This is a call to all technical analysts and a nod to my high school teachers. Just because something worked before does not mean it will keep working. The cherished mantra of “Sell in May and go away” has gone the way of the dodo bird. And the trading accounts of anyone blindly following it are doomed to the same fate.
To be sure, “Sell in May” and other market cycles are subject to outliers when they produce the wrong result. After all, cycles are tendencies, not scripts. And if they worked all the time without fail then I would be writing about them from my private island in the Caribbean.
First, let’s stipulate that the strategy used to work and worked well enough for someone to publish the results and for it to go viral in the investment community. However, just looking at the data over the past few years the discoverer would have been dismissed quickly.
We can accept two and maybe three consecutive failures, but after a string of losers we have to have a serious rethink of the strategy. I am not going to bury you with data on how the bulk of the stock market’s gains occurred in the winter months, but you can see the historical record here. The only result no matter how we go about it is that if you had your choice, you would invest in November and step aside in May.
Critics argue that even though the summer months (May-October) produced weaker results they still were positive over time. In order for the strategy to work you would have to park your money somewhere in the summertime to earn more than the Standard & Poor’s 500 gained. Don’t look towards the bond market or the bank for help today.
Rather than crunch numbers, I simply connected the S&P 500’s opening price in May to its closing price in October (see Chart 1). The visuals make it quite clear what happened after the financial crisis – out of six completed cycles, selling in May meant missing out on big market gains four times and minor risk avoidance the other two times.
Chart 1
Standard & Poor’s 500
We can argue that it was due to market distortions from super low interest rate and bond buying programs by the Federal Reserve. However, facts are facts. We can explain any failure by blaming something else.
Now we’ll look back before the financial crisis. During the bear market of 2008 it was indeed smart to get out of stocks in May. But during the bull market of 2003-2007, selling in May failed every time.
And it was same story for the prior bull and bear market cycles between 1994 and 2003. Selling in May avoided losses only in bear markets.
There is one observation that still holds true. Even in bull markets there is a greater risk of drawdowns in the summer. The most recent example was last year when the S&P 500 lost roughly 9% in September and October before rebounding. And for an extreme bull market example, in 1998 the index was down 17% before closing the summer nearly unchanged.
Of course, I cannot predict if getting out of the stock market now will avoid losses or avoid profits by October. One thing we do know is that this is not a bear market, at least not yet, so history is actually on the side of the bulls. Unless there is a better alternative to stocks – perhaps the Picasso painting that sold this month at a record high price – or you are completely risk averse, then there really is no other game in town.
With that said, there are no guarantees, just probabilities. At this stage in an aging bull market we cannot fall asleep at the wheel.