Ready for my annual late-August investment challenge? Find a plausible reason for why September should be the worst month of the calendar for average stock market performance.
This is a timely question because September is just around the corner. Since 1896, when the Dow Jones Industrial Average
was created, the Dow has lost an average of 1.03% in September. That compares to an average gain of 0.76% across all other months of the calendar.
Furthermore, September’s awful record doesn’t trace to just one or two outlier years. In fact, as you can see from the accompanying chart, in all decades but one over the last century September’s performance rank has been below average. That is remarkable consistency.
The reason why you nevertheless need to couple this track record with a plausible explanation before betting on it: Impressive statistics are not a sufficient reason to conclude that a pattern will persist.
As you try to come up with a good explanation, consider the top contenders that have been proposed in response to my challenges in past years. None withstands scrutiny, so please don’t email me to propose them again:
• Investors are more prone to sell than buy when they return from their summer vacations;
• Increased trading volume after Labor Day is bearish;
• Many mutual funds have fiscal years that end on Sep. 30, leading them to engage in window-dressing during the month;
• Investors are forced to sell stocks to pay sky-high tuition bills for their kids’ private schools and colleges.
I’m not holding my breath that a plausible explanation for September’s awful return will materialize. But even if one does exist, there’s yet another reason why you should think twice before aggressively betting that this coming September will be a downer for the stock market: Only by repeating your bet for many years in a row can you be confident that you will come out ahead.
Consider the percentage of Septembers since 1896 in which the Dow rose. Believe it or not, it’s 42%. Because those odds are only modestly below that of a coin flip, they don’t justify a strong bet in any one particular year. Last year, for example, the Dow gained 2.1% in September.
A helpful analogy is to Blackjack. A shrewd card counter can perhaps increase his chances of winning a given hand to 55%. While that’s statistically impressive, it means that his odds of winning when playing a single hand are still hardly better than a coin flip. Only by playing many hands do his odds of coming out a winner become closer to 100%.
So even if the future is like the past, which the absence of a plausible explanation reveals to be a big “if,” you’d still have to bet against September for many years in a row — 30 is a statistical rule of thumb for a necessary number — before your odds become favorable enough to make an aggressive bet. Be my guest if you decide to take that route. Let’s talk in 30 years.
For more information, including descriptions of the Hulbert Sentiment Indices, go to The Hulbert Financial Digest or email [email protected] . Create an email alert for Mark Hulbert’s MarketWatch columns here (requires sign-in).