The possibility of another crash looms large in the minds of investors but research suggests investors’ fears are overblown.
The stock market crash of 1929 ushered in the Great Depression, and the 1987 crash included the largest single-day decline in the Dow Jones Industrial Average. The possibility of another crash looms large in the minds of investors—but research suggests investors’ fears are overblown.
Yale’s William N. Goetzmann and Robert J. Shiller, along with Case Western Reserve’s Dasol Kim, find that from 1989 through 2015 both individual and institutional investors judged the likelihood of a crash as much higher than shown by historical data.
Since 1989, Shiller, a Nobel laureate, has tracked the judgments of individual and institutional investors on the probability of a severe market crash. “The definition of a crash is specific: a drop in the US stock market on the scale of October 19th, 1987 [-22.61%], or October 28th, 1929 [-12.82%],” write Goetzmann, Shiller, and Kim. The result is startling: the average estimate for the likelihood of a one-day crash over the succeeding six months—controlling for both good and bad economic times—was 19 percent. This remained true for the length of the 26-year study period.
Given that the probability of an extreme stock market collapse on the order of 1929 or 1987 in any six-month period is less than 1 percent, investors gave estimates of the likelihood of a crash that were more than 20 times the historical precedent.
The influence of news media may have contributed to the investors’ bias, the researchers argue. Journalists can “frame recent events through select[ive] reporting—emphasizing negative outcomes.”
The researchers used Wall Street Journal articles to test their theory, measuring the effect of word choice and subject matter on investor expectations of a crash. They parsed the articles for negative terms, such as “crash” and “bad news,” and positive terms, which included “boom” and “good news.” Their data suggest that articles with “crash” and related terminology correlated with higher investor expectations of a stock market crash in the succeeding six months, as measured by Shiller’s stock market confidence survey. Shiller and more recently Yale’s International Center for Finance have since 1989 collected surveys from high-net-worth and institutional investors to inform stock market confidence indices.
Using the insights of two psychologists, Princeton’s Daniel Kahneman and the late Amos Tversky, the researchers make the case that investors fell prey to the “availability heuristic,” a cognitive shortcut most of us use to make judgments based on what is top of mind—or, in this case, top of page.
Individual and institutional investors expected crashes more often than they actually occur. And their expectation determined “such critical things as stock market participation, the demand for insurance against crashes and, to the extent that the investors surveyed are representative of marginal investors, perhaps even the equity premium,” write the researchers. They point out that their findings may provide support for rare-disaster explanations of the equity premium; this historical 6 percent premium, paid to compensate stock-owners for their risk taking, may be related to high assessments on the probability of an extreme stock market collapse.
A Q&A with Chicago Booth’s Dacheng Xiu on the implications of machine learning for the financial markets.
In Finance, Humans Were the First Machines
Chicago Booth’s Sendhil Mullainathan visits the podcast to discuss if AI is really “intelligent” and whether a profit motive is always bad.
Capitalisn’t: Who Controls AI?
Research suggests frictions in the building industry could be slowing or halting projects.
US Construction Has a Productivity ProblemYour Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.