MW Why Nvidia, Broadcom and other highflying stocks are especially risky now

By Mark Hulbert

Will the market crash in October? Unlikely. But momentum stocks in highflying sectors are vulnerable to a downturn

Crash odds are elevated when an industry or sector outperforms the market significantly.

October doesn't deserve its reputation as the month when stock-market crashes occur. Though the two worst crashes in U.S. history happened in that month - 1929 and 1987 - the risk of a marketwide crash is no greater in October than in any other month.

Investors worried about a broad market crash in October can perhaps take a bigger sigh of relief. The same cannot be said of momentum investors who are betting on industries that have recently performed the best. A few of their favorite sectors are especially vulnerable to a crash right now.

I base these seemingly contradictory observations on two different academic studies of crash risk. The first focused on marketwide crashes and determined that there's just a 0.06% probability that, at any time in October, there's a one-day crash as deep as that in 1987(a drop of 22.6% in the Dow Jones Industrial Average DJIA). This minuscule probability is no different than what it would be for any month of the calendar.

This isn't to deny that October appears to have above-average crash probabilities. But this appears to be "a statistical fluke," Xavier Gabaix, the Harvard University professor who led this study, said in an email. "There's no special reason, as far as we know, for why [crash odds in] October might be worse" than in other months.

If you have trouble with Gabaix's assertion, consider a different way of measuring downside risk: the probability that a major bear market begins in a particular month. To calculate these probabilities, I turned to the bear-market calendar maintained by Ned Davis Research. As the chart above shows, the number of bear markets since 1900 that began in October is well below average. It's difficult to square this result with the belief that the market in October is more vulnerable to a crash.

Vulnerable stocks and sectors

The heightened crash risks that these industries face applies to the next two years.

Individual industries and sectors can still have elevated odds of crashing even when the overall market does not. According to another Harvard study, these odds rise whenever an industry or sector outperforms the market by a particularly large amount.

The study, entitled "Bubbles for Fama," defined a crash as a drop of at least 40% over two years. The researchers found that the more an industry or sector outperforms the market, the more likely it is to crash.

To illustrate, consider an industry that over a trailing two-year period outperforms the S&P 500 SPX by a cumulative 100 percentage points. Based on data back to the 1920s, the study determined that such an industry has a 53% probability of crashing. When an industry has outperformed the market by 125 percentage points over the trailing two years, crash probability rises to 76%, and 80% when its trailing two-year return is 150 percentage points above the market.

These findings should be especially worrying to investors in the "independent power producers & energy & traders" industry. It has beaten the S&P 500 by more than 200 percentage points over the trailing two years. Stocks in this index include AES Corp. $(AES)$, Talen Energy Corp. $(TLN)$ and Vistra Corp. (VST).

Another industry facing heightened crash risk is "semiconductors & semiconductor equipment." It has outperformed the S&P 500 by 128 percentage points over the past two years. The largest stocks in this popular industry include Nvidia Corp. (NVDA) and Broadcom Inc. $(AVGO)$.

Note that the heightened crash risks that these industries face apply to the next two years. In any case, be mindful of the contrast between an 80% probability of a crash at some point in the next two years (which applies to the power-producers industry) and a 0.06% that the broad market on any given day in October will fall by as much as it did in October 1987.

This isn't the first time I've written about the "Bubbles for Fama" study about industries' crash risk, and in only some instances did a crash subsequently occur. My most recent mention came in March, when I focused on the risks faced by the semiconductor industry XX:SP1500.4530. A few weeks after that column appeared, this industry was 21% lower - a bear market but not a crash as defined by the authors of this study.

But the study's authors didn't find that crashes are guaranteed to happen when an industry has outperformed the market by huge amounts - only that the risk is higher.

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 mark@hulbertratings.com

Also read: What to know about October's market-crash potential - and what you can ignore

More: ETFs that protect against 'painful' stock-market drops are attracting worried investors

-Mark Hulbert

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

October 01, 2025 08:40 ET (12:40 GMT)

Copyright (c) 2025 Dow Jones & Company, Inc.