Sell Stocks in May? Tempting but Not So Smart

With U.S. stocks near record highs and memories of last summer’s volatility still in mind, investors could be forgiven for wondering whether this is the year to “sell in May and go away.”

Stocks have shaken off an early-year rout, but a reading of first-quarter growth for the U.S. economy came in below expectations last week and corporate earnings have been lackluster, clouding the long-term outlook for stocks.

After falling 11% in about the first six weeks of the year, stocks have bounced back, but now at a slower pace. In the six weeks following a Feb. 11 trough, the S&P 500 climbed 11%. In the roughly five weeks after that, it rose 1.4%. Both the Dow Jones Industrial Average and the S&P 500 are off about 3% from their highs set in May of last year.

There are many interpretations of the long-running trading adage, but the underlying recommendation remains the same: Stock investors should avoid a summer slump.

It worked last year. If an investor sold the S&P 500 on the Friday before Memorial Day and rebought the index the Tuesday after Labor Day, she would have avoided a 7.4% decline.

Otherwise, the benefits are a bit fuzzier. Stocks gained during the summers of 2012 and 2014. In 2011 and 2013, they fell.

It is a controversial strategy. Some investors ridicule the idea as no better than predictors tying stock-market behavior to the length of hemlines, butter production in Bangladesh and which National Football League conference wins the Super Bowl.

“Even if last year ‘sell in May and go away’ worked, this year could be a year that it doesn’t. In the short term, you just don’t know,” said Marc Pfeffer, senior portfolio manager at CLS Investments, which manages about $6 billion.

Since 1970, the S&P 500 has gained 1% on average in the period between Memorial Day and Labor Day, according to an analysis by Ana Avramovic, trading strategist at Credit Suisse. Stocks rose during 30 of those summers by an average of 5.6%. The declines were more painful, averaging 8% in the 15 years stocks declined between Memorial Day and Labor Day, Ms. Avramovic’s analysis shows.

The summer will start off with some potentially market-moving events.

The next Federal Reserve meeting is scheduled for June. Last week the Fed held interest rates steady and gave little indication it plans to raise rates in the coming months. Investors will still be monitoring the meeting for any surprises, even though expectations for a rate increase remain low. Later that month, the U.K. will vote in a referendum on whether to remain a member of the European Union.

July will bring another batch of quarterly corporate earnings, which are expected to again contract, according to FactSet.

Negotiations between Greece and its international creditors ran into trouble last week, setting the stage for another summer showdown between the parties.

Jason Ware, chief investment officer for Albion Financial Group, which manages about $1 billion, said he understands the desire to sit out this summer’s potential volatility. But trying to optimally move money in and out of the market during a short period is unwise, he said.

“You set yourself up for errors when you try to get really tactical,” he said. If the market doesn’t fall and instead rallies, that could make it tougher to time a re-entry to stocks.

During the period between Memorial Day and Labor Day, daily stock-trading volumes on the New York Stock Exchange tend to fall about 3.1% below yearly averages, according to data going back to 1996. Fewer shares changing hands means swings in the stock market can be sharper.

In 2011, there were 31 moves of at least 1% in either direction in the S&P 500 during the summer, a period that included a downgrade of the U.S. by Standard & Poor’s Ratings Services. Between Memorial Day and Labor Day, the index dropped 12%.

Ben Jacobsen, a finance professor at TIAS School for Business and Society in the Netherlands, said he makes it an annual routine to take a much longer trading hiatus, based on his own research into seasonal market trends.

He and his colleagues found that world-wide, investors averaged 4.5% better returns between November and April than between May and Halloween.

The so-called Halloween effect of improved seasonal performance persists in 82 of 109 stock markets studied.

In the U.K., where data goes back three centuries, investors who follow that strategy for five years have an 80% chance of beating the market, based on his historical analysis. Over 10 years, it becomes a 90% chance. His results exclude transaction costs and taxes.

He has been investing according to the strategy since the 1990s, and currently commits about half his portfolio, investing the other half in stocks held long-term. In the U.S., he keeps his money in low-volatility exchange-traded funds for the summer. In Europe, he goes to cash.

“It doesn’t make you rich,” he said. “But it makes you richer a little bit faster than other people.”

Analysts at Northern Trust Asset Management found in a 2012 report titled “Sell in May…and Pay!” that investors face significant downsides getting in and out of the market around the summer.

Even if the time period is extended to include Halloween, U.S. capital-gains taxes alone erode the value of the approach, they found.

Meanwhile, investors also face transaction fees, missed opportunities and confusion about when to buy back in.

“You never know from a a calendar perspective when you’ll get your returns,” said Bob Browne, chief investment officer at Northern Trust, which managed about $900 billion as of March 31.

Write to Corrie Driebusch at corrie.driebusch@wsj.com and Aaron Kuriloff at aaron.kuriloff@wsj.com