Huge losses can be generated by bear markets without daily collapses.
Though it doesn’t make it more likely, more than half of Americans believe a U.S. stock-market meltdown is coming.
Actually, crash anxiety is a contrarian signal—that is, the stock market performs better when investors are more concerned about a crash than when they are quite complacent. If the developing investor consensus were that a crash was improbable, I would be more concerned.
Recent research by Allianz Life of a “nationally representative sample of 1,004 respondents age 18+” reveals how many investors are concerned about a crash. “More than half (51%) worry that another major market crash is on the horizon,” the insurance company’s 2025 Q1 Quarterly Market Perceptions Study states.
Analyzing poll data gathered since 2001 by Yale University professor Robert Shiller helps us to understand that crash worry is a contrarian signal. “What do you think is the probability of a catastrophic stock-market crash in the U.S., like that of October 28, 1929, or October 19, 1987, in the next six months?” asks one question on every monthly poll.
What I discovered is plotted on the above chart. After months in which crash risk is unusually high, the S&P 500’s SPX total-return index averages better than after months in which that risk is judged to be especially low.
It is impossible to know whether Shiller’s poll would coincide with the Allianz Life study indicating current high levels of crash worry. Shiller’s poll is provided with a three-month lag, hence the latest reading comes for December. But based on the numerous emails I receive from readers, I can confirm that fear about a possible stock-market catastrophe is absolutely higher than it has been in years.
Why would a contrarian indicator be crash anxiety?
One reason collapse anxiety is a solid contrarian indication is that, far lower than the subjective probabilities investors evaluate even when they are optimistic, the actual likelihood of a stock-market fall in the next six months is quite low. Investors expose a lot about their emotions since their views have so little bearing on reality.
Thanks to work by Harvard University’s finance professor Xavier Gabaix, we know the objective probability of a crash. The model he and his colleagues built indicates that the following six months have barely a 0.33% risk of an October 1987-magnitude one-day collapse (a 22.6% decline, exactly). That contrasts with the 51% of Allianz Life survey respondents who think a crash is “on the horizon.”
As I pointed out in a recent piece, the emphasis on this study is on one-day plunges rather than the long-term prospects of the stock market, which are really dismal. But a severe bear market can develop without the stock market seeing any significant one-day declines.
Consider the bear market that followed the March 2000 fall-off in the internet boom. The lowest one-day return of the S&P 500 throughout that fall was a loss of “just” 5.8%. With the global financial crisis of 2008–09, the benchmark’s worst one-day return was a 9% loss. Clearly, bear markets may cause large losses without causing one-day collapses along the way.
