Many investors worry that the stock market’s current overvaluation means that the next bear market—whenever it begins—will be a long and deep one.

It might surprise and comfort investors, though, to learn that there is little historical correlation between stock valuations and the length and duration of bear markets.

Consider that the longest stock bear market since 1900 began when stocks were far less overvalued than they were at the start of the shortest bear market. I’m referring to the bear markets that began in September 1939 and July 1998, according to a calendar maintained by Ned Davis Research. The former lasted nearly three years while the latter lasted less than two months.

At the beginning of that longest bear market, the cyclically adjusted price/earnings ratio, or CAPE—the one made famous by Yale University finance professor and Nobel laureate Robert Shiller that is based on average inflation-adjusted earnings over the trailing decade—stood at 16.45. That is below the CAPE ratio’s long-term average, suggesting an undervalued stock market.

At the beginning of the shortest bear market, in contrast, the CAPE ratio was more than twice as high, at 38.3. That reading was higher than all prior readings back to 1871, according to data from Prof. Shiller, suggesting a severely overvalued market. And yet the ensuing bear market lasted less than two months.

Note that I’m not including the bear market that began in early 2020 in this analysis, on the theory that it was caused by pandemic-induced lockdowns having nothing to do with stock-market valuations. But including it would have made my point even stronger, since it was even shorter than the bear market that began in July 1998, and when it began the CAPE ratio was even higher—indicating even Note that I’m not including the bear market that began in early 2020 in this analysis, on the theory that it was caused by pandemic-induced lockdowns having nothing to do with stock-market valuations. But including it would have made my point even stronger, since it was even shorter than the bear market that began in July 1998, and when it began the CAPE ratio was even higher—indicating even greater overvaluation.

The unreliability of forecasting the length of bear markets based on valuations can’t be traced to some defect in the CAPE ratio. Each of the seven other well-known and highly regarded valuation indicators I studied were equally unreliable. These seven, each with impressive long-term forecasting records in their own right and about which I’ve written before, are the ratios of price to earnings, price to sales and price to book value, the dividend yield, the Buffett indicator (the stock market’s total market cap divided by gross domestic product), the “Q” ratio (calculated by dividing market value by the replacement cost of corporate assets), and the average investor’s equity allocation.

Share Your Thoughts

Have you changed your investment strategy to prepare for a bear market? If so, how? Join the conversation below.

Consider what I found when, for each of these indicators, I segregated the more than three dozen bear markets since 1900 into two equally sized groups, according to the indicator’s level when the bear market began. One group had valuations below the median and one had valuations above it. In no case was there a statistically significant difference between the average length of the two groups’ bear markets. In more than half the cases, in fact, the bear markets were longer when their valuations were lower at their beginning.

By the way, I reached the same conclusion when focusing on the magnitude of bear-market losses rather than their duration. To illustrate, contrast the 1973-74 and 2007-09 bear markets, which were equally punishing in terms of the losses incurred by the major stock-market indexes. The CAPE ratio at the beginning of the former was 18.7, versus 27.3 at the start of the latter.

Valuations still matter

None of this means valuations don’t matter. Indeed, each of these eight indicators has historically been correlated in a statistically significant way with the stock market’s subsequent 10-year return. The universal message of all eight currently is that the stock market’s prospects over the next decade are at best mediocre—if not much worse. That’s because all eight currently are more overvalued than at least 90% of the time in U.S. history.

But forecasting a mediocre 10-year return is different than predicting the length or duration of any bear markets that occur along the way. After all, there’s more than one way for the stock market to produce mediocre 10-year returns. For example, it could do that while avoiding a bear market altogether, if it went nowhere in each of the next 10 years. It also could experience a punishing bear market followed by an equally strong recovery.

So what does factor into the severity of bear markets, if not valuations? Often, they can be attributed to a loss of confidence on the part of the investing public. Predicting when that confidence will begin to fade, and how long it will take before it can be restored, is as much a matter of psychology as it is valuation.

This was the key argument Prof. Shiller made a couple of years ago in his book “Narrative Economics: How Stories Go Viral and Drive Major Economic Events.” By “narratives,” he means “the stories, particularly those of human interest and emotion, and how these change through time.” Ironically, for someone who became famous for the CAPE ratio, Prof. Shiller contended that “we have to consider the possibility that sometimes the dominant reason why a recession is severe is related to the prevalence and vividness of certain stories, not the purely economic feedback or multipliers that economists love to model.”

In an email, Prof. Shiller provided the following illustration: Beginning in the late 1920s, investors became obsessed with the term “brokers’ loans,” which refers to what today we call margin credit.

He writes that in “1928, the year before the 1929 crash, use of the term brokers’ loans absolutely exploded…which gave much of the public an uncomfortable feeling about the stock market and which made the market vulnerable to panic…. I think one can blame the 1929 crash substantially on that narrative.”

The 2020 narrative

For a more recent illustration, consider what stopped the bear market in its tracks in March 2020. Most everyone is inclined to attribute that turnaround to pure economics, referring to the government’s and the Federal Reserve’s massive fiscal and monetary stimulus. But it’s worth remembering that there is no guarantee that flooding the market with liquidity during a downturn will restore confidence. On the contrary, as British economist John Maynard Keynes outlined a century ago, pumping too much liquidity too quickly runs the risk of reducing confidence rather than restoring it—exacerbating the crisis rather than helping overcome it.

With the benefit of hindsight, we today tell ourselves that there was little doubt that the government’s and the Fed’s actions would succeed in creating a new bull market. But with the benefit of Prof. Shiller’s focus on narratives, we instead realize how fragile the bull market is and how vulnerable it is to changes in the way the winds are blowing among investors.

So even though the stock market today is overvalued according to any of a number of time-tested indicators, the next bear market—whenever it occurs—may or may not be particularly long or severe. It will depend in no small part on investor psychology. And that may or may not be a source of solace.

Mr. Hulbert is a columnist whose Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at reports@wsj.com.