People have an understandable interest in patterns in stock market returns. As we head into September, we can expect the inevitable articles about the so-called ‘September swoon.’ If you look at the period since 1926, the average return in September has been negative. A 2011 paper in the Journal of Applied Finance concluded that the historical occurrence of negative returns for the stock market in September is so strong and consistent that it cannot easily be explained away. There are a range of other so-called ‘calendar effects’ in which a specific time of the year, month, or week has historically delivered returns that are markedly different from the average across all periods. There are no conclusive explanations for these effects and, in a rational world, these types of anomalies should not persist—but they do. If they expect stock prices to decline in September, savvy speculators should start to sell in August in anticipation of this drop and this selling should dilute the eventual drop in September. Over time, this type of effect should, in theory, disappear to investor anticipatory buying or selling. Nonetheless, these effects remain prominent in historical stock prices.
I pulled the monthly price data for the S&P 500 from Yahoo! Finance to confirm this effect. Sure enough, I get an average annual price return for the S&P 500 of -0.5% for September over the period from 1950 to present, the worst average return of any month. A 2012 article on the September swoon in the Wall Street Journal reports that the average return to large cap stocks in September is -0.8% going back to 1926. But what people really worry about is not a loss of 1% or so, but rather losses of the magnitude experienced in September of 2011 (-7.1% for the S&P 500), September of 2008 (-9.1%), September of 2002 (-11%), or September of 2001 (-8.1%). September has historically had a fairly high probability of big declines.
So, does it make sense to sell to avoid the September effect and then to buy back in after the anticipated decline? This is where things get interesting. Are you really willing to bet that the effect will repeat itself this year? Are you willing to incur the tax consequences and transaction costs? In any given year, it is risky to simply bet that the average will be the outcome. So, let’s assume that you really believe that the September swoon is a real effect that can be expected to continue. There are years in which September has been a great month for stocks. Most recently, September of 2010 delivered an 8.8% return for the S&P 500. To have a reasonable expectation of making money on this calendar effect, you need to plan to place this bet over a meaningful number of years. The problem is that if you place this bet this year and lose, you are probably less likely to keep making this bet in successive years. This is a behavioral bias: emotion often trumps reason. After losing money betting on the September swoon, you are more likely to question whether you really believe that this effect will persist. It may, after all, be something that occurred for some period of history by chance.
I am not asserting that the September swoon is just a persistent fluke. The numbers are what they are, and the magnitude and longevity of the September effect is notable. Even if I believe wholeheartedly that the statistics are a reasonable guide to the future, however, the potential to avoid an average loss of 0.8% in September is simply not enough to justify a major change in my portfolio. Each person must make this decision for themselves, given the historical data and research, as well as the specifics of their situations.
The views set forth in this blog are the opinions of the author alone and may not represent the views of any firm or entity with whom he is affiliated. The data, information, and content on this blog are for information, education, and non-commercial purposes only. The information on this blog does not involve the rendering of personalized investment advice and is limited to the dissemination of opinions on investing. No reader should construe these opinions as an offer of advisory services.