenon leads to “buy high and sell low” behavior. This increases their access to lower priced funding and liquidity. Stabilizing an Unstable Economy McGraw-Hill. Larger institutions confuse the ability to absorb risk provided by capital with the desirability of the risk position. The conflict between E(u) maximizing and E(r) minimizing underlies many anomalies . Finally. illustrating in this instance. Self-attribution involves internalizing success while externalizing failure. Regret is perceived to be less when risk of winnings is involved. See H. scale invariant. reject the logical alternative of acquiring additional exposure at the market price to exploit the alleged under pricing. 2008. than risk of initial capital. higher-risk assets in bull markets. The direction of the overconfidence is usually positive reflecting a related optimism bias. regret leads to confusing risk with wealth. Most institutions. Investment decisions involve 3 Rs: return. It results in an underestimation of outcome variability. Another regret-related bias is the house money effect. Investors increase their risk exposures driving bull markets until they have taken on too much. Usually this is defended as the market prices being too low. especially during a bull market when income levels are high.7 Overconfidence is reinforced by self-attribution and hindsight.6 B. price is of secondary importance relative to regret. collateral values rise during a bull market. Thus.86 Figure 2.8 Hindsight involves selective recall of confirming information to overestimate their ability to predict the correct outcome. thus. and we actively seek bad news to confirm our belief.

Control reflects the unfounded belief of our ability to influence or structure around risk.RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS 87 Figure 3. 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.? Risk Level Bear Low Low Financial Institution Profitability High learning. Knowledge is frequently confused with familiarity. Seru. not the answer. not calculation. This was especially true for mortgage default models. in part. Rajan. and uncertainty. known unknowns. • Homogenous populations: Statistical models require large homogenous populations with a long history of observations. Optimism is heightened by anchoring when disportionate weight is given to the first information received. Key model limitations include the following: sequence and is not an independent observation. The new structured finance credit portfolios were small.” Working Paper. This is usually based on the original plan. Structured Finance Risk Appetite Low High Housing bubble 2004-1H07 Bull High Market State H Contrarian 2H07 . • Statistical Loss Distribution: Loss distributions for credit are skewed. elements. to misguided overreliance on quantitative credit scoring models without understanding their limitations. and concentrated with limited histories. Risk/Market Appetite. Industry and product experts are especially prone to overconfidence based on knowledge and control illusions. • Historical basis: History is a guide. Models tend to be blinded by the mean and underestimate extreme events. heterogeneous. with unexpected event losses hidden in the distribution’s fat tails.9 This is due. Financial models adequately contemplate the former but inadequately deal with the later.10 • Uncertainty: Decisions involve both risk. 2008. They ignored the impact of securitization of mortgage originator underwriting practices. independent of the likelihood of being better or faster at identifying risk than the market. “The Failure of Models that Predict Models: Distance. A. V. This reflects an optimistic underestimate of costs while overestimating gains. Disappointment and surprise are characteristics of processes subject to overconfidence. Examples include the perceived ability to distribute or hedge risk. Risk is accepted because we believe we can escape its consequences due to our ability to control it. 9 U. History becomes less relevant as markets and underwriting practices change. Incentives and Defaults. The past represents but one possible outcome from an event Inappropriately designed incentive compensation reinforces overconfidence. Managing uncertainty requires judgment. September. unknown unknowns. Vig. 10 . which tends to support the transaction.

frequently fooled by randomness. skill. particularly when recent occurrences skilled banker. all lucky streaks come to an end as this one did underwriting standards to increase short-term income. of the potential negative long-term effects. This causes a false sense of security as risk is measures are unable to distinguish results based on luck versus underestimated. Scholes.88 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Time-delayed consequences magnify overconfidence as purchased to hedge the position. Free Press. they come prices would not decline nationwide in the US from trades by real people. Since the subjective Institutions find it difficult to accept chance and are probability of an event depends on recent experience. These types of events are ignored events. 2007).” Harvard Business Review (March-April). independent of recognizes that models can confusing beliefs for historical experience. accounting period. W. The major error focused on the belief that housing determined by random number machines. How Nature Works Springer-Verlay. the longer the time state of the world may change rendering current state data period. whereby we see patterns in random collapse. 1996. triggering unexpected losses. 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. in accordance with our many of the projections expectations. impact events are underestimated. The Halo Effect. or assumed away. For example. trades Perhaps the most dangerous statistical bias is disaster and people complicate statistical modeling. We tend to view the unfamiliar as improbable. this resulted in decisions. In fact. May. it individuals weigh short-term performance at a higher level creates incorrect capital allocation incentives. Taleb in The Black Swan: The Impact of the Highly Improbable (Random House. An example is dropping credit Eventually. We interpret short-term success as “hot hands” by a or deemed impossible. or league table status Another statistical error as occurred during the height prevalent during a boom is Behavioral finance examines how of the boom. 2007. Bak. “Crisis and Risk Management. the higher the likelihood of a “Black Swan” event obsolete. 11 12 15 P. we believe that Examples include the numerous apparently lucky real estate they will not happen to us. Formerly diversified positions begin moving together. and capital is misallocated.12 Rather. There are Actions and outcomes can be unrelated. psychological aspects of risk decision making. Since this is not priced. prices. “The New Religion of Risk Management. They are D. Herding unexpected because such movements are unfamiliar. It continue into the foreseeable Statistical bias involves future. 13 Black swans are high impact unexpected rare events. P. Unlikely events are neither impossible or remote. Princeton University Press 11. 2000 14 See P. Consequently.14 Thus. 16 . 1996. Statistically based risk management practices are inherently unlikely events are likely to occur because there are so many limited.” AEA Papers and Proceedings. Investors and Markets. influence behavior and shape chance by selecting evidence Undoubtedly. C. reserves you must reflect the values of the options not M. Prices are not proposals. The term was popularized by N. sample probability and skill for size or mean reversion. and invalidate myopia.11 Economics is underlying structured credit a social science based on human behavior. extrapolation bias. like a market or firm representative bias.15 Thus. are very small. market during the summer of 2007. 2007. These occur whenever short. A manifestation is the expectations of low-frequency events. and the improbable is The previous discussion concerned individual frequently ignored.13 As Scholes notes. Statistical trends are assumed to and process information. the than longer-term consequences. Sharpe. Feedback loops. at least during the current experts at institutions like Bear Stearns and Lehman. share. to value risk or price and ‘group think’. switching from a boom to a bust cycle occurring. Bernstein.“lucky fool” is rewarded and encouraged with bonuses and term benefits clash with long-term effects. Risk-adjusted return on capital and other are lacking. Rosenweig. This occurs when current events or risk managers gather. This occurs whenever low-frequency but highthe use of normal distributions as used in the physical sciences. They are unable to reflect the hidden risk that the unlikely events that can occur.16 impacts correlations. Although we know increased capital until luck turns and losses are incurred. interpret.

which can produce short-term self-fulfilling prophecies.RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS 89 Consequently. Investors rush in to avoid being left behind using rising asset values to support even more credit. “Inside Wall Street’s Culture of Risk. This explains why bankers continued risk practices even though they feared this was unsustainable and leading to a crisis. This belief was based on complex risk models and market diversification. including Bear Stearns and Lehman. Additionally. which triggers an asset price decline. The experts tend to play it safe by staying close to the crowd and extrapolating past performance. The recent subprime collapse illustrates this fact. the experts were taken by surprise by a supposed perfect storm. Herding reduces regret by rationalizing that you did no worse than your peers. and an exit of investors. due to a lack of critical thinking. behavioral risk factors will dominate and quantitative risk measures will be unreliable. Chevalier and G. Herding amplifies credit cycle effects. 1997. Individuals perceive these signals as information even though they may be reacting to noise. Through herding. believing they were under control.” Journal of Political Economy. 21 22 H. This causes losses. you can blame any failing on the collective action and maintain your reputation and job. enhances cognitive biases. Even though you recognize market risk. than to succeed unconventionally. It occurs when individuals identify with the organization and uncritically accept its actions. as most large financial institutions adopted best practices based on similar experts. The industry participants used the same consultants and models for their projections. Essentially. Sentiment risk is zero in an efficient market. but macro inefficient regarding the market as a whole. they focus on arbitrary market silos that may be in a downturn. During a late stage boom with high sentiment levels. you have to get up and dance”. They tend to limit information from all but other expert sources. as decisions become more uniform. 18 Relative performance measures are a form of sophisticated “me-too” metrics. individuals avoid falling behind and looking bad if they pursue an alternative action. are surveyed.20 Experts are prone to group think. Group Think Group think. When the crash occurred. Thus. 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. 19 E. It constrains both envy during an upswing and panic in a down market. Paul Samuelson has noted markets in the short-term can be micro efficient concerning individual instruments. a decline in credit. Sentiment Risk The aggregate investor error based on biases is sentiment risk. 17 This is reflected in Keynes” famous statement that it is better for a banker ’s reputation to fail conventionally. This is referred to as a positive feedback loop or momentum investing. Thornton. 20 E. Shefrin. an event occurs. or organizational pressure. 2006. mutually reinforcing individual biases and unrealistic views are validated. Ending the Management Illusion. This is reflected in the famous comment “As long as the music is playing. This is characterized as irrational exuberance where prices are driven principally by momentum and herding See J. This is illustrated in the 2006 Business Week cover story in which risk officers at numerous institutions. Carter. Rather than focus on absolute value creation. Ellison.22 Most risk models ignored sentiment risk. which strains market liquidity. . A cascade is a series of self-reinforcing signals obtained from the direct observation of others. The consultants based their reports and recommendations on the surveys of industry participants. it was reinforced and accepted uncritically. and reflects safety by hiding in the crowd. “Risk Taking by Mutual Funds as A Response to Incentives. Henry and A. such as a move by the central bank. it pays to follow the crowd. 2008. It is based on the social pressure to conform. A. F. The faith in risk management encouraged institutions to increase their risk exposures.17 In so doing.” Business Week (June 12). during the short-term the direction of the inefficiency is likely to widen due to momentum and herding. McGraw-Hill. The industry expert impact is significant. It can be either optimistic or pessimistic and is time varying as reflected in Figure 4. Once the commitment is made. The cycle begins with a credit expansion leading to an asset price increase.19 A related effect is an informational cascade. Herding occurs when a group of individuals mimic the decisions of others. 18 Herding underlies why banking experts’ forecasting abilities are poor. This is critical in banking when performance contracts are based on relative performance measures tied to peer groups. Managers learn to manage career risk by clinging to an index. inconsistent information is suppressed. they repeat statements until they become accepted dogma regardless of their validity.21 They believed that despite the risks taken they were safer than ever. Eventually. Investor responses based on the interplay of sentiment and market valuation is reflected in Figure 5. D. principal loss is converted into benchmark risk. Once the perception of a bull market took hold.

D represents irrational despondency found in market bottoms reflected in tight liquidity. II. Recognizing and dealing with biases is complicated by three factors. Behavioral Finance (Garant Uitgevers. Consequently. Price may diverge from fundamentals 23 . When sentiment is low. fundamentals will rule as in B and C. We base our actions on experience of what has happened. 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. Remedies Behavioral finance demonstrates how biases influence risk perception leading to underestimation of improbable events. we misjudge actual risk leading to surprise losses. Next. this leads to See T.90 Figure 4. 23 . Finally. but are quickly eliminated by arbitrageurs in B and C. while searching for conforming information. 2006) 183 for a discussion of various forms of behavioral finances that can occur in markets. Sentiment High Low Market Valuation Boom A B Bust D C reflected in high liquidity levels. First. Debels. we diminish information inconsistent with our existing views. bias can be amplified within organizations due to incentive misalignment and group think. This ignores beyond the data exposures leading to future blindness.

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

putting pressure on institutional accrual and trading budgets. resulting in falling risk premiums increasing credit asset prices. in 2005 it was replaced by a new team. 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. • Institutional overconfidence in risk management models based on the illusion of control.g. Setting A herding among financial institutions occurred during the last several years. Persaud Paradox is the observation of safety created by using the same models as your peers. They failed. The pressure to herd is illustrated by Morgan Stanley. Morgan Stanley refused to participate in structured products.g. leveraged buyout and structured credit markets. yield curve reduced the attractiveness of the carry trade. as the Morgan Stanley example illustrates. This was reflected in historically low credit-risk spreads in the real estate. This model allowed institutions to rationalize poorly structured. leaving them with large high risk exposures. A credit bubble formed as liquidity-driven technicals surpassed fundamentals. Its performance suffered relative to its peers.24 • Peer pressure that was not based on independent economic reasoning. Liquidity-driven technicals improved. 24 . ignoring peer pressure can be hazardous to your career. they invested too much at the same time in the same areas. which creates as illustrated by the Peraud Paradox. In the search for yield. This was done under guise of the “originate to distribute” model. These activities represented up to 75% of revenues at some institutions. Some institutions like Wells Fargo and Pittsburgh National escaped the herding. merchant banking. Thus. It is difficult to determine if they were lucky or smart. Consequently. Under its previous management.. The strategy is reflected in principal finance. to consider the impact of markets closing down. which it achieved in 2008 by recording record losses. multibillion dollar mortgage warehouse facilities) • Leverage levels approaching 30:1 • Less liquid assets (e. the key is the overall asset-class allocation and not necessarily the number of assets in an asset class.92 Figure 6. This causes an observation of safety. collateralized debt obligations) • The 5Ls strategy involves going long on higher-risk assets for the institution’s own account instead of distribution. Consequently. however. Nonetheless. Institutions responded by adopting an assetintensive carry trade strategy. mortgage backed securities) • Long tailed option type risk found in the AAA tranches of structured securities • Large positions (e. institutions adopted a procyclical asset heavy 5Ls strategy: III. risk. Spread narrowing and a flattening • Longer duration (e. They vowed to regain market share by matching its peers. A declining economy and falling markets triggered aggressive Federal Reserve interest rate cutting and liquidity injections in 2001 to 2002.. which involves borrowing shortterm to invest in longer-term risk assets.g.. which creates model risk. bridge loans and warehousing activities. Current Crisis A. 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. underpriced products by selling them to others.

Finally. Procyclical risk appetite is aggravated by ratings-based regulatory capital requirements. Thus. B. financial institution compensation is tied to peer group comparisons. Warning signs began to form during the first six months of 2007: D. The tipping point represents a change in investor sentiment based on an awareness of the risk that investors have assumed. Such cycles. The housing event occurred in mid-2007 and has continued for more than eighteen months. curtailed credit. conservative risk managers and bankers are pressured to become optimistic or leave. falling asset prices and reduced liquidity. This was subsequently discovered during the crisis. prompt investors to adjust simultaneously their positions triggering a decline in asset prices. the strategy involves incurring increased systematic or beta risk exposure versus value-adding alpha returns. Widespread credit risk under pricing existed due to an emphasis on nominal returns. Thus. leading to over optimism. liquidity risk was under reported. positive short-term results mask long-term risks.” 27 . This risk was poorly reflected in risk management models. and the warnings were ignored. Individuals and institutions succumbed to a bias of “assuming the absence of evidence implied evidence of risk absence”. are difficult to manage for several reasons. Overvalued assets. Tipping points represents triggers. NJ. The premiums appear profitable until the put event occurs. 26 See D. Princeton University Press. while predictable. 25 President of New York Federal Reserve T. Seemingly high returns can reflect the subjective probability of an event that has not occurred in the time period studied. This caused a major credit boom. Conversely.27 Figure 7 shows that the decline can trigger a vicious cycle leading to reduced collateral values. • Continued Federal Reserve tightening • Rating agency downgrades • Flattening yield curve • Increased mortgage defaults Unfortunately. Risk models also contribute to the problem by presenting the illusion of safety and control. are prone to volatile investor sentiments. These systems favor “consistent” earnings and misread low –frequency/high-impact risk “profitability. thereby encouraging credit expansion. C. risk sensitivity had diminished. feedback overwhelms fundamentals and the trend dominates. 2005.26 As risk appetite increases. 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. Thus. declining investor demand. Frequently. Regime Changes Procyclical risk appetite and feedback loops underlie credit cycles. they increase capital requirements during a bear market as ratings are pressured. The regulatory rules reduce capital requirements during a bull market as ratings increase. Investors use the increased debt capacity to bid asset prices higher. This suggests a correction when investor emphasis shifts from return on capital to return of capital. leading to a credit contraction. Firms continued to underestimate the likelihood and impact of unlikely events. institutions risk losing bankers if their risk activities are curtailed. The liquidity premium was mischaracterized as alpha. Maintaining discipline becomes increasingly difficult as the cycle continues. Consequently. costing billions in provisions. Sornette. which supports additional credit expansion creating a virtuous credit cycle with increasing liquidity. Next. Finally. First. Geither. Structured products are less liquid than market investments.25 The last market correction had occurred more than three years ago and was largely forgotten by the first half of 2007. Once the tipping point is reached. First. not causes of the change in investor actions. a beyond the data event occurs.” Such risk is similar to underwriting out-of-the-money put options. firms and individuals not following their peers suffer. May 12. Concerns The appropriateness of the 5Ls portfolio strategy depends on several factors. pricing and trading discipline is needed to ensure an adequate risk premium is earned.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. 2003. apparent success breeds an inability to imagine the possibility of failure. Therefore. tipping points are unexpected and occur during the height of an over-valued bull market. however. organizations frequently discourage pessimism. it works best early in the cycle before the opportunities are exploited by the competition and spreads narrow. Why Markets Crash. During the boom. Next. which are vulnerable to bad news. This is similar to physical events such as forest fires and earthquakes arising from “criticality. The higher asset prices increase collateral values. credit extension expands. as quoted in the Financial Times. Princeton. the return on structured products reflects compensation for liquidity risk. This recognition problem is rooted in the complex nature of cyclical risk. Investing in such instruments is profitable most of the time. Eventually. Thus. prices can exceed underlying fundamental economic values as illustrated in Figure 4. A tipping point or event can.

The process is similar to earthquake engineering. Rather. This fact underlies the procyclical bias in portfolio strategies as lower bull market capital requirements increase returns. Goodhart. This is especially true when dealing with high impact low probability risks. Persaud. 2005. Ch. Supplementing currently determined capital charges with a requirement tied to asset prices would encourage a shift to a counter cyclical portfolio strategy. Ignoring these facts substitutes an inaccurate normative model for the real world. Unfortunately.il). The observation of safety created by using the same models as your peers creates model risk. 28 during the credit cycle as risk management evolves to a more balanced system. The underlying exposure builds during a bull market as apparent risk declines. counter cyclical capital charges decrease during bull markets as ratings improve as demonstrated in Figure 3. Market-Liquidity and Risk management. while the losses materialize in the bear market cycle.28 • Difficulty reflecting out-of-sample.” where the observation of safety creates risk. London: Risk Books. Some important model issues include the following: • Inadequate consideration of the cyclical effect on. This anomaly is due to social and psychological biases resulting in bounded rationality. The higher capital allocation serves as a risk-taking budget constraint during bull markets by dampening compensation-related returns. and correlation among. financial markets are not actuarial tables. Liquidity Black Holes. models underestimated low frequency/highimpact cyclical risk. See A. it can reduce future losses This is the “Persaud Paradox. the focus is on constructing a structure to withstand a certain shock level. encouraging an inappropriate. The specific action taken depends on the source of the bias. Consequently. loss given default and exposure at default • Poor understanding of the interaction between liquidity and credit risk as bull markets create their own liquidity. Capital levels would relate to changes in asset prices. Credit Risk. incorporating human behavior. 11. Figure 8 See C.? Asset Prices Tipping Point Asset Prices Investors Collateral Values Investors Collateral Values Credit Credit Backward-looking risk models confuse history and science. “Financial Regulation. Governance mechanisms represent possible control over the bias by introducing outside viewpoints. 2003.94 Figure 7. Lessons Learned Risk decisions are at a risk from behavioral bias. which does not attempt to predict a shock. 29 E. Currently. and developing appropriate responses. 118. This requires taking low-probability-worst-case scenarios seriously. and Financial Stability. which can evaporate in a downturn • Feedback impact of models on markets is ignored. probability of default. 29 . Credit Cycle JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Virtuous Cycle (Disaster Myopia) 2004-1H07 Vicious Cycle (Disaster Magnification) 2H07 . The objective is to supplement existing quantitative risk management with developments taken from the evolving field of behavioral finance. 5L portfolio strategy. beyond-the-data possible effects.” National Institute of Economic Review ( Apr. asset-heavy. In so doing.

interviewed by N. individually. A. You ignore behavioral risk at your own peril. the situation becomes problematic. D. Merton. We chose it. we must resist the temptation to say. When both managers and markets are biased.” Technology Review (April 2).”30 Behavioral finance offers a means to choose wisely. Boards and regulators are likely to fall prey to the same behavioral biases as affecting mangers and controls are likely to fail. Biased managers operating in an efficient market. This difficulty is compounded by behavioral bias reinforced by institutional factors. 2008. Defaults are increasing and liquidity remains fragile. need to be protected by their boards of directors and regulators from overreacting to market noise through tight controls.RIZZI — BEHAVIORAL BASIS OF THE FINANCIAL CRISIS 95 Figure 8. This entails adopting counter cyclical portfolio strategies despite negative short-term revenue implications. Nickerson. compare the test results to our risk appetite and take appropriate portfolio decisions. which characterized the late stage of the boom. stress-test to determine their impact. Conclusion Presently. “On Markets and Complexity.C. products and structures. Rational managers operating in a biased market. the more difficult it becomes to manage risk. . or collectively is not a physical given constant. The key is to identify potential adverse scenarios.” The deeper we are into illiquid credits. it is difficult to consider the end to the bear market. This leads to a potentially fatal neglect of the longer-term build of risk. will exploit market inefficiencies by selling over priced claims. as it affects both individual decision making and market efficiency. “This time is different. Bias Response Choices Managemen Biased Rational Rational Investor Markets Biased A A B D C reflects the possible combination of market and managerial bias. This also requires adopting difficult infrastructure changes. IV. 30 R. Organizational obstacles inhibit appropriate responses to high-impact low-probability risks. As Robert Merton noted “The amount of risk we take personally. While no two cycles are identical. C. Classical financial theory is represented in B with efficient markets and rational managers who require limited oversight. Chief among the obstacles are short-term compensation systems which reinforce behavioral biases.

. T.. C.” Harvard Business Review 72 (No. McGraw-Hill.” Journal of Political Economy 105 (No. Quantifying and Managing Financial Liquidity Risk.” Chicago GSB Research Paper No. 1996.. F. Ending the Management Illusion.. and G. Ellison. 6). “Advances in Prospect Theory: Cumulative Representation of Uncertainty. 643669. 529-543. 1167-1200. M.96 References Bak.. . NJ.. NJ. 4). “Financial Regulation.. Taleb. 279-304. Credit Risk and Financial Stability. The Halo Effect—and the Eight Other Business Delusions that Deceive Managers. Princeton. “The New Religion of Risk Management. 2007. 2000.. Investors and Markets: Portfolio Choices. A. 4). JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Persaud. “On Markets and Complexity. E. 2008. D.. Henry. and L. N.. 1986.” Journal of Finance 41 (No. “Risk. 2006. Ross School of Business Paper No.” Business Week (June 12). 1999. P.. Scholes. H. A. 2007. 08-19. New York. “The Failure of Models that Predict Failure: Distance. Antwerpen. Why Markets Crash. Incentives and Defaults. Thornton. F. A. Rosenzweig. “Crisis and Risk Management. Springer Verlag. New York. Columbus. 297-323. Seru. Shefrin.. Savage. OH. Behavioral Finance. 2008. 118-127. Assets Prices. and Investment Advice. and A. D.C. “The Utility Analysis of Choices Involving Risk. McGraw Hill Professional. and D. 1992. Princeton. R. The Black Swan: The Impact of the Highly Improbable. Technology Review (April 2). Chevalier. Merton. Friedman.. 3). Carter. 2008. M. Ambiguity and the Savage Axioms. P. Simon and Schuster. Sharpe. “Noise. 1948. Liquidity Black Holes: Understanding. 1961. 17-21.” American Economic Review 90 (No. Sornette. 1122. U. J. Princeton University Press. London. Vig. Tversky. 2). Stabilizing an Unstable Economy. Black. Minsky. Goodhart. H. 2008. OH.. 2007.” Interview by Nate Nickerson. 4751. Risk Books.” Journal of Risk and Uncertainty 5 (No. Princeton University Press. 2006.” National Insitiute of Economic Review 192 (No..M.” Quarterly Journal of Economics 75 (No..” Journal of Political Economy 56 (No. New York. Garant Uitgevers. Random House. Kahmeman. Ellsberg. How Nature Works: The Science of Selforganized Criticality. “Inside Wall Street’s Culture of Risk. 2). Debels. 2003. 2003. and V. Bernstein. 2006. 1997. Columbus. W. 1). P. “Risk Taking by Mutual Funds as A Response to Incentives. Rajan. 4).P. D..

we’ll open up the discussion to questions from the audience. We feel very privileged to have him at the University of Rochester. Bailouts. 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. Charles Hughes. Mark. I’m not going say much about the topic itself—I’ll leave that to our panelists. I’ve asked each of our five panelists to provide a brief statement of their thoughts on the problems of the US auto industry. when the market’s up. Cliff Smith. is a “negative beta” activity. and Jim Brickley—are distinguished academics from the Simon School faculty. In other words. The Social Function of Bankruptcy: Uses and Limitations Tom Jackson: Thanks. James Brickley. I’m Mark Zupan. to those problems. and the potential role of bankruptcy in dealing with them. bailouts because I suspect we haven’t seen the last of businesses—or industries—facing such choices. NY February 2. Tom. would you please start things off for us? I. Douglas Baird. he was the provost and dean of the University of Virginia Law School. and Clifford Smith Moderator: Mark Zupan  Mark Zupan: Good evening. and possible solutions. for the kind words. is bankruptcy—and bankruptcy is a process for reorganizing troubled companies that is rooted in the economic goal of increasing efficiency. So. and so does the amount of attention and effort devoted to bankruptcy. who are the experts. and welcome to this discussion of a very topical and pressing issue: today’s problems with the US auto industry. as Mark told you. Before coming to Rochester in ’95. Our first speaker will be Tom Jackson. Charlie as an auto company executive and Joel as a bankruptcy lawyer. I want to begin this discussion by providing a broad economic framework for this issue of bankruptcy vs. including Chapter 11. 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. Since stepping down from that position. 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. My field. are a means of 97 . he has held joint appointments at both the Simon School and in the University’s political science and economics departments. But when the market’s down. Dean of the University of Rochester’s Simon School of Business. Three of them—Tom Jackson. both business school applications and bankruptcy cases tend to go down. Tom—along with his former student. 2009 Panelists: Thomas Jackson. What I will tell you is that bankruptcy.University of Rochester Roundtable on Bankruptcy and Bailouts: The Case of the US Auto Industry The GeVa Theatre Rochester. After we hear from each of them. our applications go in the reverse direction. From 1995-2005. We have five panelists tonight. by contrast. and I will be serving as moderator. Joel Tabas. Tom served as President of the University. like business school applications.

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;

CAFÉ. Fourth. every economy that we would consider to be an economic powerhouse has cultivated a strong. not one-way free trade. six out of ten constituents told their Congressmen to let them die. Ford. 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. We have gambled our money away and are left staring at our gambling debts. Fiat has volunteered—and we should see whether that marriage can work. So what’s to be done? Here is my short list of suggestions: First. Korea. is viewed by industry experts as the equal of Toyota in running the most efficient plants throughout North America. But with that said. Not just for the jobs. let’s make sure that when we think of our auto industry. for example. And. finally. at least two big names in technology would follow into Chapter 11. 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. and the car is the Cadillac Escalade Hybrid. As our government tosses around trillion dollar fixes. Britain. I think our government should continue to support Chrysler until we find out. 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. treat each of the Detroit car companies according to their degree of distress and specific circumstances. and if we get it wrong we are in real trouble. homebased car industry. France. we believe in fair trade. 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. Their car companies are vitally important to them. which I think we need to help find an international partner. with 13 cars. You can’t get much bigger than that—and the car gets 20 miles a gallon around town. Germany. which is the company that’s in the most trouble of the Big Three. or the exports. we should support them. our government has played no small role in creating the problems of our auto industry in Detroit. BRICKLEY. It’s got too much debt. and in ways we don’t fully appreciate. And. During the recent hearings in Washington. and they play the game as a team sport. but during the Congressional hearings in December. We apparently don’t understand how today’s world operates. (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. there’s the case of Chrysler. I might also add that Chrysler. All those countries fight fiercely for the success of their homegrown car companies. On too many issues. If they end up needing it. do the words hubris and quick fix come to mind? Starting with the credit crisis. What line of cars had the best JD Power rating for initial quality in mid-size cars. and now China all view their auto industries as springboards to economic growth. Chapter 11 has never been tested on an industry that is so intertwined with our entire economy. we are appallingly ignorant. a solid cash base. What got GM in trouble were its hubris and quick-fix mentality. too many brands. but because the foundation of technological development in these countries—and ours as well—is the auto industry. let’s make sure we do it with street smarts and guts. Second. GM is a different story. Let’s have one set of regulations for emissions and fuel economy nationwide. and they haven’t taken any money yet. We are at a crossroads. AND SMITH — BANKRUPTCY AND BAILOUTS 103 we don’t understand how the business that made us wealthy really works. You may be surprised to know this. 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. to the Wagner Labor Act. HUGHES. do we want to be a nation of builders—or money changers? Thank you. Let’s raise the federal gas tax. Now to my third prescription: if we are determined to push the social agendas of energy independence and climate control. I’ve worked for eight different car companies. we are a world champion athlete going to seed. is in reasonably good shape: they have a good plan. and Japan—and I have consulted for the Koreans. do no harm. and they’ve already borrowed money—and it needs to demonstrate its long-term viability to receive more. But these people are clearly unaware of some important realities. But here’s the irony I see in what’s going on today. We are mad at Detroit. I can’t imagine this country without them. and six were importers—from Germany. Who builds a large SUV hybrid that gets better mileage than the Toyota Camry? The answer is General Motors. Japan. But this is not a lab experiment. finally. We will debate tonight whether bailout or bankruptcy is the better course. Italy. The US car companies are far better than you think—though admittedly not as good as they need to be. Number two was Honda.JACKSON. TABAS. . and transplant factory tax subsidies. Since World War II. 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. Sad to say. with 16 cars.

He’s an accomplished scholar in organizational economics. The banking and financial system in an economy is like the circulatory system in a human being. But having said that. Mark. economies can’t function with major disruptions in the flow of credit. Failures in the industry will have harmful effects on many people—including people who work for other auto-related companies—and the overall economy. have problems. 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. while at the same time being resistant to the idea of bailing out the auto industry. allowing a large manufacturing company to file for bankruptcy. 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. But another critical problem is the inefficiency stemming from their number of brands and models and from their distribution systems. the Big Three auto companies developed much of their product lines and dealer networks starting back in the 1950s and ’60s. we all hope that productive solutions to these problems can be found. when they dominated the US auto market. when discussing the problems of the Big Three auto companies. 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 dealerships. and it helps to support many other jobs throughout the economy. People often ask why the government has been so quick to bail out banks and other troubled financial institutions. Wall Street analysts. even one as large as GM. As Tom told us earlier. and on their labor costs and debt. Morgan Stanley. 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. 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. is widely regarded as an expert on distribution systems. let’s talk briefly about the problems with US banks and financial institutions. 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. which are collectively less than 50% of the US market. and to our R&D effort as well. and compensation policy. well-known financial institution defaults on its agreements. competitive policy. It is thus an important contributor to our national GDP. While policy makers might view bankruptcy as a workable option for auto companies. resulted in a literal “run on the banks” that threatened Goldman Sachs. He has published extensively on the topics of franchising and vertical organization. leading to widespread lack of confidence in the banking system and even financial panic. of course.104 III. along with the near bankruptcy of AIG. the auto industry is very important. who is the Gleason Professor of Business Administration at the Simon School. It employs roughly two million people in manufacturing and in sales and service jobs. Consumers depend on banks to provide a relatively risk-free place to hold their savings—not to mention their mortgage and auto loans. and just about every major financial institution in the US As I already suggested. their major . Aren’t both industries important to the economy. and other financial services. to come back to where I started. I think that these issues of corporate strategy and structure are likely to be addressed more effectively through bankruptcy than bailouts. The bankruptcy of Lehman Brothers gave policymakers a frightening glimpse of the potential for a large domino effect when a big. Chapter 11 could well help the auto industry address some of its most pressing problems. 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. and the banks became increasingly reluctant to lend to consumers. and they are problems that unfortunately run deeper than the current economic recession. As Tom Jackson was just suggesting. But the American auto companies also. The Difference between Automakers and Banks Zupan: Thanks. That event. or dealer networks. Now. Given the importance of this industry. The Big Three now market 112 different car and truck models in the US through 15 distinct brands. given their current market shares. tend to focus on unions. Let me start by saying that the auto industry is clearly very important to the US economy. insurance. Charlie. having been a past recipient of our highest teaching award. to the business community. one of the co-authors of the leading textbook on organizational architecture. 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. corporate governance. He’s also. Jim is also a highly regarded teacher on our campus. In contrast. and even to one another. along with Cliff Smith. just as people can’t survive if their hearts fail and blood doesn’t get to vital organs. But let’s take a closer look at the challenges now facing the auto industry. Jim Brickley: Thanks. Now let’s hear from Jim Brickley. models. It is widely acknowledged that these companies now have far too many brands. But before we get into the case of the auto companies.

franchise laws are but one example of how political Now. just and very expensive.JACKSON. as owning their own system in an economy is like the Tom said earlier. the case of large banks—one that would fewer. Toyota. as Tom also said. Jim. A Bankruptcy Practitioner’s coordinated changes in these companies’ product lines and Perspective on Chapter 11 dealership systems. Of course. GM currently has some 6. What is needed instead are systematic and IV. and manufacturers from economy. a number of attempts by the Big Three to introduce new brands and models than the Japanese companies combined. HUGHES. it reportedly cost GM over $1 billion to settle flexibility to design efficient distribution systems. But relying on local business failures “car czars. As a result. state laws preventing auto models. In GM by itself has eight brands and 70 models. For supersede state laws and grant the auto companies more example. Now let’s hear from Joel Tabas. to dollars a year in the form accomplish outside of of higher auto prices. 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.200 dealers with distribution systems and just 1. 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. disputes with dealers when they stopped making Oldsmobiles And let me leave you with one final thought: Inefficient a few years ago. proceedings are much more likely to focus on economic some restructuring is going on as we speak. this is dealerships has cost US going to be difficult. marketing channels have been blocked by dealer-initiated And as Tom also told you. and thus more fact. Its closest indicates that such laws competitor. they have opposition from the local -Jim Brickley secured protective car dealers. as people can’t survive if their hearts fail bankruptcy. 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. BRICKLEY.700 dealers that I have studied the effects of franchise and dealer protection operate 14. 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. is a very protracted and costly way of addressing the basic problem. in the case of bankruptcy. 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. these laws in the face of Over time. State laws not only make it expensive to alter dealership Zupan: Thanks. the destruction of companies. local communities. all of the considerations often trump economics in legislative or company’s dealer contracts become subject to cancellation regulatory solutions.” In the long run. Auto How do we address this and blood doesn’t get to vital organs. Restructuring and consolidating the and reworking. a contracts. 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. 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. But. many dealers. the use of a similar process in and thus nearly 90% corporate values.600 franchises. TABAS. The banking and financial dealerships. they also prevent manufacturers from owning their . to consumers through other media such as the Internet. the Big Three also have far too lawsuits or regulatory actions. has lead to less efficient bankruptcy as a workable option for auto only 1. Bankruptcy dealership systems in a quick and efficient way. with 58 models and seven brands. My research While policy makers might view brands. the auto Consistent with these companies have all findings.000 laws across a broad range franchises for its eight of industries. consumers billions of circulatory system in a human being. Now.

there is some potential for conflict here. If you’re helping a debtor negotiate with creditors in a distressed situation. I have represented both debtors and creditors in the reorganization process. of the tendency of many companies to wait until it is too late to rehabilitate the business. and restaurants—and participated on creditors’ committees in complex reorganization cases involving such names as Planet Hollywood. and The Discovery Zone. like President Jackson. I would argue that the 90% failure rate is in large part the result of inadequate pre-bankruptcy planning. As part of his practice. in Chapter 11 it would be handling the litigation involving all of those franchisees in the one forum where the bankruptcy is filed. Chapter 11 could be very effective in getting concessions from not just creditors. Soloff and Miller. Cliff Smith was my finance professor when I was in the MBA program in the late ’70s. a Miami-based law firm that specializes in reorganization and bankruptcy. That’s the bad news about bankruptcy—but there is some good news here as well. In this sense. unsecured creditors—that resemble the obligations of corporate directors to the company’s shareholders. Joel has graciously agreed to join us tonight in the midst of what are pretty busy times for his business. So. I too was struck by the stage backdrop behind us. to hire professionals—accountants and other financial types as well as accountants—to help them make the managerial decisions that have to be made. such claims tend to be reduced significantly in Chapter 11. As Tom mentioned earlier. He has found himself operating airlines. Brothers Gourmet Coffee. it’s very important to understand what can be accomplished in Chapter 11. and we were invited to become part of the committee of unsecured creditors. outof-court workout process would be a nightmare—the legal fees and expenses would be astronomical. After all. Avoiding this possibility is likely to mean some cost savings for the franchisees. and for a premature liquidation of the . 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. Just the prospect of Chapter 11. 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. that is an extraordinary benefit that bankruptcy brings to a situation like this. when evaluating any kind of distressed corporate situation and the range of possible solutions. Mark. all the similarly situated. Now. with its “automatic stay” provision and the potential rejection of “executor” contracts. In Miami. by the way. we’re dealing with an incredibly distressed real estate market—and this Depression-era stage set looks very familiar. 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. 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. makes me feel right at home. as Tom also said. They have the right. they think of it as a death sentence for companies. and are often converted to equity interests. It focuses the efforts and avoids the potential for inconsistent consequences. Most people have an instinctive aversion to the word “bankruptcy”. That was how I got involved in the Planet Hollywood case that Dean Zupan mentioned. So. 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. One of the main distinguishing features of such successful reorganizations is planning and preparation. Tom Jackson’s classic article on reform of the US bankruptcy system was required reading when I went to law school. And let me start by saying that it’s a great honor to be taking part in this discussion. a private. Freedman. real estate reorganizations. the high rate of failure is not really attributable to the Chapter 11 process itself. In the case of the auto industry. When dealing with large numbers of creditors that are dispersed around the country and have the option of seeking different venues and courts. As Tom started out by saying. 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. In other words. is very helpful in getting concessions from lenders and other major claimholders. or to benefit their clients at the expense of other claimants. they are not supposed to be using the platform for personal gain. 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. instead of General Motors facing litigation throughout the country on franchise disputes. They have fiduciary obligations to their constituents—namely. They’re supposed to be trying to maximize the recovery of all the creditors.106 Simon alum from the class of 1980 and the managing partner of Tabas. My client was a creditor. 10% of the companies that file Chapter 11 do emerge as independent viable enterprises. Joel Tabas: Thank you. and are given the resources. but from the franchisees or dealers and the unions as well. instead they are sold to outside investors or end up liquidating in a Chapter 7 or similar proceeding. retailers. and healthcare workouts and bankruptcies. 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. 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. Joel has dealt with Ponzi schemes. And.

So. We’ve been involved with a few Ford franchisees in the Miami area that have recently filed bankruptcy and shut down. it could deal with its franchisees’ claims in one forum—and everyone could be treated the same. who is beholden to all the various constituents of the enterprise. including information about their plans to restructure and rehabilitate the debtors. 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. all of the top executives basically have to submit their compensation packages for approval by the court and vetting by the creditors. 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. In fact it’s more than likely that. most states have passed laws that make it very difficult and expensive for the manufacturers to shut down their franchisees. by the way. My guess is that. 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. let’s come back to this issue of franchises that everybody has identified as a big problem for the auto makers. the franchisees will become significant equity holders in the auto makers—and if this happens. BRICKLEY. 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. But even so.JACKSON. and how the company will be financed. Earlier in this decade. as Tom pointed out earlier. 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. But that practice has now been largely ended by the courts. 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. but for the creditors as well— because of their coordinated representation by the committees I mentioned. they’ll actually have a stake in the health of the underlying business. For one thing. Now. One obstacle is holdouts among creditors to a negotiated solution—and the Chapter 11 can be used to “cram down” such a solution. I’m convinced that such people are far better able to help fashion how the company will go forward than the typical regulator. HUGHES. And let me come back to the point about the creditors committees that I made earlier. Early on in bankruptcy proceedings. As has already been noted. debtors are greatly aided by the automatic stay provision I mentioned earlier. Another valuable aspect of bankruptcy is its ability to increase disclosure and transparency. While the process can sometimes . you will be dealing with jurists who handle reorganizations and feasibility determinations on a regular basis. Another obstacle is entrenched managers or owners. executive compensation is typically submitted to courts for approval. especially in a case like GM or Chrysler. or the government in its role as The United States Trustee—the right to obtain financial information from the debtors. But. this brings all compensation arrangements out into the daylight. By putting a halt to all the disputes and lawsuits. The same comment also holds. I would also argue that. Moreover. 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. an equity holder. and there are likely to be significant damages to the manufacturers associated shutting down franchises. there would be significant cost savings not only on the debtor’s side. 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. But professional fees also have to be submitted on a periodic basis for approval with the courts as well. thanks to years of litigation in high profile cases involving many of the complex issues now facing our auto makers. TABAS. 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. 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. in determining the company’s future. it provides a very effective way of eliminating obstacles to private workouts. In bankruptcy courts. I think that there are certain aspects of private workouts that have not been incorporated into the analysis. 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. 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. we used to see people filing for compensation packages with golden parachutes. AND SMITH — BANKRUPTCY AND BAILOUTS 107 business. at the end of any successful reorganization process. As already mentioned. if a manufacturer files bankruptcy. There could even be a committee for the franchisees so that they too could have an economic voice about the firm’s future. 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.

offer under these This feature is likely circumstances. I appreciate what GeVa has done for provide the super-priority financing. So. 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. 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. could be used to help US auto makers use this wonderful facility. an accomplished -Joel Tabas the Big Three— scholar. along with the it could go to the court and say. It’s become an old saying that people who do not study to work their way out from under their current burdens. Cliff is. And. which is if one of the auto makers were instead to file for Chapter 11. 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. who is the go forward than the typical regulator. But editors of the Journal of Financial Economics. They will say. I’m for the US auto Zupan: Thanks.” 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. 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. and the companies end up in liquidation. people these companies do not consider bankruptcy. have fiduciary duties that are supposed to shift from the it sounds good. let me mention one other important viable—though likely much smaller—going concerns. the directors could be facing in the US tend to point to Chrysler as an example of a success director and officer suits. That appears to be the case going to be converted into equity. Thank I’m a believer in having people with the to be most valuable. markets are otherwise Run) Costs of unwilling to provide especially if a big portion of their claims is Bailouts new capital. capital structure. to be giving money to all. “Chrysler got their act together and now. who expansion of the Louise and Henry bailout money already Epstein Professor of is beholden to all the various constituents of provided. 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.” directors if you say that fast enough. It’s good to be here. I want to thank them for letting us code that. which is why Joel. process that leads up to confirmation of a plan generally tends In sum. they could be facing a D&O suit. in cases economic interests involved in the key where the capital V. and don’t think about it very hard. our Executive MBA Teaching Award an amazing 19 times! Cliff Smith: Thanks. advantage of Chapter 11—a feature designed to help debtors Bankruptcy. He has long they’re unable to raise been one of the main equity or debt—and so they’re going to the government. Let’s just do it again?” Now. there are a number of features of the US bankruptcy the local arts community. for all its flaws and bad press. as President Jackson suggested. headquartered at the Simon School and. they should be weighing all history are doomed to repeat mistakes that have already been their options very carefully. But it’s important to remember that Chrysler shareholders to the creditors. of course. I’ve found that the adversarial value of their companies. Batting cleanup the government is tonight on our panel is able to help fashion how the company will contemplating an Cliff Smith. first of the enterprise. may have a lot to raise new capital. The capital Finance at the Simon markets are not going School. 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. 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. The (Longdecisions about the future of the business. which is a fertile area of law right story. “To raise new capital. What those suits are alleging is that is when a company things worked out wonderfully. One reason they should be made. Mark. As a If that fails. through their problems and preserving their companies as Before I close. the other option would be to have the government long-time subscriber. enters what is known as “the zone of insolvency. convinced that such people are far better makers. in my view. I need Journal of Finance. .108 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 get a bit heated and hostile. you.

” and “innovative. We are going to see lots of outsized bets being funded not by private investors. You would have seen many leaving that industry. and innovative. or maybe three—and US farmers would have had a tough row to hoe. our government turned what would have likely been relatively modest losses into much larger ones. If you’re ever invited to a poker game and allowed to play with someone else’s money. If the government did nothing. It was those transactions that ended up doing most of the damage. when interest rates on Treasury notes and bonds got into double digits. despite the best efforts of the Resolution Trust Corporation ten years later. and perhaps even less to lose. fund them with CDs paying 12%. My point here. But after a few years. People with college degrees and more opportunities in other industries would be more likely to go. So. Think about the history of US agriculture since World War II. So the resulting warranty claims didn’t eat us out of house and home. and we would have been back in normal operation. when companies use words like “bold. 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. After Chrysler’s debt was guaranteed by us. What was to be done? One option was to do nothing. for the next few years. Thus. then.” What happened next? Well. Since most of these S&Ls were holding mainly long-term fixedrate mortgages with rates around 5-8%.E.” I would recommend a fair amount of skepticism because those costs are regularly understated by what can turn into scary amounts. As the political process is unfolding and people are saying. we had lots of “zombie” S&Ls—they were dead. US taxpayers ended up footing a bill that has been estimated at about $130 billion. 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.” It is like flipping a coin where heads I win tails you lose. the cost that we’re forecasting for this bailout is X dollars. 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. and expect to make it up on volume. AND SMITH — BANKRUPTCY AND BAILOUTS 109 Remember the US savings and loan industry and what happened to it? In the early ’80s. new. in a sense. So you would have seen younger farmers leaving while older farmers stayed. My dad was a banker in Greensboro. we’re giving the company to the Treasury. we’re heroes. “Well. as things turned out. the wrenching adjustments would have been behind us. to play poker with someone else’s money. The net result of this regulatory “forbearance” was that. executives from many S&Ls went to Congress looking for help. but were still walking around underwriting risky mortgages and investing in risky commercial real estate. the people with the most opportunities other places. My third point is that the forecasted duration of this bailout is something that can easily expand. product warranties in the industry covered 12 months or 12. HUGHES. But think about this from Chrysler’s perspective. Who would have been most likely to leave? Well. though with . Chrysler management decided to expand Chrysler warranties to five years or 50. those with the most flexibility. I’ve got a piece of advice: increase your bets. the European battlefields back into wheat fields.” “risky. most European wheat fields were turned into battlefields.” “risky. If you go back and look at accounts in AutoWeek of Chrysler’s post-bailout success story. agricultural prices would likely have remained low for two years. Before Chrysler got its bailout package in the ’70s. but bailouts can have the effect of increasing risk within the system.” And they did a marvelous job. In response. TABAS. “You can’t write 8%. As a finance professor. Bringing out risky new products is one way automakers can do it—but there are others. economically speaking. Day? The swords were turned into plowshares.” And that leads to an interesting problem for regulators—and of course the rest of us as taxpayers. 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. you will see articles in the late ’70s and early ’80s about Chrysler’s bold. If it doesn’t work.000 miles. what seems to have largely vanished from the collective memory is any sense of the eventual cost of that initial act of forbearance. innovative models. my second history lesson is that bailouts allow companies to play poker with the taxpayers’ money.JACKSON. Now. Georgia in those days.000 miles. During the War. “We’re going to try something that is bold. BRICKLEY. those bold new products generally were well-received and well-produced. If it works. By failing to deal with the troubled S&Ls effectively in the early ’80s. 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. Roosevelt granted draft deferments to US farmers along with instructions to “crank up production and feed the Allies. This huge increase in supply and crash in prices put the US at a political crossroads with respect to its agriculture industry. new. the first lesson from history is that bailouts are a risky business—and not only is the outcome uncertain. and he liked to tell people.” what I hear is “risky. and the US auto industry is clearly worth more than that. they were effectively insolvent. So. 30-year mortgages. They are being given the opportunity.” “new. the taxpayers. But what happened after V. is that although the S&L bailout is today widely viewed as having been a good thing.

(And the same. Think about Fannie Mae and Freddie Mac—not to mention the US Postal Service.110 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 far fewer people working in the industry. Thus. government poured likely never have reached their current levels. Nancy Pelosi is talking about forcing them to start making “green” cars—and she’s not the audience. the U. bailouts can persist—sometimes for decades. That pieces being sold to happened in the ’40s foreign auto makers. Morris. would likely have been true if the government had government engineered a bailout of British Leyland. on that one? Jackson: Any time you’re looking at a large financial Unfortunately. Thus economy. 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. In the case of Lehman be especially good at running businesses. and decided to pay our eventually went out of farmers not to business. I’ll now invite the other panelists constraints on what they can do. 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.) about $16. In the 1970s. We British Leyland into their business models and strategies. it had suffered a because the industry. it’s always been a very large industry. Mini. 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). 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. ’80s. the government’s track record in running institution. perhaps of whatever policy you need to think choices get made— again. by has provided us with an example. Bailout advocates in Congress regularly announce. the way. there are many more linkages with the rest of the businesses is not the best. going to do once and long-run consequences -Cliff Smith be done with. an incredibly important large portfolio of options. Ford and Chrysler. I think our regulators learned a lesson from that failure. Rover. that we are discussing ’70s. Yet this same logic it go. magnitude of problems Detroit is forced to deal with— and inefficient manufacturing: moreover. we would not now be facing the Leyland had a weak balance sheet. If you believe US auto industry. MG. with select putting their own capital on the line to back produce. and Jaguar. tough choice back then. and how many of our taxpayer dollars we are perpetual annuity. We’re going to put constraints to join us on the stage. “We’re VI. and the overcapacity problem. The U. In that this bailout of the making our policy dramatically more valuable than an option auto industry is choices. Cliff. As any finance We’re still doing it set of problems for the professor will you. We were told earlier that we’ve never had a bankruptcy Brothers. a portfolio of options is today. and we will take some questions from on how they can pay people. this begins to have been forced to go bankrupt?” Tom. And we all it ultimately failed to know how that one save the company: number of people putting their own intuition turned out. the not bailed out Chrysler in the early ’80s. To me. Bank Bailouts and the Credit Crunch not planning on just handing suitcases full of money to General Motors.5 billion (in current dollars) into the company The other choice facing US policymakers back then was during the ’70s and ’80s. we need to something that we’re think carefully about the on a single portfolio. their bets. can you start us off sound like allowing the government to run the industry. contentious labor relations. and things are much more complicated. would substantial loss in market share. But if we look overseas. Here’s the first one: “Should Lehman Brothers talking about her favorite paint color. my fourth history lesson is that the government is unlikely to Commercial banks can’t use bankruptcy.K. We’re going to put Zupan: Thanks. Had we made the maker of Austin. and ’90s. 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. history they quickly saw there would be huge problems unwinding .K. they need to go through some other regulatory process. the third history lesson is that willing to use to see if we can make it happen.

and what almost always happens during these kinds of financial dislocations. Jackson: I think that cleaning up the banks’ balance sheets is a necessary but not a sufficient step in dealing with our present problems. while I think it was good that the government pumped in some cash and kept other institutions from collapsing. Hughes: I’d like to jump in here. 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. Things look awfully murky out there. and the amount of new loans—particularly real estate loans—are down as much 90% in some cases. BRICKLEY. 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. Now I think it has overshot on the way down. in thinking about this question. I think there are two main ways out of this problem: the Japan model and the Swedish model. this is a very complicated and multi-faceted problem— and cleaning up the balance sheets is. I think it’s important to start with an understanding of what banks have a comparative advantage in doing. that’s especially been true of the smaller regional banks—I think it makes more sense to recapitalize those banks . That’s why I’m frankly skeptical about the government’s plan to buy troubled assets. And that suggests that this idea of cleaning up bank balance sheets so they can start trusting each other has some important limits. So the curtailing of access to credit has been most pronounced for businesses with weaker credits. So it hasn’t been a big surprise to me that the bailouts have failed to produce an immediate increase in bank lending. too. you’re going to have trouble persuading a commercial bank to make you a loan. But other things have to happen too. Smith: Well. 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. But market participants tend to overreact—and in some cases perhaps bank regulators. as I said. if you’re a start-up company with little in the way of tangible assets and not much of a track record. Banks. we went back to a model of stepping in. a necessary part of the process of getting credit flowing again. it’s very hard for them to make new loans. AND SMITH — BANKRUPTCY AND BAILOUTS 111 all the claims. creating a downward spiral. The Japanese approach was to accommodate the banks. on the other hand. HUGHES. even if publicly traded. as a result. Tabas: I represent some local banks in Miami. 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. Even if you clean up their balance sheets. a well-known appraiser in Miami. is that credit spreads have risen dramatically. Our real estate market clearly overshot on the way up ’04 and ’05. That’s going to take time.” Sweden came back pretty quickly while Japan was in a recession for over ten years. In cases where insiders have an advantage over outsiders in valuing bank-originated assets— and as I say.JACKSON. It’s like a game of liar’s poker. But what has happened. Thus. since I think we’re avoiding the biggest issue with the banks—namely. If you are a fairly large business with a good track record of producing earnings and cash flow. Financial institutions—and particularly smaller banks—are by their nature somewhat opaque institutions that hold many assets that are difficult for outsiders to value. their unwillingness to lend to each other because they don’t trust each others’ balance sheets. Now the real problem here is all the uncertainty about how long it is going to take before the economy recoveries and. and how much debt it is really capable of supporting. Wall Street will take your debt to public markets if you’ve got a triple A credit rating. more opaque enterprises that have always been their bread and butter. it’s these kinds of smaller. The Swedes said. For regional and community banks in particular. they’re having a tough time getting people to step up and make appraisals on the properties. I think we’re avoiding the big issue. your first choice will typically be to go to Wall Street and have them package your debt as a public issue. You can’t have a banking system where institutions can’t trust each others’ balance sheets. So. TABAS. As Joel said. to allow them to continue to operate and make more loans while cleaning up their balance sheets very gradually. tend to finance smaller companies that. 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. One of my best friends. So. have substantially less information produced about them. Because of this situation. 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. “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. then we’ve only added to the existing troubles. The result is that right now people in Miami are not able to borrow money for real estate from banks. And I suspect it was probably the right thing to do under the circumstances. or an investment bank to help you raise debt capital. about how much collateral lies behind the business. 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.

M. But. in the service sector rather than manufacturing. if I can be a little patriotic. “We should be supporting cars that are built somewhere else over cars that are built here. Some of G. in the caveman era. Smith: And to add to what Jim’s just said. Jackson: I don’t think this question of domestic versus foreign production is nearly as simple as you make it out. the companies would be in much more trouble than they are now. 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. such as Holden in Australia. VIII. I think it make sense to view our entire financial system as engaged in a kind of Darwinian process of trial and error. Now. I find it bizarre to say. I think that using taxpayer dollars to subsidize that overproduction is a terrible idea. I’m not sure it even makes sense to talk about a US-produced car. 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. I agree. My point is that. I think they were pretty smart and decisive in doing that. 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. making the same mistakes over and over again. but before the credit crisis began to set in. Again. and that we have to figure out some way to take capacity out of the system. Now. I find it very odd that we would be having any conversation where people say. 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. So distinguishing between US versus foreign production is not straightforward. This way. When you hear it. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 VII. both General Motors and Ford went through massive restructurings that took out almost half of their production capacity. That’s a matter of getting the incentives right inside organizations—something that I believe is incredibly important. But we still import a huge number of cars. I would guess. One of the strengths of capitalism is that it tends to prevent people from persisting in error. Brickley: Greed is a pretty loaded term. The Role of Greed Zupan: Another question: “It seems that all the problems we’re currently dealing with can ultimately be traced to greed. Had they not done that. When the physicists figure out how to repeal the law of gravity. As Tom said. 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. say. they can become the efficient producers that we want. the Big Three have already made huge efforts to take out excess capacity. we would be the only country in the world to take that approach. 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. we make mistakes—and then we learn from our mistakes and make adjustments. in the past few years. it’s important to keep in mind what another guy named Smith—not an economics professor. Since there are lots of American investors who own shares in Toyota.000 a shot because of contracts with the UAW. not all of that capacity is sitting in the United States. When will we learn how to deal with this? Smith: I’ll tell you when. as the environment becomes more complex. one important lesson underscored by recent experience is that problems are going to arise whenever individuals and companies are granted a lot of “free .” I’m not talking about putting tariffs on imports. Honda. “We’ve got three million units worth of excess capacity. all the new products and services that are the real source of prosperity.’s most efficient operations are manufacturers in other countries.” If we were to do this. They were forced to buy out thousands of workers at $140. Hughes: That’s all true. We will no doubt make mistakes in the future. We keep trying different things. and although we may well have three million units of excess capacity in the US. 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. Toyota. the economists will be right behind them repealing the law of demand and abolishing greed. We will continue to have boom and bust cycles of the kind we’re now going through. let’s take it out of the US producers.112 with infusions of equity than to buy individual assets. and given some time. but a Professor of Moral Philosophy—told us over 200 years ago. something like 6070% of the Toyotas that are sold in this country are also assembled in this country. one question we are asking is whether people are any more self-interested now than they were. It drives innovation. You couldn’t have done these things 20 years ago because the financial instruments just weren’t available. 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. it’s no longer even clear what it means to be a foreign company. Obviously a lot of the foreign companies have now built US plants that employ US workers. most of them. Brickley: To expand on Tom’s point. But.

General Motors’ need to deal with 14. 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. It’s a fairly wellestablished principle in political science that these kinds of “collective action” problems are generally likely to be intractable. Solving the Dealer Problem Zupan: Ok. But the way things are now. Hughes: I want to jump in here. On the other hand. I think both of these prohibitions are sources of inefficiency that increase the cost of automobiles—and. To have a chance of becoming a competitive producer. you’re the expert on franchising. One has to do with the states. So I think that the federal government can address some of the restrictions on the auto makers’ dealings with their dealers. 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. and here it is: “Instead of relying on bankruptcy. There are more and very urgent problems that cannot be handled through legislation. GM must renegotiate these contracts. 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. people are going to be fighting over pieces of the pie instead of trying to preserve the overall operating value of the firm. I think it would be great if we could remove some of these inefficiencies through legislation—and without resorting to Chapter 11. HUGHES. the rejection of executor contracts in bankruptcy suggests that Chapter 11 is the ready-made solution to these franchise problems. I agree that we should have . pushed by Republicans and Democrats alike. But getting political action on this is likely to be difficult. why don’t you take this one? Brickley: Well. across-the-board solution would be preferred if possible. That’s a clear prescription for too many mortgages and too many securitized deals. 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.JACKSON. each bearing a relatively small cost and having little interest in the issue.000 franchise contracts—is one that I don’t think can be addressed effectively by legislative action. That problem keeps a lot of politicians from forgetting about their commitment to the public good. in my view. But. I see two different issues here. Smith: Let me add to Tom’s point. The second issue raised by the dealers—by. 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. It was government policy. AND SMITH — BANKRUPTCY AND BAILOUTS 113 options”—that is. I agree with the premise of the question that a legislated. It’s this uncertainty about the political process that makes me think that bankruptcy is the right way to go. but it wasn’t just opportunistic or greedy lenders and homeowners at work here. 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. As I said earlier. almost all of which have these laws that make it difficult for the auto companies to operate efficiently. 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. whenever they can acquire assets or do deals without putting any of their own capital at risk.” Hughes: That’s all true. I think that at that point you can say that the driving force was greed. When people and institutions respond in predictable ways to those policy initiatives. If they try to accomplish this outside Chapter 11. we have time for one more question. the people hurt by these laws—namely. and ending up with leverage ratios of 40 to one—you have to ask how that all came about. there’s no doubt that such changes would be blocked at the state level. I don’t think we can get it done. Jackson: True. So this is the collective action problem at work. this renegotiation is going to be very difficult outside of bankruptcy. TABAS. as Tom said before. I don’t know many bankers that are comfortable with the idea of operating with that kind of leverage. I’m not sure we learn much from identifying the source of such behavior as “greed. I think it’s important for the government to back these contracts. I agree that we probably don’t have the will to make a lot of changes that we should. In talking about the dealers. At the same time. government policy clearly had a hand in producing the housing and mortgage bubble. But given the realities of the political process. anybody who ever bought a car— are a widely dispersed group of individuals. But whether they could be accomplished at the federal level is also highly questionable. IX. 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. say. Jackson: Like Jim. they should be overridden by federal legislation. So. BRICKLEY. 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. 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. the profits from the dealerships. As Jim mentioned.

you would not be putting them out of business. Brickley: Most companies use a mix of both arrangements—say. Inc. but at one point in the past.2 billion to $2 billion. Smith: From the manufacturer? Hughes: Not from the manufacturer. there should be a contract that says.114 a federal franchise law. Brickley: Well. it hasn’t made any money in the last dozen years. it sure sounds sweet to be able to do that. Even Toyotas sometimes have to go back. In fact. by tying your hands behind your back and saying you can’t try something. let me tell you a bit more about what the dealers actually do. I agree with Cliff that. 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. if it was merely a matter of General Motors going out and saying. the industry has a network of dealers that in most instances has been willing to give the cars away for almost nothing. publications like Business Week and Fortune and the Harvard Business Review were all talking about how Japan. and the probability of repeat business. although excellent dealers. It does seem to work. is that there is no longer any Blue Sky in the Saturn franchise. and pay you all the money that we owe in accordance with our contracts.” then we wouldn’t be talking about anything like $1. If you look at unregulated or less regulated distributor relationships in other industries. Since they also own a lot of other franchises. 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. or is willing to reveal. One thing we know is that. Now. There are few things more common than believing you can do something as well as somebody else. But the state governments have taken that option away from the auto companies. however. but is there to service and trade them and help buyers sort out their finances in a way that manufacturer cannot do. Hughes: That isn’t the real obstacle. they would then have to contract with some of their dealers to deliver them to the buyers. it’s probably not the last time you have to go into a dealership. let me give you an example of something Ford tried and then got blocked by regulation. 80% dealerships and 20% dealer-owned stores—depending on variables such as location.” The problem. Smith: I wasn’t making that assumption. That subject’s got Rick Wagoner so afraid he won’t even touch it anymore. So if we did pass a federal law—though I realize it’s unlikely to happen. There are now some 440 Saturn dealers that.” We talked about example of GM’s shutting down Oldsmobile earlier. “This is what we’re going to pay you. Hughes: Right. by the way. or the company wants to stop doing business with a certain dealer. Brickley: Well. Ford had a bunch of used cars that they wanted to be able to market directly to buyers over the Internet. I don’t think bankruptcy. they succeeded in passing legislation that prevented the automakers from owning dealerships. No one knows. Smith: You mean the manufacturers needed a law to protect them from themselves? Hughes: That’s basically right. behind the scenes. you will never know what might have worked best. But the dealers are asking for a lot more than that—they want “Blue Sky. Let me mention one other interesting piece of auto industry history. at the moment. Hughes: Well. But the dealers are still asking for it—and that’s where the problem becomes intractable. The idea was that if they sold the cars. The interesting thing here is that. There’s no question there are some hidden costs and inefficiencies in the system.” you take away that opportunity to learn something you didn’t know. I’m assuming that allowing people to experiment with a different model is something that has a lot of value. are not making any money. would be the solution to this problem—though when you’ve called on and negotiated with as many dealers as I have. When the dealers were getting their way with state governments. But this experiment never got off the ground. when you buy a car. “We will buy back the parts and tools. Smith: For those of you here who are old enough to remember. But there are other issues that also need to be recognized and addressed. I think that’s a mistake. and maybe as high as $2 billion. was . “You can’t turn down that street. If a dealer goes out of business. And. People do buy cars over the Internet every day. let me weigh in on this one. 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. In Texas around the year 2000. They should be put to rest. 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. the actual costs of ending relationships with the dealers— but in that case it was reportedly over $1 billion. this all reminds me of those discussions back in the ’80s about Japanese industrial policy. In those days. 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. So there’s a whole array of services in a car transaction that go beyond just buying a car. That by putting a regulatory stop sign at the intersection that says. like a lot of other things we talked about tonight—we could solve that problem. the manufacturers once had the right to their own car dealerships.

My point is more narrow: The problem here is that we will never know. we’re not going to let you see if that would work or not. That’s something the US economy has been pretty good at—cutting its losses when necessary and moving on to something more promising. let’s leave it at that—and let me thank all of the panelists for taking part in an instructive and entertaining discussion. 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. BRICKLEY. I’d much rather have the American business community continue to make thousands of calculated bets. 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. 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. Now. Zupan: Well. I just want to say that stopping that kind of experimentation is not without costs. 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. it may well have blown up in their face. putting their money where their mouth is. “As a regulatory matter. HUGHES. the problem with these dealer laws we’re talking about is that they absolutely prevent certain kinds of experimentation. TABAS.JACKSON. You are legally prohibited from trying certain business models and practices. When those bets turned out well. AND SMITH — BANKRUPTCY AND BAILOUTS 115 competing the US right off the map. than having somebody in Washington or Albany say. I’m not arguing that if Ford had been allowed to sell cars directly on the Internet. 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. As any finance professor will you. . 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. a portfolio of options is dramatically more valuable than an option on a single portfolio.” What you wind up with when you allow that kind of experimentation is a very large portfolio of options. But a lot of them crash and burn—and you rarely hear about them. Some of these bets turn out wonderfully—take Google for example. In fact. but cut your losses and move on when you’re failing. But that approach seems to have lost its magic in the last two decades. it would have been a multibillion dollar product line for them.

Furthermore. Sherman Fairchild Distinguished Scholar at the California Institute of Technology.116 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Pioneers of Finance An Interview with Vernon L. especially concerning human judgment and decision-making under uncertainty”. he has been a Ford Foundation Fellow. Professor Smith is a distinguished fellow of the American Economic Association. finance. Economic Design. and the Journal of Economic Methodology. Brown University. Fellow of the American Association for the Advancement of Science. Dr.1 Vernon Smith is widely regarded as the “father of experimental economics” for his pathbreaking work in this area. especially in the study of alternative market mechanisms” – 2002 Nobel  On January 9th 2009. the Western Economic Association and the Association for Private Enterprise Education. the 1995 recipient of the Adam Smith award. the University of Arizona. He serves or has served on the board of editors of the American Economic Review. he is past president of the Public Choice Society. currently serving as Professor of Economics and Law at Chapman University School of Law. He has previously held faculty positions at George Mason University. and the University of Massachusetts. After decades of research. Fellow of the Econometric Society. Journal of Economic Behavior and Organization. 3 116 . especially in the study of alternative market mechanisms”. In addition. 2 Daniel Kahneman was the other co-recipient of the 2002 Nobel Prize “for having integrated insights from psychological research into economic science. and Fellow of the American Academy of Arts and Sciences. Science. Economic Theory.2 He has written or cowritten more than 200 articles and books on capital theory. the Journal of Risk and Uncertainty. the Economic Science Association. Smith remains very active in the economics profession. and an elected member of the National Academy of Sciences. Terry Odean and Betty Simkins interviewed Vernon L.3 1 A video of the interview is available at the Journal of Applied Finance website. Games. and Economic Behavior. Purdue University. and natural resource economics. to the field of behavioral financial economics. experimental economics. Vernon Smith was a co-recipient for the Nobel Prize in Economics “for having established laboratory experiments as a tool in empirical economic analysis. The Cato Journal. Smith: 2002 Nobel Laureate in Economic Sciences and Father of Experimental Economics Terrance Odean and Betty J. 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. Smith for this issue of the Journal of Applied Finance. Simkins “For having established laboratory experiments as a tool in empirical economic analysis. In 2002. Fellow of the Center for Advanced Study in the Behavioral Sciences.

5 Some people tried to come up with elaborate theories of what was going on. they would miss out on all of the payments. With the Federal Communication Commission (FCC) auctions. you have the danger of paying more than you need to. experimental and behavioral economics are actually complimentary. 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. the increments were $5 and $6 million. and related issues. never bid again yourself – that is. If you go back before Danny Kahneman. 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. decision making under uncertainty.” Even though I have interacted with Danny Kahneman over the years. They were first focusing on decision making in markets and market exchange situations and the work was primarily interested in the performance of markets. What lay behind it ought to be an important part of the picture. I see behavioral economics as having evolved out of the early cognitive psychology work with the emphasis being on fundamental decision making. SMITH INTERVIEW 117 In this interview. It doesn’t bluff anyone out.” But as far as we can see. Danny is more utilitarian than I am and that sounds odd – because we economists ought to be utilitarians. Most experimental economists that came in the 1960s and 1970s were not primarily focusing on individual decision making under uncertainty. You’re not suppose to jump bid. This refers to FCC auctions of the electromagnetic spectrum.ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. Vernon Smith shares his insights on markets. 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. Among the issues he addresses are: * the relation between experimental economics and behavioral economics. They’ll tell you: “My desire to bid was. by jumping the bid. 4 5 A jump bid is a bid higher than necessary to reach the next bidding level. who were early psychologists but were not part of the cognitive psychology development.fcc. You should only bid by the minimum increment. Can you tell us about these early findings? Smith: In my early work. How are experimental economics and behavioral economics related? Experimental and behavioral economics are actually complimentary. don’t raise your own bid. * the insights his research on speculative bubbles in experimental markets provides for understanding the recent bubble is US residential real estate. A person behaving rationally in the English auction should always raise the bid. If you bid more than that. Now in English: This gives you outcomes that are completely efficient. I would advertise — They (other traders) had better get out because I was going to win. Vernon Smith: Well. There is no way this is going to work. It doesn’t work. People don’t follow these rules. if the starting bid is less than the value. Small stakes. the predictions of behaviors in different auction formats. that is completely ineffective. if you had multiple units. The idea was “Let’s just do experiments with it (the human mind) and just see what comes out of it. They were seen as the “Skinner behaviorists” – that view the human mind as a kind of black box. 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. They have various rationalizations about it. wasn’t holding up. In particular. They thought there is some sort of signaling going on. we’ve always been interested in different questions and issues. See: wireless. you find people like Sidney Siegel Ward Edwards. You can’t believe how many experiments we did to understand that.4 It was that sequential nature of the bidding that led to the problem. we began to see and realize in experiments that what we got from utilitarian and formal analysis. people would get into what they called “jump bidding” and in that jump bidding process. and * his view as to reasons for the dramatic rise and fall in oil prices last year.gov/auctions/ . It is also characteristic of some of the work in experimental economics.

You can only describe the equilibrium with simultaneous nonlinear equations. Because right from the beginning. There are a lot of stories about bubbles in history going back to the South Sea Bubble and the Tulip Bubble. even though by definition of an equilibrium. you’ve studied experimental asset pricing markets that lead to speculative bubbles. Given my traditional economic training. That is actually an experiential process. One of the things we were interested in doing was studying the possibility of price bubbles. People have this ability. they can’t. The companies participating in the auctions had behaviorists advising them. This led to the idea of looking at asset trading markets in the laboratory by the early 1980’s. It was jump bidders. We brought people back a 2nd and a 3rd time.118 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 It was irrational behavior. Well. They don’t have any idea of that’s what they did. Odean: Did you do that in your classes at Purdue? Smith: Yes. using institutions that somehow survived in our society using those rule systems. From what everyone else is doing. It turns out that people without any People have this ability. We had this incredible contrast. that research program never got off the ground. in one case – markets worked far better than you expected based on the theory. As far as I could see. because economic theory never predicted the weak conditions under which it prevailed. I think the important thing we learned was that the common information wasn’t sufficient to give you common expectations. Odean: In contrast to the early experiments in which markets performed surprisingly well. They will deny there is any kind of model that can predict the convergence properties. We thought: “Let’s see what we can learn in the laboratory?” The original idea was to begin with an environment that was transparent. without any understanding of supply and demand. It is hard for them to imagine anyone modeling what it is they have done. because they don’t have a concept of what there means to be an equilibrium. using institutions that somehow survived in our society using those rule systems. that people would trade at the fundamental value because of the information we gave them. the first ones I ran performed so well I didn’t believe the results. 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). They also will report. It really goes back to William Stanley Jevons’ writing in the early 1870’s that contributed the basic supply and demand . Not exactly. you can’t improve your position. They have a capacity to do well for the group while doing well for themselves. It was what eventually got me hooked into experimental economics. They don’t have the vocabulary. 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. we were getting bubbles. this came as an astonishing surprise. Given my traditional economic training. it led nowhere. Many people will say this is a great victory of economic theory. They have a capacity to do well for the group while doing well for themselves. People find those equilibria too. Traditional theory from financial economics was that all the information in the markets gets quickly incorporated into the price. this came as an astonishing surprise. And we would disturb them to see if we could produce a bubble. training in economics. 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. In my early work on markets. We see what happens to the bubbles in the stock market all the time. This is robust all types of subjects. 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. without any sophistication at all walk into a room and they find the equilibrium of these markets.

Smith: In experimental markets. The model predicts this. 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. 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. This gives you a differential equation model that gives you bubbles. It was. and group A has a larger endowment of cash. SMITH INTERVIEW 119 theory. You can model it by postulating there are two kinds of investment: fundamentalists and momentum traders. no result that said if people have complete information. It is really common in stock market bubbles to put on price change limit rules. We’ve tried all kinds. If you model that we have two groups. This gives you a differential equation model that gives you bubbles. and if they don’t. you get bigger bubbles. The fundamentalists buy in proportion to the discount from true value (fundamental value) and proportion to the premium. It is a beautiful example that if subjects want to dissatisfy the experimenter. Odean: They do get hit occasionally. . Smith: Most of the time. even though we have ways of modeling them. I think. they (the traders) did not understand the instructions. We quickly shot that down with people who had exactly the same endowments. I’ve seen a couple of papers looking at what happens when you hit the limit. the market doesn’t converge. The momentum traders are the type of trader who simply buys in proportion to the rate of change in the price. that performed very badly compared to our expectations. The bubbles can go longer and can carry further. It is still is an experimental process and the traders have to come back for a 3rd time to finally create near fundamental value. 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). That never inspired confidence in it being a very believable model. some may be willing to sell for less than fundamental value because they get a more balanced portfolio and are less exposed. Since people have different endowments of cash and shares. They ignored that. There was no theorem. The momentum traders are the type of trader who simply buys in proportion to the rate of change in the price. 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. what we think is going on is people feel there is a downside limit. what the remaining dividends left were and the holding period. We still don’t know what it is about asset trading that leads to the contagion of bubbles that we observe. they completely falsify his result. This helps us understand why the market is sensitive to monetary policy and what the Federal Reserve is going to do. we expected markets to work better than the theory. You can model it by postulating there are two kinds of investment: fundamentalists and momentum traders. 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. The momentum traders are very much influenced by the amount of money that is around them. By then. they are wide enough that they are not binding. We still don’t know what it is about asset trading that leads to the contagion of bubbles that we observe. Odean: Many of the world’s exchanges have these such as Taiwan and the Chinese exchanges. Essentially. even though we have ways of modeling them. What we did was to compute for them and then remind them in each period. Can you share more on this work? Smith: We thought somehow.ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. and you got bubbles. We are able to come up with models that lead to testable propositions. The fundamentalists buy in proportion to the discount from true value (fundamental value) and proportion to the premium. the market converges.

had been looking at the structure of those prices — at the low priced. Odean: I think of this in terms of liquidity. such as local ones as in Alaska where the oil pipeline was built. I really believe that an important distinction in this housing bubble than previous ones. so real estate has to go up in price. 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. The index started up in 1997. This capital gain wasn’t restricted to once. is that we had a tax law change in 1997. but as a thing that was a collateral effect. went up faster and further and more sharply. bubbles have gone on for maybe 3 or 4 years and then the prices decline. That was a speculative bubble and these types of bubbles have been local. I think if there was a spark that explains why this bubble was bigger than our previous ones. It has never succeeded so dramatically as in this last bubble.S. mid-tier. I don’t see the financial magic that was being done in the subprime as so much a cause. Smith: Yes. the I really believe that an important distinction in this housing bubble than previous ones. people sometimes believe that there is some downside protection. The low priced homes consistently in market after market. I think these two things together. At the low end. My conjecture is. I think what you are getting there is self-reinforcing beliefs about price change. that if you give favorable tax treatment. People had been expecting prices to rise and then the subprime lending industry created so much difficulty for the whole financial system. There is much more than the tax change of course. because a lot of people in the industry realize that some of these assets are risky enough…. The lower tier had the biggest bubble component. of course. Real estate prices just went up. about another 1/3 is in all listed securities) and you are taking one of them and giving special capital gains tax-free treatment. There is evidence that there is pressure on FM&FM to take substandard mortgages. it was in those homes. There was a two year holding period. That is similar to the housing market in 2005 and 2006. is that we had a tax law change in 1997. My colleague. . 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. Or people talked about real estate and said “There is only so much land. both from above (politically) and below (industry). Odean: This was the largest bubble here in the U. Steve Gjerstad. People talked about how you can’t lose money in the real estate market. This tax law change allowed home buyers to receive capital gains from their home that were tax free up to $500. In past housing bubbles. This is by far the largest bubble on record. the ability to buy or not buy was most effected by liquidity. There was Japan.” You get this belief in the markets.S. We’ve long tried to create conditions for people to own their own wealth through home ownership. So the housing bubble had components. you could do it more than once. say to 1/3rd of U. it is without a question the largest housing bubble. This is one of the things I want to look at incidentally. If we could just get the government to take them over. In national statistics.120 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 Odean: In experimental markets.000. 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. that would be one of the things that I would want to look at. Smith: You have the phenomena of boom towns where the real estate goes crazy. The current bubble has hit national proportions and is astonishing — the movement from 1997 to the peak in the CaseShiller index during 2006. wealth (1/3rd of all US wealth is in a form of homes. than this last bubble this time around. and high priced homes. Now – you never had as much trouble with FreddieMae and FannieMac before. These are exactly the people that are much more vulnerable to the decline in prices because whatever the wealth they had.

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

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

Now it has come back down to around $40. 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.  .ODEAN & SIMKINS — PIONEERS OF FINANCE: VERNON L. I think it is very hard to put your finger on what that is. We had a run up in oil prices that almost destroyed the motor home industry and the big gas guzzlers. Regarding how to prevent bubbles. Then prices dropped back down. that precipitate the movement up or the current movement down. 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. SMITH INTERVIEW 123 not sustainable. I think that is as mysterious as ever. Bigger cars are what people wanted to buy. The sparks that actually lead to these bubbles. I’m not sure we ever know the root cause.

“Fannie Mae” and “Freddie Mac” are widely used nicknames for the Federal National Mortgage Association and the Federal Home Loan Mortgage Corporation.3 Over time. § 1451 et seq. Brown (2001) and Bunce (2002). The views expressed do not necessarily reflect those of the Federal Reserve Bank of Atlanta. and politically powerful. Scott Frame Fannie Mae and Freddie Mac are government-sponsored enterprises that play a central role in US residential mortgage markets.1 Together.ofheo. 3 124 . see the results of special regulatory examination reports at: <http:// www. implied public-sector support for their obligations) and define the scope of their permissible activities. and had policymakers voice concerns that they posed a systemic risk to the global financial system. (Freddie Mac). Wayne Passmore. outlines the measures taken by the federal government. the Federal Reserve System. I would like to thank Michael Hammill for research assistance and Mark Flannery. and presents some evidence about the effectiveness of these actions. Mario Ugoletti. § 1716 et seq. and Larry White for providing helpful comments on an earlier draft. Recently. GA. however. for example. 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. as of mid-year 2008. Both Fannie Mae and Freddie Mac have been the subject of a great deal of attention and controversy in recent years. Looking ahead. 2 The charter acts may be found at 12 USC. policymakers became increasingly concerned about the size and risk-taking incentives of these two institutions. For an analysis of the GSEs funding of mortgages for low-income borrowers and underserved areas earlier this decade see. In recent years. policymakers will need to consider the future of Fannie Mae and Freddie Mac. Scott Frame is a Financial Economist and Policy Advisor for the Federal Reserve Bank of Atlanta in Atlanta. Fannie Mae and Freddie Mac are publicly traded financial institutions that were created by Acts of Congress in order to fulfill a public mission. the two institutions held or guaranteed about $5. among others.5 W. Fannie Mae and Freddie Mac are enormous governmentsponsored enterprises. profitable. US housing markets became increasingly stressed through 2007 and resulted in severe disruption to mortgage markets. In September 2008. 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. respectively. 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.2 At the heart of these (and other) issues is the GSEs’ incentive structure. that play a central role in US secondary mortgage markets. (Fannie Mae) and 12 USC. This paper describes the sources of financial distress at Fannie Mae and Freddie Mac.gov/Regulations.aspx?Nav=199>. as well as the appropriate scope of public-sector activities in primary and secondary mortgage markets. Larry Wall. described the systemic risks posed by the GSEs in testimony before the US Congress. or their staffs. 1 trillion in US residential mortgage debt – slightly more than the $5. Diana Hancock. been criticized for not sufficiently targeting their activities toward low-andmoderate income communities and households.124 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 The 2008 Federal Intervention to Stabilize Fannie Mae and Freddie Mac W. Fannie Mae and Freddie Mac became exceptionally large. Former Federal Reserve Chairman Greenspan (2005).3 trillion in publicly held US Treasury debt at that time. or GSEs. These charter Acts imbue the two GSEs with important competitive advantages (most notably. Each GSE has: faced accounting scandals.

On September 7.7 By contrast. 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.” an agency within the Department of Housing and Urban Development. 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. As a result of these developments. 8 According to Frame and White (2005).and moderate-income households and/or in areas viewed as historically underserved (central cities and rural areas). Nevertheless.5 The remainder of this paper will proceed as follows. Congress in 1970 created Freddie Mac. corporate objectives: fulfilling certain social policy goals (and assisting related political constituencies) and maximizing shareholder value. A similar statement for Freddie Mac is located at 12 USC. 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.8 Freddie Mac became publicly traded in 1989 as part of the thrift crisis resolution. These developments resulted in a significant reduction in the availability and cost of mortgage credit for homeowners. Fannie Mae and Freddie Mac have been funded with private capital and their shares are traded on the New York Stock Exchange. which those authors refer to as the “other” housing GSE. and GSE-backed mortgages saw liquidity pressure as evidenced by unusually wide yield spreads. 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. or HUD. 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. 2008.6 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. By law.10 On the other hand. The National Mortgage Association of Washington. with a particular emphasis on housing for low. 7 See Flannery and Frame (2006) for a history and overview of the Federal Home Loan Bank System. financial markets have long viewed the GSEs’ obligations as carrying an “implicit” government guarantee. as Fannie Mae By law. as discussed further below. 4 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.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. from financial institutions around the United States. or VA).(d)-(e) (Fannie Mae) and 12 USC. 10 .FRAME — THE 2008 FEDERAL INTERVENTION 125 was first known. § 1716. 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. today Fannie Mae and Freddie Mac are quasipublic/quasi-private financial institutions. Who are Fannie Mae and Freddie Mac? Fannie Mae’s roots stem from the Great Depression. 9 The maturities of new debt issues by Fannie Mae and Freddie Mac also increased as a result of these policy actions. Fannie Mae and Freddie Mac were placed into conservatorship by their federal regulator: the Federal Housing Finance Agency (FHFA). I. § 1455(h)(1) (Freddie Mac). Its business was to purchase mortgages insured by the Federal Housing Administration. Nevertheless. On one hand. the federal government was compelled to intervene to stabilize both GSEs and mortgage markets more generally. 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. was created within the federal government in 1938. 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. § 1719(b). sometimes opposing. § 1451 [Note]. Fannie Mae and Freddie Mac are limited to operating in the secondary conforming mortgage market and dried-up. See 12 USC. This unusual governance arrangement has resulted in two. by issuing debt and purchasing and holding FHA-insured residential mortgages.9 Hence. or FHA. or “Ginnie Mae. the obligations of Fannie Mae and Freddie Mac must state that they are not guaranteed by the federal government. Concurrent with this action. 6 Fannie Mae was replaced within the federal government by the Government National Mortgage Association. 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. which was owned by the 12 Federal Home Loan Banks and the savings and loans that were members of these Banks. Some concluding remarks are offered in Section IV. each GSE was created by an Act of Congress and is broadly charged with providing liquidity and stability to the secondary residential mortgage market.

is most often done through each institution’s “swap Third. and Stevanovic (2002). whole been estimated empirically to be about 40 basis points. 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. 16 By contrast.) with some past government GSE-backed MBS are very liquid (relative to other asset. allows Fannie Mae institutions. Fannie Mae and “guarantee fee”.15 This belief. of loan pools into federal charters. coupled tradable securities.25 billion of Fannie Mae’s and Freddie Mac’s enhancement of mortgage-backed securities. the conforming loan limit for single-family properties is $417. and Passmore.gov/Regulations. and politically powerful. US General First. Fannie Mae and Freddie exchange for a monthly depository institutions. and liquid fixed-income investment securities. See Frame and Wall (2002) for a discussion.ratings from Standard and Poor’s in terms of their “risk to the government”. 12 See. in the late 1980s.ofheo. 2001) for further discussion. for example.000. This involves the two GSEs holding strength ratings would warrant. which is often referred to as their federal line-of-credit. or MBS. of the comparison bonds and the sample period studied. these ratings incorporated whatever government support or intervention the entity typically enjoyed during the normal course of business. 1719 (Fannie Mae) and 12 USC. However. Second. First.14 Mac became exceptionally large. during the late 1970s and early 1980s. This securities. US Congressional Budget Office (1996. commercial banks (as a group) maintain a ratio of total equity to total assets of about 10 %. Notably. both GSEs are exempt from state and local income taxes. 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. and Burgess (2005).13 that they are not obligations of the federal government. 13 See Ambrose and Warga (1996. 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. for purchase by the Federal Reserve in openensure the timely market operations.126 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 their activities take two broad forms. 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.” as classified programs. 90–91) discusses two past episodes during which 11 See 12 USC. 2002). the Secretary of the Treasury has the authority to guarantee” businesses involve the creation and credit purchase up to $2. (This Freddie Mac use the profitable. however. 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. Nothaft. 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”. although mortgages (purchased from originators under their “cash such estimates vary depending upon the maturity and credit rating programs”).12 and fees of the Securities and Exchange Commission (SEC). but can be as high as $625. process is commonly Federal Reserve as their Recently. Fannie Mae’s and referred to as fiscal agent. in turn. Fannie Mae and Freddie Mac long received AA. the Farm Credit System (another GSE serving the agricultural sector) required a taxpayer bailout totaling $4 billion. 1454 (Freddie Mac).aspx?Nav=128>. Fannie Mae was insolvent on a market value basis and benefited from supervisory forbearance. 17 . Conforming mortgages are those with balances below the legal limits on the size of residential mortgages that Fannie Mae and Freddie Mac can buy.11 The GSEs’ “credit Second. Accounting Office (1990. The GSEs agree to public deposits. For 2009. 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 2008.17 The Fannie Mae and Freddie Mac largely fund these assets with so-called “Federal Agency” debt. premium known as a Fourth. the federal government assisted troubled GSEs.16 This borrowing advantage has MBS that they have purchased in the open market. 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. mortgage market. See <http://www. Pearce.” whereby mortgage originators present pools of under the Securities Exchange Act of 1934. Sherlund.500 in certain high-cost areas. the GSEs’ issue “government securities. and for payment of principal and unlimited investment by interest on the MBS in federally insured Over time.actions.

McKenzie (2002). conduct examinations. OFHEO supervised only two institutions making it prone to regulatory capture. 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. Sparks. The GSEs’ US Congressional Budget Office (1996. . The perceived implied federal guarantee also distorts the GSEs’ risk-taking incentives in a way that may increase the probability of financial distress. the federal government imposes a two-part regulatory structure on Fannie Mae and Freddie Mac. (2) has authority to set minimum leverage and risk-based capital requirements.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. see US Congressional Budget Office (2001). Recovery and Enforcement Act of 1989. Congress formally established a safetyand-soundness regulatory and supervisory regime for Fannie Mae and Freddie Mac in 1992. data sample. and other “underserved areas”. or OFHEO. Prior to the passage of the Financial Institutions Reform. See also Heuson.21 GSE regulatory reform was an active legislative item this decade following the accounting scandals at both Fannie Mae (2004) and Freddie Mac (2003). This $5. However. and Ingpen (2002) for theoretical analyses of the relationship between GSE securitization and mortgage interest rates. 2005).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.e.8 trillion in assets (almost entirely MBS and whole mortgages) and had another $3. and Sparks (2001) and Passmore. or HUD. The Office of Federal Housing Enterprise Oversight. p.FRAME — THE 2008 FEDERAL INTERVENTION 127 equity holders thus perceive a greater-than-normal benefit from risk-taking. and the Federal Home Loan Bank System. and the Federal Housing Finance Board. was authorized to set riskbased capital standards (subject to important statutory limitations). 2001) reported an MBS advantage of 30 basis points. OFHEO lacked the authority both to adjust minimum capital standards and to resolve a failure of either Fannie Mae or Freddie Mac. Fannie Mae and Freddie Mac together held almost $1. 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. LaCour-Little and Sanders (2004). and Ambrose. which is more focused on promoting housing than contending with GSE safety-and-soundness. 2008. In particular.18 The perception of an implied federal guarantee conveys a subsidy on Fannie Mae and Freddie Mac. although the advantage is difficult to estimate. HUD maintained exclusive regulatory oversight responsibilities over Fannie Mae and (for 1989-1992) Freddie Mac. the FHFA succeeds the OFHEO. part of which is translated into lower mortgage rates for consumers. Fannie Mae’s and Freddie Mac’s share of residential mortgage debt outstanding was only 7% (Frame and White. Passmore (2005). 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. 9) critiques the approach that generates this estimate and alternatively argues that the advantage is in the range of 0-6 basis points. 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. but Passmore (2005. and time period studied. OFHEO’s structure and authorities proved deficient in many respects. HUD’s GSE mission regulation.5 trillion in obligations represented almost half of all residential mortgage debt outstanding at that time. and (3) has receivership powers.20 Unfortunately. Freddie Mac. in 1980. and take enforcement actions if unsafe or unsound financial or management practices were identified. The GSEs have become enormous financial institutions – both in absolute terms and relative to the mortgage market as a whole. Prior to the Federal Housing Enterprises Financial Safety and Soundness Act of 1992. net of those held in their own portfolios. The two GSEs have also grown much more rapidly than the residential mortgage market as a whole over the past three decades. Frame. As of June 30. (A similar situation is wellunderstood in the context of federally insured depository institutions. 19 As discussed in Eisenbeis. housing in central cities. Freddie Mac was the responsibility of the Federal Home Loan Bank Board.) 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. Passmore. 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. 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. With respect to supervisory tools. For example. 21 22 By doing so. The US Department of Housing and Urban Development..and moderate-income families.7 trillion in net credit guarantees outstanding – i. In order to maximize benefits to homebuyers and minimize taxpayer risk. The agency was also an independent arm of HUD. 20 perceived implied guarantee also affects the interest rates on MBS that Fannie Mae and Freddie Mac issue. and Wall (2007). The new law created the Federal Housing Finance Agency (FHFA). OFHEO was also subject to Congress’ annual appropriations process and sometimes fell victim to political meddling. which consolidated the mission and safety and soundness oversight for Fannie Mae. 18 For an introduction to this literature. and the references in these papers.

and then jumped markedly to exposed to US residential mortgages. largely Freddie Mac certainly fit this bill. This combined $65. One index fell 4. See Leventis (2007) for an analysis of the differences between the two indices.gov>. Fannie Mae and Freddie Mac were not only singularly the first half of 2007 alone.com/portal/site/sp/en/us/page. In any event.7 trillion in combined net differences by geography. largely as a result of house financial crisis).5 billion (Fannie Mae) and -$5.1.0. 23 24 See Foote. and Willen (2008) and references therein. and expanded to almost $1. owing to a statutory minimum half of 2008. Greenspan 2005. 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. Other borrowers may default after finding Mac reported book values of equity of $41. and for and 0.2 billion and $12. As of June 30. Fannie Mae and businesses incurred rapidly increasing expenses.9% capitalcontrast. Bernanke 2007).8 these repeat-sales indices trillion in combined assets and the weakening global economy. During 2006.5%. in turn.5% for on-balance-sheet assets related expenses almost double to $12.. 2008:Q3 alone they totaled $15.5 billion during the second half of that year.6 billion during Washington Mutual. While the year later.9 themselves in a situation where their expectations of future billion.8 billion. Information about the OFHEO house price index can be found at: <www. “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. According to Financial Accounting Statement (FAS) Number 157. Fannie Mae and Freddie Mac again saw creditcapital requirement of only 2.0 billion in new equity) and Freddie Mac lost often referred to as “negative equity”.com/media/pdf/newsreleases/q22008_release.7 trillion in combined assets (3. During those four intervening House price declines quarters. By respectively. mounting foreclosures. Fannie Mae operating on a nationwide basis like Countrywide and and Freddie Mac together incurred about $1. but their capital will remain elevated for some time.1. significant problems emerged in both of that.2 trillion in net off-balance nationwide house price Given the ongoing decline in house sheet credit guarantees.topic/ indices_csmahp/0.3 billion. the market value of positive equity in the foreseeable future. mark-to-market accounting evidence suggest that negative equity is a necessary condition losses on ‘available-for-sale’ mortgage-backed securities for mortgage default. mounting foreclosures.0.128 II.1 billion in creditrelated expenses.fanniemae.html>. Fannie Mae and Freddie Mac maintained price declines in many parts of the country. Gerardi.0.5 billion in equity stood against almost $1. respectively. Economic theory and another $4. 2008. 25 .0. These expenses rose to $1. 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. the two GSEs had magnitudes of decline in prices.pdf> (Page 17) and <http://www. During 2008. Fannie Mae and Freddie living expenses. this national decline in cushions had begun to erode. 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.7 billion. 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.g. Moreover. The credit guarantee riskier subprime borrowers and investors.pdf> (Page 13).com/investors/er/pdf/2008fintbls_080608. it differ – owing to coverage $3.8 billion. Between 2007:Q2 and 2008:Q3 house prices declined 18.0% on a nationwide book equity values of $39.7 billion and $25.standardandpoors.6 billion (Freddie 25 in a tremendous wave of mortgage defaults and foreclosures Mac). and loan quality – guarantees. see: <http:// www. and the Information about the S&P/Case-Shiller house price index can be found at: <http://www2. the OFHEO $3. 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. over the same to-assets ratio) and another period.e.0.0. For the GSEs’ fair value balance sheets. is likely that the GSEs credit losses off-balance sheet credit loan size.. For the first a high degree of leverage. 23 house prices is unusual.0.5 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. basis based on the S&P/Case-Shiller composite index.0. as did thrift institutions owing to loan loss provisions.freddiemac. As of mid-year 2007 (and prior to the beginning of the stressed during 2007 and 2008.0. but also operated with $6. Fannie Mae posted $9. Given the ongoing Concerns about the GSEs’ concentration of residential decline in house prices.45% for net off-balance sheet credit guarantees.0.0.ofheo.0.0.

Figure 3 shows prices for credit default swaps on Fannie Mae and Freddie Mac senior and subordinated debt between January 2007 and July 2008. 26 III. Debt investors also sought clarity from the federal government about whether bondholders would be shielded from any losses that might ultimately arise. and Morgan Stanley. the Federal Reserve. Figure 2 illustrates how the GSEs’ share prices fell during that time (and following an even more dramatic decline during the fall of 2007). 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. such as measured by credit scores. Interestingly. However.7 billion in mortgage securities backed by subprime and Alt-A mortgages — virtually all of which were rated AAA.FRAME — THE 2008 FEDERAL INTERVENTION 129 and Freddie Mac.29 In response to increasing concerns that Fannie Mae and Freddie Mac would be unable to rollover their debt. Figure 4 presents relevant weekly data based on holdings in custody accounts at the Federal Reserve Bank of New York.26 As of mid-year 2007. the two GSEs held $252. 30 . Federal Intervention According to former Treasury Secretary Henry Paulson (2009). it is likely that the GSEs credit losses will remain elevated for some time.27 The GSEs’ holdings of such securities likely reflected at least two factors. Given that US mortgage markets had already been disrupted for almost one year at that time. of this amount. securities classified as “hold-to-maturity” are not marked-to-market unless there is an “other than temporary impairment” to value. the same data indicates that holdings in foreign private accounts peaked sooner – as of 2007:Q4. 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. Solomon and Paletta. Fannie Mae and Freddie Mac account for about 70% of all Federal Agency obligations outstanding. 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”. However. One is the distorted risktaking incentives faced by Fannie Mae and Freddie Mac because of the perceived implied federal guarantee of their obligations. investors became increasingly concerned about the financial condition of both Fannie Mae Private-label mortgage securities are those not guaranteed by Fannie Mae. foreign official institutions. The observed widening is believed to be primarily caused by the financial market turbulence.3 billion. or GAAP. one especially significant and risk-averse investor constituency. began decreasing their holdings of Federal Agency obligations at that time. the Treasury began a comprehensive financial review of Fannie Mae and Freddie Mac in conjunction with the FHFA. 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. Subprime mortgages refer to those loans made to borrowers with riskier credit characteristics. The other key issuer is the Federal Home Loan Bank System. 2008). former US Treasury Secretary Henry Paulson requested that the federal government be given broad authority to invest in the two GSEs. weakening global economy.) 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. the aforementioned credit problems at Fannie Mae and Freddie Mac also likely played a role by pushing-up required yields on the GSEs’ MBS. Alt-A mortgages refer to loans made with little or no documentation. loan-to-value ratios.. (Under Generally Accepted Accounting Principles.30 The GSEs believed in their solvency and thought that any capital deficiency below regulatory minimums could be rectified by significant asset sales. which led to a heightened demand for US Treasury obligations that was reflected by lower Treasury yields. 27 28 The GSEs’ housing goals were established in the Federal Housing Enterprises Financial Safety and Soundness Act of 1992. Mark-to-market losses also occurred in each GSEs’ holdings of “private-label” mortgage securities backed by subprime and Alt-A mortgages. the prospect of Fannie Mae and Freddie Mac retrenching was not an appealing policy option.107) corroborates the trend illustrated by showing that foreign official holdings of Federal Agency obligations peaked in 2008:Q2. Of particular note are the spikes in March 2008 (just prior to the Bear Stearns rescue) and then again during June and July 2008.g. Another factor was the HUD-regulated affordable housing goals that mandated a certain percent of each institution’s business devoted to affordable housing. Fannie Mae accounted for $81. Quarterly Flow of Funds data (Table L. The data comes from a memorandum to the Federal Reserve’s H. 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. immediately following the passage of the new housing legislation. During the summer of 2008. Freddie Mac.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. both Fannie Mae and Freddie Mac released their second quarter earnings. As of June 30. or debt-to-income ratios. In early August 2008. That provision was included in the Housing and Economic Recovery Act that passed in July 2008.1 release: Factors Affecting Reserve Balances.4. This responsibility was transferred to the FHFA as part of the Housing and Economic Recovery Act of 2008. 2008 Data provided in US Office of Federal Housing Enterprise Oversight (2008) indicate that.4 billion and Freddie Mac $174. or Ginnie Mae.

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 .

100 1.FRAME — THE 2008 FEDERAL INTERVENTION 131 Figure 3: Fannie Mae & Freddie Mac Credit Default Swaps p basis points.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 . 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.

8 billion. 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. 2010) to be determined by the Treasury and the FHFA (as conservator) in consultation with the Federal Reserve. or merging into another entity. there was a compelling case that – on an economic basis – both were actually insolvent. and is priced at LIBOR plus 50 basis points. in exchange for the senior preferred stock agreements. Each of these agreements is of an indefinite term and for up to $100 billion. Second. officers.F. see: 12 C. (2) warrants for the purchase of common stock representing 79. (2) terminating conservatorship other than in connection with receivership.35 As of year-end 2008.pdf. Eligible collateral is limited to Fannie Mae and Freddie Mac mortgage-backed securities and Federal Home Loan Bank advances. and (4) acquiring. Preliminary figures suggest that Fannie Mae will require as much as $16 billion and Freddie Mac as much as another $35 billion (Kopecki 2009). Frame. The senior preferred stock accrues dividends at 10% per year. (2) having the Treasury enter into senior preferred stock agreements with both GSEs. The MBS purchase program. 31 See <http://ustreas. the FHFA assumed the responsibilities of the directors.132 both GSEs were both technically solvent insofar as the book value of their equity capital was positive. 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.4 billion for Freddie Mac. or (2) 10% of Tier 1 capital. 35 . the Treasury will contribute cash capital equal to the difference in exchange for senior preferred stock. for state member banks. as mentioned previously. 32 Credit must be collateralized. First. See Eisenbeis. which had become widely viewed as posing a systemic risk to the financial system and providing little social welfare benefit.6 billion (Fannie Mae) and -$5. the Treasury received from each Fannie Mae and Freddie Mac: (1) $1 billion of senior preferred shares. Fannie Mae had $41. and Wall (2007) for an overview of the policy concerns and the related literature.34 The senior preferred stock agreements also included various covenants. Freddie Mac. (At that time. Both GSEs are expected to require significant Treasury capital infusions after the announcement of their respective year-end 2008 financials.9% of each institution on a fully diluted basis.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. Specifically. made a compelling case for swift federal action.31 These facts. For example.2 billion in book equity and Freddie Mac $12. The Treasury’s GSE credit facility is for Fannie Mae. redeeming or issuing any capital stock or paying dividends. Treasury approval is required before: (1) purchasing. Freddie Mac drew $13.R. and (3) a quarterly commitment fee (starting March 31.5 billion (Freddie Mac). had 33 While acceptable under GAAP.) However. Treasury Secretary Henry Paulson. can be extended for one-to-four weeks. This provision was intended to assuage policymaker concerns about the GSEs’ investment portfolios.9 billion. no credit had been extended through this program. These amounts were $20. 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. By becoming a conservator.6 billion for Fannie Mae and $18. a rate that steps up to 12% if in any quarter dividends are not paid in cash. Section II(B)(4). This included: (1) placing both Fannie Mae and Freddie Mac into conservatorship. 208 Appendix A. taken together with deteriorating mortgage market conditions and a view that the GSEs had been especially conservative in estimating expected future losses. FHFA Director James Lockhart. 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. 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. (3) increasing debt to greater than 110% of that outstanding as of June 30. Also. the GSEs’ reported fair values of equity were much lower – and in Freddie Mac’s case fair value was actually negative.32 And on September 7. both institutions were carrying relatively large “tax deferred assets” to allow them to reduce future income taxes. 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. consolidating.gov/press/releases/reportspspa_factsheet_090708% 20hp1128. by contrast. 2008. providing counter-cyclical support to mortgage markets and financing affordable housing. 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. 2008. Concurrent with the conservatorships. After its 2008:Q3 earnings release. New CEOs were named to act as agents of the conservator. and Federal Reserve Chairman Ben Bernanke outlined a plan to stabilize the residential mortgage finance market. the Treasury entered JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 into a senior preferred stock agreement with each GSE. and the Federal Home Loan Bank System and is operated by the Federal Reserve Bank of New York.

the senior stock purchase agreements have become an appendage to the GSE charters. the Federal Reserve announced on November 25 that it was establishing new 36 IV.html>. However. 37 The NYSE Listing Manual (Part 802. 36 Consistent with the GSE investment provisions in the Housing and Economic Recovery Act of 2008. Indeed it is unclear what role Fannie Mae and Freddie Mac will ultimately play in the US housing finance system. resulted in conforming mortgage rates falling by about 50 basis points. Hence.fms. Nevertheless. Figures 5 and 6 also show a positive market response to these announcements.39 The positive bond market reaction.5 billion.gov/mts/index. the federal government may need to redefine its role in supporting primary and secondary mortgage markets. 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. additional research and policy analysis should commence quickly about the public-sector’s role in mortgage markets. Indeed. the efficacy of the GSE model of financial intermediation.00 over a consecutive 30-day trading period. Policymakers then searched for additional tools to lower and stabilize the cost of mortgage finance. suggested that the imposition of conservatorships at Fannie Mae and Freddie Mac was.FRAME — THE 2008 FEDERAL INTERVENTION 133 facilities to: 1) purchase up to $500 billion in mortgagebacked securities guaranteed by Fannie Mae. Once notified. and the reasons for this uncertainty do not solely rest with the two GSEs. accumulated $71. and monoline bond insurers. down from $60 just 12 months earlier. and the future of Fannie Mae and Freddie Mac. 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. in turn. . 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. and Ginnie Mae. In response. See <http://www. mortgage rates and yields on Fannie Mae and Freddie Mac obligations had climbed back to pre-conservatorship levels because of worsening financial market conditions. these actions were expected to lower and stabilize the cost of mortgage finance. In some sense. the Treasury issued a press release intended to clarify the status of the senior preferred stock agreements. a company has six months to bring its share price and average share price above the $1.37 By extension. coupled with a relatively smooth operational transition. private mortgage insurance companies. The senior preferred stock agreements may only be terminated by either: 1) full funding by the Treasury ($100 billion). the two agreements had significant negative consequences for the GSEs’ common and preferred stockholders. Of course. Indeed. 38 39 See McGeer (2008) and Blackwell and Flitter (2008) for some discussion. or 3) GSE satisfaction of all liabilities. a success. by November 1. so far. Fannie Mae and Freddie Mac were each notified in November 2008. Freddie Mac. Fannie Mae and Freddie Mac common shares quickly fell below $1.gov/press/releases/hp1131. 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.htm>. However. Figures 5 and 6 illustrate the announcement effect for Fannie Mae and Freddie Mac 5-year debt spreads and current coupon MBS spreads. Today’s consensus appears to be that the previous publicprivate business model is inherently flawed and unstable. The tighter spreads on mortgage-backed securities. As real estate prices fell and mortgage defaults and foreclosures mounted. See <http:// www. credit extensions and MBS purchases must be made by year-end 2009 (although previously purchased securities may be held beyond that time). the two GSEs now face delisting. respectively. and the Federal Home Loan Bank System. as a result of trading at such low levels. 2008. and 2) purchase up to $100 billion in debt obligations of Fannie Mae. 2) GSE liquidation. 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. Federal Reserve Chairman Bernanke (2008) and former Treasury Secretary Paulson (2009) have offered some initial thoughts about various policy options.ustreas. In response. several community banks became financially distressed themselves as a result of having to write-down the value of their holdings of GSE preferred stock. Freddie Mac. Conclusion Fannie Mae and Freddie Mac play a central role in the US residential mortgage finance system. 2008. The financial distress at Fannie Mae and Freddie Mac has occurred along with significant and well-publicized problems at a host of mortgage originators. the two highly leveraged GSEs became financially distressed. On September 11.treas.38 Preferred shares suffered a similar fate.00 threshold.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.

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 .

Sparks. “A Reconsideration of the Jumbo/NonJumbo Mortgage Rate Differential.” American Banker (September 11). Sparks.S. Gerardi.. Willen. “Fannie. “Negative Equity and Foreclosure: Theory and Evidence. http://www.” Journal of Financial Services Research 31 (No.org/publications/hsgfin/ workpapr13. B. 2007. A. (March 6).FRAME — THE 2008 FEDERAL INTERVENTION 135 Frame. 2).” Federal Reserve Bank of Atlanta Economic Review 87 (Q1). “The Effect of Housing Government-Sponsored Enterprises on Mortgage Rates. 2002. Wall. “The Federal Home Loan Bank System: The ‘Other’ Housing GSE. B. Blackwell. “An Analysis of the Systemic Risks Posed by Fannie Mae and Freddie Mac and An Evaluation of the Policy Options for Reducing Those Risks. and J. Passmore. 541-569.W. 29-43.L. 2005. Housing. Warga. 2) 215-242. 2002. “GSE Portfolios. 169-204. “GSEs. Ingpen. 2005. M. http://www. 2005.pdf.gov/boarddocs/ testimony/2005/20050406/default.” In US Department of Housing and Urban Development. 151172. Paulson. Eisenbeis. Bernanke. 2004. How Much Fire?” Journal of Economic Perspectives 19 (No. Brown. Nothaft. and R. “Measuring Potential GSE Funding Advantages.” Federal Reserve Bank of Atlanta Economic Review 91 (Q3). D.B. 2002. Flannery.. and A.” http:// www.... B. Freddie.” Journal of Urban Economics 64 (No. the Economy. “The Role of the GSEs in Supporting the Housing Recovery.” Journal of Real Estate Finance and Economics 25 (No. A. 3).” Journal of Real Estate Finance and Economics. “Debt Spreads Between GSEs and Other Corporations. Passmore. Flitter. 2). Burgess. “The GSEs Funding of Affordable Loans: A 2000 Update. 2008. Passmore.” United States Senate (April 6). Warga. 2). and W.federalreserve.D. Mortgage Rates. 2008. “A Note on the Differences between the OFHEO and S&P/Case-Shiller House Price Indexes.S. Serving Two Masters Yet out of Control. B.. J.htm. H. and E. Sanders. Freddie Funding Needs May Pass $200 Billion. Frame. 4).S. S. 2009.S. “Financing Housing through Government-Sponsored Enterprises. 2002. “Regulators and Bankers at Odds Over GSE Seizure. Sherlund. and L. “Testimony before the Committee on Banking. 153-165. Leventis.ofheo. 2008.” American Banker (September 9).D.” Journal of Real Estate Economics 32 (No.E. B. Frame. K.” Journal of Real Estate Finance and Economics 25 (No. and G. W. Heuson. 2009.W. and A. D. LaCour-Little. 2007. W. B. 2007. Frame.E. Foote.” In Peter J. 129-150.. “Fussing and Fuming over Fannie and Freddie: How Much Smoke. and D. “Preferred Exposure Fallout.” Wall Street Journal (August 6). 2-3)..gov/media/research/notediff2. Ambrose. Stevanovic. W. 2008. R.” Real Estate Economics 33 (No. D. McKenzie.. FHFA Says. Studies on Privatizing Fannie Mae and Freddie Mac Washington. Kopecki.. 2). 3). 159-184.C. Ambrose. 2002.A. Bernanke. Paletta. 2008. Greenspan. H. Solomon. Wall. 1996. “The GSE Implicit Subsidy and the Value of Government Ambiguity. Editor. 427463. 2006.. HUD. AEI Press.. 2001. C. W. “Implications of Privatization: The Costs to Fannie Mae and Freddie Mac. and Public Policy (October 31).. . Washington D. 2001. and A. “Reform of GSE Housing Goals..html.C.” Journal of Real Estate Finance and Economics 25 (No. 2).S. and L. J..” Bloomberg (February 10). 3354..” Remarks to the Independent Community Bankers of America.” Journal of Real Estate Finance and Economics 25 (No. M. 2005. 3).S.J. F. and Urban Affairs. Ambrose. Passmore. W. J.. 23 (No. “Treasury Hires Morgan Stanley for Advice on Fannie. 465-486. 75-99. A.. Bunce. and Affordable Housing.” Remarks to the Economic Club of Washington (January 7). R. “Credit Scoring and Mortgage Securitization: Implications for Mortgage Rates and Credit Availability. and S. R. 2002.L. White. 197-213. 234-245.” Remarks to the UC Berkeley/UCLA Symposium: The Mortgage Meltdown. and P. and L. McGeer..” US Department of Housing and Urban Development Housing Finance Working Paper HF-013 (April). W. W.” Real Estate Economics 33 (No.. Pearce.S. and the Long-Run Effects of Securitization. “The Effect of Conforming Loan Status on Mortgage Yield Spreads: A Loan Level Analysis..huduser.J. “The Future of Mortgage Finance in the United States. Wallison. 337-363. Systemic Risk.

D.C. Federal Subsidies and the Housing GSEs. US Office of Federal Housing Enterprise Oversight. GovernmentSponsored Enterprises: The Government’s Exposure to Risks. D. .. CBO. US Congressional Budget Office. Washington. Washington.C.C. 2008. 1990. CBO. Washington.C. 2001. Assessing the Public Costs and Benefits of Fannie Mae and Freddie Mac. D. 1996. JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 US General Accounting Office.136 US Congressional Budget Office.. D.. Washington.. Report to Congress. GAO. OFHEO.

“Involving the Workforce in Financial Planning”. McGraw Hill: 2008. Most chapters. 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. Chapter 5. but the level of discussion is a bit advanced. powerful individuals to make biased decisions and introduces Shefrin’s recommendations for improving the way corporations are managed.Book Review: Ending the Management Illusion: How to Drive Business Results Using the Principles of Behavioral Finance By Hersh Shefrin. The second chapter discusses the financial metrics the author would include in each employee’s training. returns to the text’s strongest feature. is quite dense. even for finance professionals if they do not use valuation principles on a regular basis. “Narrow Financial Focus on Projects and Financing: Traditional Approach”. are concise and interesting. 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. 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. 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 3. Chapter 6. “Motivating the Workforce Through Smart Carrots and Sticks”. It uses excellent real-world illustrations to explain the psychological arguments and traits that induce educated. The middle chapters of the text are its weakest components. 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. The first chapter of the text is one of the strongest. which is the link between business decisions and behavioral 137 . The text is written at the middle-management level and will also be useful for academics that teach or consult for executives. which can easily be digested in separate sessions.

2) that all stakeholders should be educated about how financial information is presented and evaluated. “Integration: The Whole Ball of Wax”.1 is a list of forces that should appear in every conference room in corporate America where managerial decisions are made. but the real-world focus here is quite narrow. and 3) that everyone makes biased decisions but that these biases can be identified and corrected. Its main messages are 1) that sharing information. 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. Andrea Heuson Professor of Finance University of Miami . (especially financial information) with employees at all levels improves productivity and efficiency. “Sharing Information Throughout the Organization”. with most of the examples being drawn from a single local firm. manage or train individuals to battle the “gremlins”.138 psychology via a discussion on constructing appropriate bonus and incentive plans. 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. Table 8. 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. Shefrin discusses various types of information that firms post on their walls to incentivize employees. 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. The text ends with chapter 8. Throughout the text. The chapter could have easily been combined with chapter 7. This issue is intriguing for an academic. (a favorite term the author uses to describe typical behavioral glitches that bias human decisions) that confront us all every day.

While it’s an injustice to the author to attempt to summarize his work in a single paragraph. Former Chancellor of the Exchequer in the UK. however.3 million jobs in the service sector would be moved to other countries by 2015. Here too. the result of this decline in the US’s scientific and technological competitive advantage is a weakening of the US economy. along with the economy-wide meltdown of the second half of 2008. scientists. The techno-nationalist position is supported by numerous statistics. Gordon Brown. some of which Dr. he cites a 2006 study by Alan S. According to the techno-nationalists. Bhide outlines in his book. 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. Dr. For instance. statisticians. mathematicians. refreshingly optimistic.” The techno-nationalists believe this is already occurring. Dr. European Union countries were granting 54% more of these doctorates than the US. Dr. and researchers. “Leadership in science and technology gives the United States its comparative advantage” (p. 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. Bhide. By 2001. 262). The ability to “out innovate” the rest of the world drives the economy. the recent troubles of the Detroit automakers. 262). Bhide gives particular attention to what he terms the techno nationalists – those who believe. Bhide reports on p.Ds as the US.Book Review The Venturesome Economy By Amar Bhidé. Freeman and Prestowitz’s arguments Dr. summed up the argument thusly (as Dr. China went from granting almost no doctorates to granting roughly 1/3 as many Ph. “Every advanced industrial country knows that falling behind in science means falling behind in commerce and prosperity. Blinder in which Blinder claims that 40 million jobs in the service sector were at risk of being off-shored. Over that same period. Freeman’s work is especially representative of the cause. Amar Bhide offers an in-depth. rejects the techno-nationalist paradigm and rebuts the fundamental thinking behind this technologically protectionist movement. Bhide outlines are consistent with the much publicized recent studies showing the US school children slipping in international academic comparisons. Surely. For instance. 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. Dr. so will the economy. and insightful counterpoint to the increasingly loud voices decrying the trend of economic globalization. On this point. Freeman also notes the decline in publications and citations of US researchers. 264). they offer empirical support. He notes that in 1970 US educational institutions granted over half of all doctorates awarded in science and engineering fields. “The share of papers (by US researchers) counted in the Chemical Abstract Service fell from 73% in 1980 to 40% in 2003” (p. because an increase in the supply of highlevel know-how helps mid-level innovators . Bhide writes of a similar report from Forrester Research in 2002 that estimated 3. ix + 499 pages  In The Venturesome Economy (2008. He cites the work of respected scholars such as Clyde Prestowitz and Richard Freeman to outline the empirical foundation of the techno-nationalist movement. Princeton University Press). he quotes a New York Times article that states. Techno-nationalists argue that if that competitive advantage erodes. Bhide devotes more discussion to alarming forecasts than actual data. Princeton University Press: 2008. add fuel to the techno-nationalist fire.

which readers will appreciate. or they can simply jump directly to the policy debates of Book II and use Book I as reference material. Dr. 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. 500 List. It is the country where all innovators seek to land their products and services. The example he gives is that the invention of the silicon-based micro-processor represents a high-level innovation. Mid-level innovation is the application of a high-level innovation. Readers can choose to read the foundation in Book I and then proceed to the prescriptions of Book II. 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. Bhide makes up for with enlightening qualitative analysis springing from countless hours of interviews with CEOs of important mid-level companies. I found the chapter prefaces and conclusions to be concise and content rich. He goes on further to argue that mid-level innovators are relatively unconcerned with the location from which the high-level innovation emanates. and again represents an innovation per se. it is possible that there is no more important issue than economic globalization. For that reason. Literally. almost every page in the book contains an example that relates the points Dr. This iterative development process requires. the creation of the motherboard is a mid-level innovation. 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. On this point. Bhide argues in Book I. Regardless of where the high-level innovation comes from. The first point highlights the importance of an economy filled with “venturesome” consumers who are willing to try new innovation. regardless of whether the reader accepts or rejects Dr. Readers will benefit greatly from the bountiful mini-case studies throughout. He stratifies innovation into three levels: high. 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. High level innovation is the kind that deals with molecular or physical advances. not to the same degree as the mid-level innovation. through his interviews and surveys of CEOs of VC-backed businesses. And although his study is intentionally short on econometrics (a point I return to at the end of my review). but again. It represents an innovation per se. and ground-level innovation. The book is written in easy to digest modules. The second part of the book (Book II) is devoted to a discussion of policy debates related to the tide of globalization. Such venturesome consumers attract all levels of innovation. but not to the same degree. further intensified by the collapse of the US auto industry. and the development of the personal computer represents a ground-level innovation. Dr. Regarding the continued prosperity of the United States. 59).) 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. These points present critical ramifications: 1. Dr.140 based in the United States develop products that increase productivity and wages in the United States (p. Bhide’s thesis. Bhide is convinced that midlevel innovation is not currently and will not in the near future be something that can be moved off shore. and 2. 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. what he lacks in numerical analysis Dr. mid. 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. Dr. In other words. geographic. Bhide is articulating to the real-world companies he studied. In Book I. Bhide’s current work with his previous work focusing on firms from the well-known Inc. I believe Dr. Ground-level innovation represents the application of a mid-level innovation. 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. each chapter is written to allow readers to spend as much (or as little) time as necessary. we cannot afford to take the wrong position on this issue. 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. for the most part. The first part is supplemented by numerous comparisons and contrasts of Dr. the mid-level . so long as they are able to transform it into something marketable and beneficial to their consumers. But the US’s attitude toward and response to economic integration on a global scale is something we must get right. that mid-level innovators play a key role.) having consumers that embrace new technology makes an economy the natural magnet for all levels of innovation. His book is both timely and relevant. especially the debate regarding the importance of developing high-level innovation on the US soil (and by the US citizens). while the bodies of the chapters were persuasive and intriguing. Bhide believes the US represents an innovation Mecca. language. and cultural proximity. On the latter point. Similarly.

Dr. it is important to recognize that much of Dr. The overarching message of the book is one of optimism.D. There is little in the way of rigorous econometric analysis. 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.) In short. Regarding the second point. 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. Dr. He is very clear about why he chose not to do that. The innovation cycle creates jobs. 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. he cites a study by Gene Epstein in which Epstein notes that from 2002 to 2006. Bhide even provides convincing statistical support for this rosy cycle of innovation and economic prosperity. reap the lion’s share of the benefits. scientists. Bhide. employment in the two sectors most vulnerable to off-shoring had increased 7. Bhide provides evidence from his study that Ph. Bhide’s conclusions on the topic are derived from qualitative analysis and critical thinking. facilitating mid-level innovation. 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. I come away feeling as though Dr. which will not be outsourced or off shored. Given the importance of the subject. Further. (While he concedes ground level innovation and basic production can be and is off-shored. high-tech products. Bhide argues that pursuing a policy of protectionism and isolationism will do nothing but hinder this process. from Mumbai to Paris. both by creating jobs and by providing its citizens with a beneficial new product or service. In fact although I don’t see this explicitly stated in the book. he is adamant that mid-level innovation is difficult to offshore.’s are not necessary for mid-level innovation. Most of Dr. Colby Wright Assistant Professor of Finance Central Michigan University innovation that results will benefit the country where it occurs. This willingness of consumers creates a magnet market for all levels of innovation. his reasoning and qualitative evidence is sound and convincing.7%.D.WRIGHT— BOOK REVIEW OF THE VENTURESOME ECONOMY 141 the means and gusto to buy new. 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. 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. to have been a career mistake (it was unnecessary in performing the roles they fill inside the company). mathematicians. Bhide’s work would have benefited from a more rigorous econometric approach to his study. Any decisions we make and any policies we enact must be done only after careful and measured investigation. 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 . Instead. mid-level innovators in the US will use the imported or domestically produced innovations to develop even more innovations. I believe Dr. Bhide’s discussion on this topic is derived from his own work with the CEOs of VC-backed businesses. In spite of this one perceived deficiency. He even quotes one CEO who states frankly that most Ph. For instance. which was a conscious choice by Dr.’s at his company consider the attainment of their Ph. Nevertheless. while employment in the overall service sector increased 4. statisticians. 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. and researchers.5%. As all levels of innovation come from various sources. innovation breeds innovation.D. But equally important. which he describes as being almost universally mid-level innovators. the innovation cycle creates products and services from which consumers. not innovators. 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. Dr.

Financial Research and Management Decisions. but with diversifiable noise added. Thus NPV(C) has a lower  and a lower E. Assume a constant future tax rate T. (a) Book depreciation always equals economic depreciation. —The Editors Financial Puzzles Stewart C. A project amounts to a long position in PV(Revenue) and short positions in PV(Fixed costs) and PV(Variable costs). C’s variable costs are a partial hedge against C’s uncertain revenues. p. Condition (b) was first derived in Solomon and Laya (1967). Think of a project balance sheet. Project C’s costs are variable. Project A’s costs are fixed. NPV = PV(Revenue) – PV(Fixed costs) – PV(Variable costs).142 JOURNAL OF APPLIED FINANCE — FALL/WINTER 2008 With the previous issues of JAF.5. Project B’s costs are fixed. 131 11. 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. A’s and B’s costs do not provide such a hedge. 1 142 .1 12. (b) The firm settles into steady-state growth and the growth rate equals the IRR. 2007 issue of the Journal of Applied Finance. New York. but does not do so.. with PV(Revenue) on the asset side and PV(Fixed costs) and PV(Variable costs) on the liability side. The issuing company can call and force conversion now. Book return equals the true rate of return (IRR) in the following cases. ed. Proposed Answers for Problems 11 – 13 In the Fall/Winter. Net cash flows and NPV are safer (lower  ) for project C than for A and B. John Wiley & Sons. Myers 14. we began publishing Financial Puzzles by Stewart C. Laya (1967). This set (problems 14-15) is the fifth installment. A.. This problem considers how corporate income taxes can affect IRRs for capital investment projects. Inc. 15. (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. The solutions for the previous set are also given below. Robichek. A convertible bond emerges from call protection with the conversion option well in the money. The overall position is safer (lower  ) if the short position is in a positive- asset. “Measurement of Company Profitability: Some Systematic Errors in Accounting Rates of Return. so that there is no cash-flow advantage or disadvantage from conversion. Economic depreciation would be calculated as the change in the PV of project cash flows using the IRR as the discount rate. Myers. Solomon and J. with  = 0.” in A.

poor performance can actually benefit the institutional investor. A slightly lower cash flow could leave IRR < RCAP.MYERS — FINANCIAL PUZZLES 143 cost of capital than NPV(A) and NPV(B). when earlier cash flows are already money in the bank. A large project cash flow for month 61 could be just enough to achieve IRR > RCAP and put the investor out of the game. but not ex post. Two or more cash flows can be better than one. (a) NPV > 0 when x is close to zero. so NPV > 0. so the value remaining to the institutional investor must be less than $100 million. even if the first cash flow is not as high as it could have been. Suppose that IRR after 60 months is just shy of RCAP. This residual claim has positive value. the investor may not get all future cash flows. The ideal outcome for the investor would be a series of cash flows. but the payoff is risk-free. each as large as possible without bringing IRR > RCAP. (b) NPV at first increases as x increases and then declines. 13. The value of the investor’s claim would have to be calculated from a Monte Carlo simulation of certainty-equivalent cash flows. When x = 1. This would be a disappointing outcome ex ante. (d) Financing from the institutional investor could require RCAP > r. The risk that the investor will not earn RCAP increases. NPV > 0 for small values of x. However. the institutional investor contributes $100 million to a project with PV = $100 million. If the institutional investor only has to put in $1 for a $100 million project. The residual claim that gets all cash flows after the institutional investor is out of the game is also a path-dependent option. But a larger x also means that the investor has to put in more money and wait longer to earn IRR = RCAP. NPV < 0 at x = 1. The return r includes a risk premium. . at the start of month 0. (c) Once the project is up and running. 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. and value finally declines. C is the more valuable project. There is another residual claim that gets all subsequent cash flows if and when IRR reaches RCAP. The simulation is necessary because the test for IRR > RCAP in any given future period is path dependent. in order to give the investor additional upside to offset the negative NPVs noted in (a). for the reasons given in (a). then the investor is almost certain to earn the return r = RCAP immediately. NPV < 0 when x = 1. assuring the investor of cash in both months 61 and 62 and possibly in later months.

Tampa. When algebraic terms do appear in the text. 7-26. 1990.org-Journal of Applied Finance. and $130 for doctoral students who are not FMA members. Minimize extensive content footnotes. New York. Chicago. The New Corporate Finance. authors. Of particular interest are manuscripts with an applied orientation falling into one of the following broad categories: traditional research articles (empirical. 90-97. 1). Harper Business. Stuart C.. Names of authors should not appear on the article itself. NY. cardholder's name. “The Impact of Asset Stripping on the Cost of Deposit Insurance. and students. McGraw-Hill. FAX 813-974-3318. and synthesis).   Cite references in the text by citing the author(s) name(s) and then the year of publication in parentheses. Any questions about disk preparation should be directed to the Managing Editor. If you prefer to pay by credit card. 1994.” Financial Practice and Education 4 (No. and Charles W. “A (Relatively) Brief History of Finance Ideas. FL 33620-5500. and expiration date. NY. practical. at the end of the manuscript.         Manuscripts are evaluated anonymously. The Handbook of Financial Engineering.” in D. Attach a separate cover page that includes the title. year. Weston. When submitting a paper for review. Submission Procedure: Manuscripts should be submitted electronically as a PDF file at www. credit card number.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. Place references in an unnumbered. Examples of references are provided below: References Baldwin. Myers. Authors should ensure that their identity does not appear under the “Properties” of the file they are submitting.. Label figures at the bottom. season/month. Each paper must be accompanied by a submission fee for manuscript evaluation: $100 for FMA members. Double space the text with ample margins. New York.edu .e.. double-spaced. Jr. and education (wellmotivated.H.” Harvard Business School Working Paper 92-053 (December). Smithson. Provide all relevant publication information available (i. please provide supplemental sheets showing all steps in algebraic derivations so that the reviewers do not have to re-create them.. TEL 813-974-2084. Tables and figures should appear on separate pages labeled in numerical order and grouped at the end of the text. and title and affiliation of each author to one copy of the article. Place derivations and proofs in an appendix.S.fma. please include in your submission email the following information: type of credit card. survey. Clifford W.usf. place all footnotes. Chew Jr. Carliss Y. If paying by check. Master Tax Guide. city and state. “Finance Theory and Financial Strategy. scientifically sound manuscripts that represent major contributions to the field of financial education). 1994 U. accompany them with a clear explanation. 1993. one-paragraph (approximately 100 word) abstract of the articlethat provides a brief overview of the paper. Ed. $200 for non-FMA members. clinical studies (characterizations of real world situations using unique sources of data). with the number in parentheses and flush with the right margin. Smith. The non-member submission fees include a one-year membership in FMA for the submitting author. please make checks payable to Financial Management Association. Each equation should be numbered consecutively. 1993. academics as teachers. Authors of accepted papers must supply an MS Word copy of their article. IL. 1991. College of Business Administration. etc. alphabetical list at the end of the manuscript. The journal is devoted to the publication of original manuscripts that are accessible to a broad audience. author(s) full names. 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. J. Avoid tedious mathematical expressions. including practitioners. Include marginal notation in the article for the approximate placement of all tables and figures. E-mail:cconno@coba. University of South Florida. Fred..). When preparing accepted papers. Commerce Clearing House. The cover page of the manuscript should show only the title of the paper and a short. Financial Management Association International.