April Showers Bring May Myths

Story Highlights:

• Predictably, the old adage “sell in May” has re-emerged with the new month.
• Most of the persistence behind this saying isn’t based on fact, but behavioral biases.
• Markets are not seasonal—they’re cyclical—and knowing what month it is won’t help in forecasting.

It’s May! And here at MarketMinder, May always holds excitement. Each year at this time, headlines encourage investors to sell stocks because the calendar has moved, and it’s an annual tradition of ours to show the flaws with this myth (like here, here and here.)

Versions of “sell in May” have existed for decades. Some cite an origin from over a century ago in Britain when the adage had a touch more instructional advice (“sell in May and go away, don’t come back till St. Leger’s Day”). Many attempt to rationalize their belief in this theory by claiming volume drops off steeply while traders take summer vacations—which was possibly true (though unlikely and unprovable) in earlier generations, but certainly isn’t today considering one can trade stocks in seconds from a Florida beach using one hand while slathering on sunscreen with the other. In modern markets, there’s no dire lack of liquidity in summer (nor are trading volumes predictive of market direction). However, despite years of disproof, googling “sell in May” will yield about 12,100,000 results. (Guess it’s 12,100,001 now.)

To be fair, historically, there is some evidence it works—if reviewing solely index price returns for European stocks. Summer might appear bad historically because of a proliferation of companies paying dividends then (unlike the US, many European stocks don’t pay on a quarterly schedule). But that’s price returns only and for European stocks—no doubt a bit of an odd thing to conflate to the entire market, particularly since dividends paid lower the stock price but not the total return (and you’re also ignoring many areas of the globe). Another way it works? “Sell in May” no doubt works pretty darn well—during bear markets. But if you could dead-to-rights forecast every bear market, you wouldn’t need sell in May.

This myth is mostly driven by behavioral biases pushing investors to attempt to create some easily understandable system for assessing markets. Psychological biases drive proponents to either explain away or twist data to fit the belief. That’s likely why many timing iterations of “sell in May” exist. (Do you sell May 1^st, 15^th, 31^st or earlier? And when to re-enter: St. Leger’s Day? Mid-September? October? Later?) No matter how you slice it, selling in May and buying later increases the probability of missing positive returns—which facts show aren’t unlikely during summer.

Since 1926, May’s S&P 500 total returns on average are positive. June even more so. July is, on average, the best month of the year! August is likewise positive. Taken together June, July and August are the best three consecutive months of the year! (Hooray for May!) While that doesn’t necessarily mean stocks will be positive this year, it should call into question the accuracy of “sell in May.” Even last year, when May and June weren’t terrific for stocks, the correction began in April and ended in early July. And we’re quite sure neither the correction’s beginning nor its end had anything to do with the time of year. Last, consider the flipside: If investors are re-entering in fall, shouldn’t September historically be great for stocks? It’s not. (And again, that’s just an average and not predictive in any way of what September—or any other month—will do each year.)

If “sell in May” worked before, rational market participants would be fast to sell prior to May. And then before April. And so on. Ultimately, one could be in the rather ridiculous position of selling in June—because May’s only 11 months away.

Equally puzzling, this myth runs headlong into another: While proponents push selling in spring and re-entering in fall, believers in another seasonal myth argue fall is “crash” season—like 1929, 1987, and 2008. (True, though these three examples still aren’t indicative of seasonal market behavior—there’s no inherent reason to fear fall.) So these myths are basically mutually exclusive. Cumulatively, adhering to all the seasonal myths surrounding stocks would likely result in owning stocks for just the couple weeks around Christmas, as if Santa sprinkles bull market in your stocking only then. Which, if it isn’t obvious, is a terrible strategy for equity investors.

No matter how data are divided up, massaged, and spliced together, markets are cyclical, not seasonal. Whether it’s the January effect, Santa Claus rally, Diwali (a supposedly bullish Indian holiday occurring smack in the middle of supposedly crash-prone September), or the coming of May, rest assured that whatever markets do, it won’t be because of anything dictated by the calendar.