Although it's easy to doubt the predictive power of an oversimplified adage like "sell in May and go away," there is surprisingly more truth to it than some might expect. Several historical trends lend credence to this saying, and there is no shortage of explanations for why "selling in May" is an idea worth considering. But of these trends, perhaps the most interesting is the idea of "summer months vs. winter months" market seasonality.
The premise of this particular market seasonality study is that, from a historical standpoint, the stock market performs far better during the 6-month period from November through April than it does from May through October. To evaluate this trend, we will look at the returns from investing $10,000 in the Dow Jones (DIA) on May 1, 1950 and then selling on October 31st of that year, repeating every year until the present. The same is done for the November 1 to April 30 period.
The chart below is a reproduction of the "Six-Month Switching Strategy," first published in the Stock Trader's Almanac.
click to enlarge
As shown, if you were to invest $10,000 in the Dow Jones on May 1 and sell it on October 31 each year since 1950, you would have actually lost money as of 2011. On the other hand, if you were to invest $10,000 on November 1 and sell it on April 30 for every year since 1950, your investment would have grown to over $700,000, representing an average annual rate of 7.1%.
Looking across the 62 years since 1950, we see that only 13 of these years saw a negative return during the November to April "winter" period. And only three times (1969, 1973, and 2008) did the Dow Jones fall by more than 10% during the "winter" 6-month period. As for the "summer" months from May to October, during this period the Dow experienced a negative return for 25 of the 62 years, and saw the Dow decline more than 10% during 11 of the 62 years.
Although the historical trend is pretty clear - that growth in equity markets tends to be greater during the winter months" period, or conversely, that markets tend to sustain greater losses during the "summer months" period - there will always be exceptions. Big one-off macro events, like a recession or a breakup of the euro, will impact markets with little regard for these trends.
So how should investors incorporate their understanding of this trend into the management of their portfolios? First of all, it helps to look at this issue from a slightly different lens. Looking at average monthly returns of the S&P 500 (SPY) (IVV) from 1926-2004, we see that the two months of the year with the lowest average returns are September and October. As such, it seems that a fair amount of the poor historical performance during the 6-month "summer" period discussed above can be attributed to the negative average return for the month of September.
In conclusion, although clear patterns do exist in the historical returns detailed in the "six-month switching" study, in many ways the trends we see are highly dependent upon the way we present and examine the data. That being said, there is undoubtedly great value in being familiar with these historical trends when making investment decisions.
One doesn't need to literally employ the six-month switching strategy in order to benefit from understanding these patterns. Knowing what we do about the market's historical performance during the May through October period, for example, and coupling it with our understanding of current economic conditions and general market sentiment, we might expect some bearish market activity in the coming months, and consider reflecting this in our portfolios.