During the COVID-19 Crash, Investors’ Beliefs Didn’t Match Their Behavior

Between February and March 2020, the stock market cratered as the reality of the COVID-19 pandemic came into focus. The Wall Street Journal dubbed Monday, March 16—when the Dow Jones Industrial Average dropped 13% and the S&P 500 nearly 12%—“the day the coronavirus nearly broke the financial markets.”

What were investors thinking during this exceptionally volatile economic moment? A new paper by Stefano Giglio of Yale SOM offers a rare real-time look at how the COVID crash shaped not only beliefs about the market, but also actual trading behavior.The biggest takeaway: beliefs and behaviors don’t always match up as closely as you might expect. Though investors became significantly more pessimistic about one-year stock returns, most didn’t do much about it. “They change their beliefs, but something like 70% of the investors [in our sample] don’t trade,” Giglio says. “Even those that do trade don’t change their portfolios very much.”

Giglio and his co-authors—Matteo Maggiori of Stanford University, Johannes Stroebel of New York University’s Stern School of Business, and Stephen Utkus of Vanguard—gained this surprising insight via surveys of Vanguard investors administered in February, March, and April of 2020, which they matched to respondents’ actual trading behavior.

The researchers launched the survey in 2017 with the aim of understanding how investors’ views of the market interact with their choices. “Very often in economics, we look at what people actually do and try to infer their beliefs,” Giglio says. “But there’s been a recent movement in finance to use surveys to see more directly what’s going on in their minds.”

The survey includes questions about short-term (one-year) and long-term (ten-year) expected stock returns and short-term (three-year) and long-term (ten-year) annualized GDP growth, as well as questions about the likelihood of short-term stock market and GDP disasters—all aimed at gauging how people evaluate both the financial markets and the broader economy, and how those evaluations translate (or don’t) into their portfolios.

Giglio says there are a variety of possible explanations for “the amazing inertia in the translation of beliefs into action” he and his co-authors uncovered. Why, seized by a newly gloomy outlook for the upcoming year, did so few investors change their strategy? “It could be actual costs of trading. It could be mental costs of trading. It could be long-term optimism,” he says. “We don’t know. But it’s certainly fascinating.”

Although the majority of investors made minor or no changes to their portfolios despite what Giglio calls “a general wave of pessimism” following the crash, there was one group that did change their beliefs and behavior more than others: those who had previously been most optimistic. (The researchers identified optimists, pessimists, and “neutrals” by their beliefs in February about expected one-year stock returns.) “Where you see the most correction is from the optimists,” Giglio says.

Of all the survey respondents, optimists decreased their equity shares—that is, the proportion of their portfolios invested in stocks—the most from February to March, by 1.05%, as compared to .98% for neutrals, and .63% for pessimists.

Interestingly, while the investors who’d been the most pessimistic in February did not change their portfolios much in March, more than half of them became, surprisingly, more optimistic about expected one-year stock returns. One plausible interpretation is that February’s pessimists were simply ahead of the curve in seeing how COVID-19 would wreak market havoc. “They certainly seem to feel vindicated,” Giglio observes. Those who predicted a dire future in February “revised expectations upward after the crash. Those people could be saying, ‘I told you the crash was going to come, it came, and now it’s time for recovery.’”

Still, the survey revealed a note of caution: investors became significantly more pessimistic about the prospect of short-term stock market and GDP disasters. In some respects, that’s not surprising, Giglio notes: existing economic models predict that, following a crash like the one that shook the world in March, the perceived likelihood of further economic disasters increases.

But it was somewhat unexpected that investors expected such poor short-term returns following this spring’s jolt. After all, what goes down must come up; expected returns should, logically, increase following a crash.

“People see the market drop and they don’t understand that going forward, in fact, returns are going to be high. That’s a kind of tension,” Giglio says—one that classical economic models don’t predict, but that behavioral economic models, which factor in a hearty dose of human irrationality, do.

While respondents’ opinions on the prospect of short-term recovery varied, a belief in long-term recovery—both to stock returns and GDP—was widely shared by optimists and pessimists alike. In fact, investors’ expectations actually grew modestly over the survey period: in February, they expected ten-year returns of 6.9% per year. By April, that number grew to 7.2% per year. Similarly, 10-year GDP growth expectations increased from 3.1% to 3.5% per year.

Giglio and his co-authors are continuing to tweak their survey in their quest to get to the bottom of the disconnect between beliefs and actions. That friction, he says, “has huge implications for the way we think about why prices move and what happens in financial markets.”