The stock selloff of the past month is forcing investors to think about whether the market remains too high, and if so, how far it might fall.
One way to answer that question is to see if stocks are unusually expensive relative to their fundamentals and then estimate the declines required to bring them back to more normal levels. I did just that following Monday’s 3% slide in the S&P 500 Index, expecting to find a frothy market packed with overbought stocks. What I found surprised me.
I looked at analysts’ consensus long-term earnings growth forecasts for each company in the S&P 500. This is the estimated earnings growth over the company’s business cycle, which is typically three to five years. I used those growth rates to calculate each company’s future earnings per share—let’s call them long-term earnings—and then used that to calculate a price-earnings ratio for each company—I’ll call it the long-term P/E ratio.
I had to exclude 58 companies for which estimates were not available, but collectively their weighting is just 6% of the index, so most of the S&P 500 by market value is represented in my calculations.
With my long-term P/E ratios in hand, I then had to settle on the “right” ratio. There are many disagreements about where fair value lies, but the market hints at an answer. One clue is that the S&P 500’s average P/E ratio based on forward earnings is 18 times since 1990, the furthest back the data goes. A second is that the average long-term P/E ratio for S&P 500 companies tends to hang around that multiple and is now only modestly higher at 20 times.
With that as background, I decided on 18 times as my bogey for long-term P/E. To my surprise, of the 442 companies I looked at, 63% trade at less than 18 times long-term earnings. Some have a larger weighting in the S&P 500 than others, with heavyweights such as Amazon.com Inc., Meta Platforms Inc., Alphabet Inc., and yes, even Nvidia Corp. on the list, in part because of the recent selloff.
If all those stocks were to rise or fall as needed to approach an 18 multiple of long-term earnings, the S&P 500 would climb 33%. That’s not only because two-thirds of the companies trade below that level but also because, weighted on market value, which is how the S&P 500 is constructed, some of the moves higher would come from the biggest companies.
Let’s focus on the 165 companies that appear to be overbought, among them giants such as Broadcom Inc., Tesla Inc. and Costco Wholesale Corp. If these overbought stocks declined to an 18 multiple, it could well sour sentiment, making it harder for cheaper stocks to fetch higher multiples. But even if just the overbought stocks repriced while the others stayed where they are, the S&P 500 would only slump 9%.
Why does the market seem so overbought when most of the S&P 500 companies are trading at reasonable multiples of long-term earnings? One answer is that headline P/E ratios tend to be based on last year’s earnings or next year’s expected profits rather than longer-term estimates that are almost always higher and therefore result in lower P/E ratios.
A more worrisome explanation is that the market occasionally takes issue with analysts’ earnings estimates. Indeed, some of the expected growth of the big technology companies is built on the hype around artificial intelligence, which is increasingly in question. And if the U.S. economy slows, as now appears to be the case, earnings growth may not be as robust as analysts expect.
And when the market becomes skeptical of earnings estimates, it’s usually not good for stocks. Analysts have been forced to slash estimates in the past, usually after the market signaled its displeasure by sending stocks markedly lower. Notably, analysts slashed estimates for forward one-year S&P 500 earnings by 41% during the 2008 financial crisis and 29% during the Covid pandemic.
If long-term earnings were to be written down, say, 35% across the board, many more stocks would be overbought at current prices. Still, the potential decline in the S&P 500 would only approximate a normal correction. In that event, 69% of S&P 500 companies would trade higher than 18 times long-term earnings. Assuming all stocks were to rise or fall as needed to approach that 18 multiple, the S&P 500 would be down 11%.
But again, a correction in some stocks tends to infect or at least restrain others. If the declines in overbought stocks were not offset by advances among the third of more reasonably priced S&P 500 companies, the market could tumble closer to 27%, resembling a painful bear market.
If earnings come through, the market as a whole may not be as expensive as some of its individual excesses might lead investors to believe, although some repricing is still warranted. For now, analysts are holding firm to their long-term earnings estimates, in part because most S&P 500 companies have beaten expectations for the second quarter. But the market will have its say about the quality of the estimates in the coming days and weeks.
Nir Kaissar is a Bloomberg Opinion columnist covering markets. He is the founder of Unison Advisors, an asset management firm.
This article was provided by Bloomberg News.