Historical data shows that volatility in the U.S. stock market is indeed slightly higher in October. However, experts point out that statistical phenomena lacking a reasonable explanation may be driven by market psychology, and investors should focus more on fundamentals rather than seasonal superstitions.
The so-called ‘October curse’ in the U.S. stock market currently appears to be unfounded, as there is no clear evidence suggesting that U.S. stock market volatility in October would be higher than in September.
Marketwatch analyst Mark Hulbert summarized monthly data based on the Dow Jones Industrial Average since 1896. He found that the standard deviation of the Dow in October each year was significantly higher—actually 21% higher than the average of the other 11 months.
Hulbert noted that while part of the above-average volatility in October stems from the fact that it was the month of two major market crashes—those of 1929 and 1987—even after removing these two years (along with the exceptionally volatile October at the peak of the 2008 financial crisis), October remains the most volatile month, albeit with a reduced magnitude of 11% instead of the original 21%.
Hulbert pointed out that the latter remains statistically significant at the 95% confidence level commonly used by statisticians. If statistical significance alone were sufficient grounds for investment decisions, we would undoubtedly have solid reasons.
Unfortunately, statistical significance has never been a sufficient condition. Financial history is replete with statistically significant yet meaningless spurious correlations. For those who remain skeptical, one may refer to David Leinweber’s article ‘Stupid Data Mining Tricks: Overfitting the S&P 500,’ published in The Journal of Investing. The research by the former director of the Center for Innovative Financial Technology at Lawrence Berkeley National Laboratory profoundly reveals the pitfalls of data mining.
Hulbert believes that any investment decision must meet a dual condition: it must be both statistically significant and theoretically consistent. Regarding October volatility, we must ask: Is there a reasonable explanation for why the stock market tends to be particularly turbulent in this month?
He has yet to discover any theory that stands up to scrutiny. The most frequently cited explanation relates to the impact of the 1986 Tax Reform Act—which required mutual funds to use October 31 as the end of their tax year and distribute 98% of capital gains to shareholders. Many believe that the large-scale portfolio adjustments triggered by the tax reform led to October volatility.
However, if this explanation holds true, the volatility in October after 1986 should have significantly increased, but the data shows the opposite result: the statistical support for the post-tax reform period is slightly weaker than in prior years.
Another perspective argues that there is greater economic uncertainty in October. For instance, this month marks the beginning of the final earnings season of the year, and Wall Street analysts typically base their full-year forecasts for companies on the October earnings reports. While this explanation seems plausible, it lacks empirical support.
The Economic Policy Uncertainty Index, compiled by Scott Baker of Northwestern University, Nicholas Bloom of Stanford University, and Steven Davis of the University of Chicago, shows that since 1900, the uncertainty index for October has been 3% lower than in other months.
Ultimately, there may exist some yet-to-be-discovered theory that explains the volatility in October. However, until conclusive evidence emerges, all related discussions remain mere market noise. Investors should focus on corporate fundamentals and macroeconomic trends rather than obsessing over seasonal statistical illusions.
Editor/Doris