赢家的诅咒

By Michael Peltz

Richard Thaler lifts his wineglass, swirls its contents-a 1990 Chateau Cos d'Estournel-and takes a slow sip before setting It down next to a glass containing the 1989 vintage. An empiricist by nature, Thaler has arranged this wine tasting for fellow economists Andrei Shleifer, Kent Womack, and Luigi Zengales over dinner at a trendy Chicago restaurant on the eve of the National Bureau of Economic Research's behavioral-finance meeting. Most of the meal Is spent discussing their latest research, but at this moment, the four academics are focused on claret. "The two wines clearly come from the same vines," says Thaler, the 53-year-old Robert P. Gwinn professor of behavioral science and economics at the University of Chicago Graduate School of Business. "The more recent vintage, however, seems to have matured very quickly."

The same could be said of behavioral finance. in less than two decades, the study of how investor psychology affects markets has captured the imaginations of some of the world's finest young economic minds (Including those wine tasters, ages 35 to 42), made its way into business-school curricula, influenced the strategies of professional investors, and given conniptions to the economic old guard. Mostly this is due to Dick Thaler, the leader of the behavioral pack. Says Stanford University finance professor and Nobel Prize winner William Sharpe, "He's the one who was out there in our midst, just pushing, pushing, pushing."

In December, Thaler even began putting his money where his mouth is by offcially joining the ranks of professional money managers. He and longtime friend (and former academic) Russell Fuller hung out a shingle for Fuller & Thaler Asset Management, based in San Mateo, California. Their goal is to take advantage of quirks in investor behavior that they see as systematic-for example, the tendency of people to overreact to unexpected and dramatic news events concerning a particular stock, driving its price down too low when the news is bad and too high when it's good. The strategy of buying beaten-up stocks and avoiding high- priced glamour ones is a lot like the traditional Graham and Dodd value- Investing approach, but Thaler sees a purely behavioral rationale for why it works.

At least that's the theory. Anything can happen when theory hits the street, as Nobel economists Robert Merton and Myron Scholes learned last summer when Long-Term Capital Management, the hedge fund derived from their economic models, nearly imploded. Behaviorist Thaler observes, "It may be that they were guilty of underestimating some risks."

In the 1970s, when Thaler was writing his Ph.D. thesis in economics at the University of Rochester, there was no such thing as behavioral finance or behavioral economics. Yet he saw quirky, unexplainable economic behavior everywhere. in researching his dissertation on the economic value of a human life, he discovered that people would pay only $200 to slightly improve their odds of living yet they'd want thousands of dollars in payment to slightly increase their odds of dying. Traditional economics says that the willingness to pay and the willingness to be paid shouldn't diverge by much.

As an assistant professor at Rochester, Thaler also observed some highly curious economic behavior on the part of the faculty. Most of them chose to receive their salaries over 12 months rather than 9, even though that made far less economic sense. Then, in 1977, he met the israeli psychologists Daniel Kahneman and Amos Tversky. They introduced him to behavioral psychology and their seminal work on "prospect" theory-which in essence states that people's aversion to loss is about twice their desire for gain. in 1978, Thaler moved to Cornell University and began applying his newfound knowledge to economics.

His theories were a controversial departure from the dominant thinking of the time, led by Nobel winner Milton Friedman and other effcient-market theorists who operated out of the University of Chicago. They believed that human behavior was governed solely by the principles of rationality and self-Interest. Markets were effcient because rational investors couldn't help but drive the prices of securities toward their intrinsic value as they tried to maximize their own wealth. in a rational world, future prices were unpredictable and changed only when there was genuine news.

The heretical Thaler tended to view life and markets through a psychological prism, a throwback to economics before World War II. in the 1930s, John Maynard Keynes likened investing to the newspaper contests of his day in which competitors would be asked to pick the 6 prettiest faces from 100 photographs. "It is not a case of choosing those which, to the best of one's judgment, are really the prettiest," wrote Keynes, but "anticipating what average opinion expects the average opinion [of the prettiest to be]." Thaler thinks Keynes's beauty contest says a lot about how people should look at markets: "Good investing must combine good analysis and good psychology."

Despite the critics-University of Chicago Nobel laureate Merton Miller has deemed behavioral finance "a fad that doesn't have anything to offer"; Chicago finance professor Eugene Fama calls it "dredging for anomalies"-Thaler was gaining admirers. in 1992, he published The Winner's Curse, which, among other economic irregularities, describes the tendency to bid too aggressively in an auction as the number of competing bidders increases and why gamblers go for long shots at the end of a bad day.

Three years later, in 1995, the citadel of effcient theory itself offered him an endowed chair, despite opposition from Miller, now finance professor emeritus there. As part of the enticement, the university also offered a professorship in marketing to Thaler's fiancée, France Leclerc, who had been teaching at MIT. (They married a few months later, 35 feet underwater in HawaII.) "It speaks to the school's ecumenical views that it is willing to tolerate a guy who is not in the church," says Sherwin Rosen, Thaler's Ph.D. adviser at Rochester, who has since joined Chicago's economics department. Rosen is himself somewhat skeptical of behavioral finance but is impressed with Thaler's success. "Dick seems to be passing the market test," he says. "His stock looks very hot right now."

Acceptance into Chicago's business school was a turning point for Thaler and behavioral finance. Bright young students who came to Chicago to learn hard-core finance could now also sign up for Thaler's elective course, called Managerial Decision Making and described in the curriculum guide as "only recommended for those students who expect to have to make decisions during their careers." Thaler begins the class with an experiment that's a takeoff on some famous behavioral research: a study that found that 90 percent of Swedish drivers consider themselves better- than-average drivers-a mathematical impossibility. For his lesson, Thaler asks students to anonymously write down where they think they'll rank in the final grading. Of 125 MBA students taking this course last fall, not one thought

he or she would finish in the bottom half. "Obviously, half of them were wrong," quips Thaler. (Students typically say they'll finish in the second decile.)

These experiments speak to overconfidence, which is one of the pillars of behavioral finance. Study after study shows that people tend to overestimate their abilities and their knowledge. "We all think we are pretty good at sizing things up, just the same way we all think we are good judges of character," explains Thaler. "We should know that we really are all hopeless. When some terrible crime is committed and the neighbors are interviewed, they always say what a nice guy the alleged perpetrator was and how he couldn't have done it. The fact is, we can't tell if the guy is a serial killer or not. We also have a hard time telling a good stock from a bad one, though we think we can."

Investor overconfidence may explain what is sometimes referred to in economics as the Groucho Marx theorem. Just as Groucho wouldn't belong to any club that would have him as a member, a rational investor should be reluctant to be part of any trade where someone is willing to take the other side. After all, if the investor on the other side of the trade is also rational, he or she must have a good reason for buying or selling. According to Thaler, the Groucho Marx theorem should result in very little trading, but of course that is not the case.

Thaler's most recent paper, delivered at last October's NBER meeting, looks at asset-allocation strategies used by investors in defined- contribution retirement plans. "We discovered that many investors have incredibly naive ideas about diversification, the most extreme of which is what we call the 1/n heuristic," he says. "Basically, people tend to divide their contributions evenly among all the funds offered in their plan." Sound familiar? Although heuristics, or mental rules of thumb, are often useful in making decisions, sometimes they can be misleading, even harmful.

Eugene Fama, Thaler's friendly rival at Chicago-they play tennis together; Fama usually wins-argues that effcient-market theory actually predicts that there will be some anomalies but that they are generated by chance. "I'm willing to stipulate that investors don't always behave rationally," he says. "The issue in my mind is to what extent does irrational behavior affect pricing, and it's dIffcult to find any evidence that it does. That's what all the fighting is about."

Thaler knows he can't convert Fama, but he also understands that behavioral finance needs a mathematical theory to supplant the effcient- market hypothesis. "We're still at the stage of making fun of the geocentric model and having some theories for some facts, but we don't have a complete theory of the world," he says. "Maybe there can't be one. There certainly won't be anything as neat and tidy as the planets all circling around the sun or that everyone is rational." Thaler himself isn't likely to play a big mathematical role, but his student disciples, now teaching in business schools across America, are on the case. "Dick is more interested in the ideas than in the techniques needed to solve the problem," says Kent Womack, a former student who teaches at Dartmouth's Tuck School. "His one weakness is that he's not particularly strong in statistics or mathematics, so we do the numbers work and he thinks the big thoughts."

It might seem as if Thaler's observations couldn't find practical application within the markets, but they do. indeed, Richard Bernstein, chief quantitative strategist at Merrill Lynch, says behavioral finance is redefining the basic rules that drive the investment game. "We don't advertise it as such, but nearly everything we do in our group-our asset- allocation models, our stock-sector selection models-Is based on behavioral finance," he says. "One of the most important things in any investment strategy is to

understand the frame of reference with which people are approaching the markets, and behavioral finance helps you do that."

Thaler has every intention of taking advantage of people's frames of reference in his new money- management venture, which evolved from RJF Asset Management, a firm that had been solely owned by Fuller. "Everybody is trying to buy cheap stocks," says Thaler, who had been an adviser to RJF but is a full principal in the new firm. "In a sense, we're trying to do that, too, but we're trying to distinguish between ones that are cheap and ones that are dogs by looking explicitly for signs of systematic errors on the part of the analysts." Their firm already manages $370 million in mostly institutional assets and has five years' experience using behavioral finance. it also has a small- to mid-capitalization mutual fund, the Behavioral Growth Fund, which earned a 23 percent re- turn in its first year, and in January the firm planned to launch two new funds under the Undiscovered Managers Fund family. "Russ invests using concepts that are near and dear to my heart," says Thaler. "If I can't help improve that, there's something wrong."

Maybe that's overconfidence. Then again, maybe it's not.

Benefactor