Is the U.S. stock market a poor performer between the two holidays that mark the start of the Jewish new year? In other words, should you “sell Rosh Hashanah and buy Yom Kippur?”
A Wall Street adage indeed suggests that the stock market is weaker over the 10 days these two holidays are observed. Belief in this seasonal pattern has a long history; it was mentioned in the financial press as early as 1915.
Many are asking about the statistical basis of this folk wisdom because Rosh Hashanah begins this Friday evening. To follow the adage, investors would reduce their equity exposure at the market close on Sept. 15 and wouldn’t return to their target exposure until after Yom Kippur ends on Sept. 25.
My analysis of the Dow Jones Industrial Average
since 1900 yields a mixed answer. It’s true that the stock market in the past has declined, on average, between Rosh Hashanah and Yom Kippur. But these two holidays typically occur during September — the worst-performing month of the year.
It’s not clear that the Rosh Hashanah to Yom Kippur stretch is any worse than other days in September. So, as is so often the case, the answer you get from a statistician depends on how you frame the question.
- If you ask whether the U.S. stock market is a worse performer between Rosh Hashanah and Yom Kippur compared to all other periods of similar length throughout the year, the answer is “yes.” And the difference is significant at the 95% confidence level.
- You get the opposite answer if instead you ask: Is the stock market a worse performer between Rosh Hashanah and Yom Kippur than at other times in September? Now the answer is “no,” at least at standard levels of statistical significance.
“ The odds are hardly in your favor if you are interested in betting on this pattern. ”
These results suggest that, to the extent you didn’t already reduce your equity exposure at the beginning of this month, you would be on solid statistical ground reducing it between Rosh Hashanah and Yom Kippur. But you should also know that statistical significance is not the same as economic significance.
Consider that in 48% of years since 1900 the stock market bucked the seasonal pattern and actually rose between Rosh Hashanah and Yom Kippur. While that is below the 56% of all other comparable periods of the year in which the market rose, this 48% is hardly different than the probability associated with a simple coin flip. So the odds are hardly in your favor if you are interested in betting on this pattern in just one particular year.
A good analogy is to a game of Blackjack, where if you’re a good card-counter the odds of winning a particular hand are slightly better than 50%. The key is to play many hands, in which case the card counter has decent odds of coming out a winner.
In the case of the “Sell Rosh Hashanah, Buy Yom Kippur” pattern, however, you get to bet just once a year. The functional equivalent of playing many hands of Blackjack is to bet on this pattern each and every year for several decades.
When you think of it that way, betting on this pattern seems a lot less interesting — no matter how statistically significant it may be.
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