Tell ‘Sell in May’ to Go Away

Between Mothers’ Day, the Kentucky Derby, the Stanley Cup playoffs, the Indy 500 and the Monaco Grand Prix, there are a lot of great traditions in May. Alas, there is also a more dubious one: the annual admonitions to sell stocks and stay away until Halloween. Or, as the saying officially goes, Sell in May and go away and don’t come back until St. Leger Day, which is a mid-September British horserace. These days, most pundits argue the endpoint is October 31, tied to the April 30 through Halloween period being the calendar’s weakest rolling 6-month stretch. In our view, it is all trivia mixed with silliness. Calendars don’t predict.

Sell in May might seem extra tempting this year, what with stocks’ springtime volatility. After the best Q1 in five years, the S&P 500 has taken a step back in April alongside global markets.

Yet this is a pretty powerful demonstration of the myth’s shallowness: Stocks don’t run on schedules. Sometimes pullbacks happen in summer. Sometimes they happen later. And sometimes they happen earlier, presaging a summertime rebound. Short-term moves run on sentiment, and sentiment is fickle and unpredictable. Markets can react to any old thing, at any old time, for any old or even no reason. But if you sell in May because stocks wobbled in April and you are feeling nervous, you risk locking in those declines and missing the rebound that follows. We often make the point that you can’t buy past returns—well, you can’t avoid past volatility, either.

But also, stocks’ performance history doesn’t back up Sell in May once you dive in. Consider Exhibit 1, which shows the S&P 500’s average total returns by month since good data begin in 1926.

Exhibit 1: Monthly Returns at a Glance

Source: Global Financial Data, Inc., as of 4/29/2024. S&P 500 total returns, monthly (average), 12/31/1925 – 12/31/2023.

Yes, as the highlighted row shows, May has the third-weakest returns, beating only September (more on that later this summer, natch). But that third-weakest return is still positive, and returns in the month are overall positive much more often than not—63 out of 98 Mays since 1926, to be exact. That is a 64.3% frequency of positivity, which is a shade higher than the 62.9% frequency of gains in all months since then.^[i]

Moreover, summer returns have historically been quite nice. July is the calendar year’s best month! Not that this is predictive, either, but one would think a seasonal saw would at least consider that it is encouraging people to sit out the strongest month on the calendar, if only for the sake of intellectual consistency.

Slicing this by 6-month returns to account for the full Sell in May window doesn’t make the case for sitting out stronger. Exhibit 2 shows this, highlighting the Sell in May window’s returns.

Exhibit 2: Rolling 6-Month Returns at a Glance

Source: Global Financial Data, Inc., as of 4/29/2024. S&P 500 total returns, trailing 6-month (average), 6/30/1926 – 12/31/2023.

Yes, it is the weakest of the calendar’s rolling 6-month spans. But it is still positive. We aren’t aware of anyone who met their long-term financial goals by routinely sitting out up markets because they weren’t up as much as other arbitrary stretches. The frequency of positivity is even higher here, at 71 of 98 years—72.4%.^[ii] This, too, is a mite more often than the 71.5% frequency of positivity in all rolling 6-month stretches since 1926.^[iii] Nothing here shouts stay away!

Seasonal sayings sound fun, especially when they rhyme. And, volatile markets mean there will be enough negative stretches to give at least the appearance of a half kernel of truth. But they don’t hold up. Calendars aren’t market drivers.

Even if seasonality ever worked a lot of the time, it would get priced in so quickly that it would stop. People would get out in April to front-run Sell in May. Then in March. Then February. Round and round and round, chasing their tails. And all for periods so short as to be utterly meaningless when you consider that enduring short-term volatility isn’t some massive return headwind. Rather, it is the price we all pay for stocks’ stellar long-term returns. Over time, all the noise evens out into a big, uneven upward-sloping line.

If you can identify a bear market early enough, exiting markets can make sense—if you can identify the eventual rebound in time to reinvest and reap the benefit. But seasonality will tell you nothing on this front. Bear markets happen for fundamental reasons—not random chance. They start one of two ways: when euphoric markets have finished climbing the wall of worry and are primed to overlook a deteriorating backdrop, or when some huge, shocking, unseen negative wallops them before the climb is done. We don’t see either today.

Sentiment, though warmer than a year ago, is hardly euphoric. Meanwhile, the fears unsettling people lately—geopolitics, inflation and Fed chatter—are stale and reheated several times over. No surprise power here, just a nice wall of worry for stocks to climb as fears fade.