At the zenith of the dot-com craze when Palm Inc. spun off from parent 3Com Corp., its March 2, 2000, IPO touched off a stampede of over-eager investors elbowing their way into the action. When the dust settled at the end of the day, Palm’s worth surpassed its parent, even overtaking such titans as Ford Motor Co. and General Motors Corp. Stock market watchers scratched their heads and wondered why investors didn’t run up 3Com stock as a cheaper route to Palm.
“3Com owned 94 percent of Palm,” points out Ming Huang, an associate professor of finance at Stanford GSB, who has conducted research in the emerging field of behavioral finance. “Yet Palm was trading so high right after its IPO that 3Com’s value was much lower than Palm’s. That didn’t make any sense.”
In academia, financial economists were hard put to explain this and other overheated stock behavior in the frenzied months leading up to the bursting of the dot-com bubble. The highly irrational pricing of Internet stock — especially at IPO — did not square with the traditional approach to the study of financial markets, which assumes that markets are always efficient and participants always rational. The traditional view that “smart money” in the market will take advantage of mistakes and human foibles and thereby drive prices back to equilibrium could not explain the collective faulty judgment of a broad base of investors and their impact on stock prices.
“Financial economists have been aware for a long time that in laboratory settings, humans often make systematic mistakes and choices that cannot be explained by traditional models of choice under uncertainty,” says Paul Pfleiderer, the William F. Sharpe Professor of Financial Economics and codirector of the Business School’s Financial Management Program. “Until recently, most researchers believed these mistakes or deviations from rationality did not have a significant impact on pricing in the financial markets. Now many of us are willing to entertain the notion that some of the behavioral phenomena uncovered in laboratory settings may affect pricing in financial markets.”
The question is: Which behaviors? And more important: How do they affect pricing? These are the issues at the core of behavioral finance.
“The tag ‘behavioral finance’ is stuck to anything that’s an unorthodox approach to finance,” says Harrison Hong, an associate professor of finance and a leading voice in the rise of the field’s popularity. “The good news is that behavioral finance is at the heart of what finance is all about, and that is: What drives stock prices? What drives investors’ decision making? The price part is the key. But that step between investors’ psychology and stock prices is a very big, hard leap to make.”
Hong and colleague Huang have been perhaps the most visible agents provocateurs of the finance status quo. “They are at the forefront in developing rigorous models that will give us insight into how deviations from our standard assumptions might have important effects on pricing,” says Pfleiderer.
While Hong’s research has studied asset pricing from the angle of collective investor beliefs, Huang’s research has examined the impact on pricing of investors’ preferences toward risk.
“The research my coauthors and I worked on proposed to explain an array of pricing phenomena, or anomalies, by merging the traditional asset pricing approach with some evidence from psychology literature of human beings’ preferences when faced with uncertainty — what’s called the prospect theory,” says Huang. Developed by the late Stanford psychologist Amos Tversky and Daniel Kahneman, the theory demonstrates, among other findings, that people care about changes in financial wealth rather than absolute value, and that they are more sensitive to losses than to gains relative to certain reference points. “For example, when we invest in a stock we tend to remember at what price we bought it,” he says. “That affects how we feel toward future risk-taking. In particular, we may be less afraid of stock market risk after having accumulated a lot of prior gains — even if there is a small drop in the stock market, we will still be ahead overall.”
Huang and his collaborators specifically studied the effects of such preferences on asset pricing. They showed that their framework can help explain why stock investment returns historically exceed interest rates by 6 percent, and why price-earning ratios tend to fluctuate more widely than can be justified by cash flow news. They also have applied this framework to study the predictability of individual stock returns.
Hong and Jeremy C. Stein of Harvard University have modeled and measured the collective effects of investors’ beliefs on asset prices. “Rather than saying, ‘investors make errors,’ we know it’s more accurate to say there are different degrees of rationality,” says Hong. “Some people are very rational — the professional money or hedge fund types — while others have limitations in their formation of accurate expectations. What happens when you put these people together and you look at the equilibrium prices that result? My research explores that interaction.” Specifically, his research shows that simple models of asset pricing featuring these interactions can efficiently reconcile a number of stock price phenomena such as momentum in stock prices, the overvaluation of IPOs such as that of Palm Inc., and market crashes.
Hong’s 1999 study on stock return momentum suggests that past winning stocks continue to perform well, and past losers continue to perform poorly because prices adjust too slowly to information about the company. The research also found that bad news about a company travels slowly and has a less immediate impact than good news on stock prices.
His research in 2000 found that when investors have different opinions about stock valuation and some are short-sales constrained, breadth of ownership in a company’s stock is a valuation indicator. In other words, an increase in the number of investors in a stock demonstrates agreement among a broader base of opinions and forecasts higher returns. Conversely, a reduction in the number of investors in a stock forecasts lower returns. An earlier, related study found that within a six-month period, heavy trading volume in a firm’s stock signals intense differences in opinion among the firm’s investors and presages the stock’s crash.
Hong’s most recent research tries to understand what determines investors’ beliefs or opinions. It tests the idea that stock market participation is influenced by social interaction, or the effects of investors’ peer groups. Using data from the Health and Retirement Study administered by the University of Michigan’s Institute for Social Research beginning in 1992, Hong and his fellow researchers found that sociable households that either know their neighbors or attend church are substantially more likely to invest in the stock market, all other factors being equal. Moreover, sociability is stronger in geographic areas where stock market participation rates are higher.
“There is a lot of evidence on prices that indicate markets are predictable,” says Hong. “Now we’re taking steps toward bridging the gap between hypothesis and rigorous, empirical testing. I think that’s what behavioral finance is about.”
For media inquiries, visit the Newsroom.
