I have no idea how long the bear market on Wall Street will last, or how far down the major averages will go. But I do know what won’t help you forecast the bear market’s severity: valuation. Bear markets that occur when the stock market is more overvalued do not last longer, on average, than those that begin when equities are less overvalued. Nor do they produce bigger losses, on average.
That should come as a relief, since the stock market currently is either overvalued or extremely overvalued, according to almost any valuation indicator with a solid long-term track record.
These are the conclusions I reached upon analyzing all U.S. bear markets since 1900, according to a calendar maintained by Ned Davis Research. I determined where the stock market stood at the beginning of each bear market,according to eight well-known stock-market valuation models. I’ve written before that each of these eight historically has exhibited a statistically significant ability to forecast the stock market’s subsequent 10-year inflation-adjusted total return.
I measured the correlation between these models’ readings and the length and severity of the bear markets, and came up empty at traditional levels of statistical significance. Moreover, to the extent that correlations did emerge, they usually were just the opposite of what you’d expect: bear markets that began when equities were more overvalued were shorter, on average, and produced smaller losses.
My findings are not as unexpected as they might appear. Think back to the early-2020 bear market that accompanied the beginning of the COVID-19 pandemic. Despite the U.S. stock market then being more overvalued by almost every measure than at the start of any other bear market of the past century, the decline lasted just over a month.
“ Bear markets in recent decades have been shorter and less severe.”
The pandemic-induced bear market is consistent with the historical pattern: bear markets in recent decades have been shorter and less severe, on average, than those in the early decades of the 20th century. At the same time, valuations overall have become more stretched.
What about inflation? Surely bear markets are more severe when they begin when inflation is higher? Once again, no. I l divided the past 120 years’ worth of bear markets into two groups according to the CPI’s 12-month rate of change when those bear markets began. There was virtually no difference in those two groups’ average bear-market loss.
Great care must be exercised in interpreting these trends, since correlation is not causation. The proper conclusion to draw, in my opinion: valuation indicators tell us little about the stock market’s prospects over the next year or two. Thankfully, the typical bear market is relatively brief — 1.4 years is the average of those in the Ned Davis Research calendar.
Valuation indicators instead are most helpful in forecasting the stock market’s long-term return — over the next 10 years, for example. So the market’s cumulative prospects between now and 2032 are well-below average. But that doesn’t necessarily mean the current bear market will be especially long or severe.
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at email@example.com