The "sell in May" seasonal pattern, as famously celebrated by the Stock Trader's Almanac and academic research papers such as Sandro Andrade et al.'s "Sell in May and go away just won't go away" and Bouman and Jacobsen's "The Halloween indicator, sell in May and go away: another puzzle," has proven to be a rigorous statistical phenomenon over more than a century and is pervasive in financial markets throughout the world. The seasonal effect is very real and has been written about extensively in popular media (and to a lesser extent in academia), but relatively little has been written examining what may be the etiological factors behind this phenomenon. So this article attempts to briefly address the question: what are the causes of this phenomenon?
My research indicates that the seasonal effect of "sell in May" is the result of a rather complex combination of factors. So I will postulate a number of causes behind this seasonal pattern. One of the causes of this seasonal pattern is the run up in Dow stocks having dividend increases in April and May, which encourage investors to take the plunge and buy stocks at their April and May highs but peter out after May as profit takers arrive. May tends to be a month when many corporations report their earnings, pay out handsome dividends, and concomitantly announce they are increasing their dividend payouts. It stands to reason that May is thus when stocks, particularly dividend paying stocks, are most popular, and market sentiment most bullish, and thus these stocks are selling at relatively high prices. But after earnings season comes out in May, the bullish sentiment starts to wane and many investors or traders do not want to wait around for the next dividend and next quarterly earnings season to come around and so they take profits in the interim, thus driving stock prices down.
My research also indicates that weather is a big factor. Individual workers tend to work harder and be more productive when it is colder but tend to relax and take it easy and vacation and enjoy the outdoors more when the weather is nice outside, thus corporate and overall productivity is substantially higher in the colder months, whereas productivity is substantially lower in the warmer months. This also comports with the fact that economies of nations with tropical climates like the Philippines or Indonesia or Ecuador or Colombia or Congo on average tend to be less productive because the individuals are relaxing and enjoying the weather on the warm sunny beach or wherever, while economies of nations in cooler temperate-zone climes like Japan or Germany or China or Canada or the United States are more productive because individuals sometimes feel like working hard just to stay warm or because they are cooped up inside their offices and have nothing much better to do and the weather outside is cold and nasty, and it may be raining or snowing outside; so instead they work hard during the cold and bad weather months so they can store up more than enough for themselves to relax and enjoy the warm good weather months. See e.g. here.
Another factor is that the "sell in May" thesis has become a self-fulfilling prophecy to some extent. Many people who know of the "sell in May" seasonal pattern deliberately and somewhat robotically sell in May simply because they know about the historical strength of this pattern and they don't want to be caught on the losing end of a major downturn or correction in the markets come the summer doldrums. In fact, as more and more press coverage has resulted in more and more investors and traders knowing about this seasonal pattern, the pattern has only become stronger in more recent years (since 1951) than it was many decades ago. Thus, this strengthening of the pattern in more recent years coincident with increased media attention focused on the pattern supports the self-fulfilling prophecy explanation. See chart below sourced from cxoadvisory.com. (click to enlarge)
Another factor is that some people will sell in May at whatever the market price is so that they don't have to worry about or monitor the markets and instead can go on vacation and just have fun during the nice warm weather without any anxiety about their investments. A secondary factor closely related to this is that school gets out starting in late May and so moms and dads with children go on vacation in the summer months, and again, they don't want to have to worry about or monitor their investments while on vacation so they just sell in May at whatever the market price is so they can just relax and enjoy their vacation time with their kids. Another related factor is that some teachers only get paid during the months they work and during the summer vacation months, they don't get paid. Thus, the aggregated effect of millions of teachers not getting paid is less consumption in the economy.
In conclusion, whatever the exact cause or causes of this seasonal pattern, it is a very rigorous and pervasive and rather intriguing phenomenon. I have postulated above a number of factors that coincide to cause the sell in May annual seasonal cycle, and I am sure that my readers, who collectively are much smarter than I am, can postulate a number of other factors that may be at play feeding into this well-known seasonal effect.