Trading Nation

It sounds crazy, but 'sell in May and go away' is good advice

Should you sell in May?
Should you sell in May?

Rhyming couplets can rarely be expected to serve as sound investment advice. But "sell in May and go away" may be a prominent exception.

"This old saying was proposed in the past … and it actually wasn't looked at until probably the late '90s. We looked at it, and indeed, it is persistent and it's economically just as strong as it was then," University of Miami associate professor Michael Fuerst told CNBC's "Trading Nation" this week.

Fuerst, along with fellow University of Miami professors Sandro Andrade and Vidhi Chhaochharia, reported in a 2012 paper that stock returns were 10 percent higher in the November-to-April half of the year than in the May-to-October period.

Importantly, this result isn't solely based on historical American stock returns. In that case, the academics could be making the all-too-common mistake of "proving" an adage by using the same evidence that was used to bring about that line of thinking.

Rather, they examined returns across 37 markets within a 14-year time period that was not tested in a prior paper that also found support for the sell in May effect.

In each of those 37 markets, returns in the May-to-October period were found to be smaller than those in the November-to-April period. The sample size here is relatively small, but by pooling the data, the authors were able to show statistical significance.

"This out-of-sample persistence indicates that the effect is enduring and not a statistical fluke," the authors conclude.

So what's the best way to capitalize on this effect?

Well, the problem with actually selling in May is that one is well-served by maintaining exposure to a full year's worth of equity returns. The approach Fuerst would suggest, then, is to double one's market exposure in the November-to-April period, and then simply hold Treasury bonds in the other period of time. By doing this, an investor could expect to outperform the market by a significant measure, Fuerst says.

Incidentally, the question of why this effect actually exists remains an open one.

Outperforming the market shouldn't really be this easy. For starters, in order for one set of people to cash in by buying their stocks in November and selling them in April, another set needs to be doing the opposite. One expects that at some point, these "buy in May" folks would get sick of lagging the market, and would start demanding a higher price for their sacrifice (until the effect disappears).

Read More The bull market is about to set a major milestone

Still, the 37-country analysis does clarify at least one potential misconception about the sell in May effect.

"It used to be that people would say, 'Oh, investors go away for their summer vacations, that's why the market does this.' But if that was the case, this shouldn't work as well in Rio [where summer starts in December] as it does in New York," Fuerst pointed out.

Indeed, in the U.S., selling in May generated 6.9 percent of outperformance in the authors' sample. That compares to 10.5 percent in Brazil, and a stunning 25.9 percent in Russia.