STANFORD GRADUATE SCHOOL OF BUSINESS—Investors who lived through the dot-com bubble are unlikely to forget it. But although the still-raw details of that era have been documented in countless publications—and are likely to be ensconced in MBA textbooks for decades to come—those who didn't experience it will probably not learn from history.

That's because intellectual knowledge of the risks of over-hyped and overvalued stocks doesn't seem to do investors much good when it comes to avoiding future speculative disasters. Instead, first-hand experience of a bubble appears to be necessary to make investors wary of returns that seem too good to be true.

"After a crash, people are quite aware of the potential dangers and adjust their investing behavior accordingly. But after 20 years or so, that memory is erased, and you again have the conditions that make another bubble possible," says Stefan Nagel, an associate professor of finance at the GSB who, with coauthor Robin Greenwood, an assistant professor of finance at Harvard Business School, has written a paper titled "Inexperienced Investors and Bubbles."

Nagel was interested in how investors formed expectations for the future, as well as in market conditions that lead to market bubbles. "In this study, we were trying to connect those two things," he says.

Although there have been some theoretical studies and laboratory experiments examining these topics, "there is a big gap between these experiments in artificial laboratory settings and the empirical research we performed," says Nagel, who teaches the core finance class to MBAs, as well as a course in empirical finance in the PhD program.

In the study, Greenwood and Nagel examined the portfolio decisions of experienced and inexperienced mutual fund managers during the technology bubble of the late 1990s. They used data from Morningstar, a provider of mutual funds research and ratings that maintains a database of funds and profiles of their managers.

By categorizing fund managers by age, Greenwood and Nagel found that at the start of the bubble in 1997 younger managers bet on technology stocks at rates that actually trailed those of older managers. But leading up to the bubbles' peak in March 2000, younger managers dramatically increased their holdings of such stocks.

"The funds by younger managers took aggressive positions in technology stocks—much more than those by older managers," says Nagel. Particularly after quarters when technology stocks had done well, younger managers shifted their holdings further into technology stocks. "Young managers chased the trend, but older managers did not," Nagel says.

And as the bubble developed further, new money was flowing into the funds run by the younger managers. This seemed to be driven by the fact that investors tend to "chase" funds that have been performing well. "At the beginning of the bubble, younger managers tended to perform better than older ones, thus attracting more investors to their funds and perpetuating the cycle," says Nagel. The result was that although younger managers started out in 1997 with relatively small funds, the amount of money they controlled had quadrupled by March 2000 when the bubble finally burst.

After the bubble ended, younger managers experienced "outflows"—investors taking money out of their funds—but, interestingly, not more so than comparable funds run by older managers, despite the worse overall performance by younger fund managers. "During the upturn, there were substantially more inflows for younger rather than older managers," says Nagel. "But after the crash, younger managers didn't necessarily experience more outflows. Thus for mutual companies, the failed dot-com investments of younger managers turned out to be not that costly," says Nagel. "For those investors who piled into young managers' funds before the peak of the bubble, the picture looks different, of course."

The findings of the study point to a difference in beliefs of younger and older fund managers. "It seems that older managers were more skeptical about how well these technology stocks would do in the future," Nagel says. That might surprise economists who, in theory, believe there should be no difference whether a fund manager has first-hand experience or not. Knowledge is what matters, regardless of how it has been acquired. "And after all, even if you're a younger manager, you should still have learned about past bubbles and what happened in the stock market over the past 100 years. In fact, MBA programs teach this," Nagel says. "But it turns out that on-the-job experience is what counts."

Nagel currently has another related project in progress with Ulrike Malmendier, an assistant professor at the University of California, Berkeley. They are analyzing the actions of individual investors, examining how their investments in stock versus other assets correlates with their personal experience. The researchers see evidence that personal experience dramatically impacts investment decisions. "If someone lived through the Great Depression, or was young in the 1970s or 1980s when stock market performance was very poor, and experienced lousy returns, he or she is much less likely to put money in stocks," says Nagel. "Almost inevitably, they choose other assets to invest in."

The implications of the study are a bit ominous. "Go out 20 or 30 years from now, and it's possible we'll see another bubble," Nagel says. Is there a moral here? Always invest in funds with older managers? Not necessarily. "After all, if you had ridden the tech boom for the short run only, you would have done very well," Nagel says. Over the long term, of course, it would have paid off to have gone with an experienced manager. "So it's very difficult to generalize," he says.

Related Information

"Hedge Funds and the Technology Bubble," Markus K. Brunnermeier and Stefan Nagel, Journal of Finance, 2004.