Brett Arends’s ROI: Halloween: It’s the best time of the year to buy stocks

His advice was to steer clear of taking on more risk right then—which turned out to be an excellent tip, as the stock market promptly plunged.

Whether or not the timing was lucky, I reflected at the time that those were pretty smart questions generally. And they only cost me $10, as opposed to the hefty fees you’ll pay on the street of shame for serial underperformance.

No wonder Warren Buffett says stock market analysts make fortunetellers look good.

Meanwhile, ghosts and ghouls stalk the stock market this October, yet again. Halloween is so often the scariest time of year for investors, who must look at the gyrating Dow Jones Industrial Average and ask themselves: Trick or treat?

So it’s worth revisiting this weird, spooky and inexplicable paranormal fact: Halloween is the best time of the year to buy stocks. According to financial research, most or all of the stock market’s long-term returns have come during the winter months.

Professors Cherry Zhang of Nottingham University Business School in China and Ben Jacobsen of the TIAS School for Business and Society in Europe compiled all the stock market return data they could get from around the world. This wasn’t just the U.S.A.

“Our data consists of all 114 stock markets with stock market indexes in the world for which price indexes exist and in total we have almost 63,000 monthly returns. The sample starts with the U.K. stock market in 1693 and ends with the addition of the stock market of Rwanda which starts in 2013,” they wrote.

(Freaky note: 1693 was the time of the Salem witch trials.)

Their conclusion? Stock markets overall beat cash handily over the winter months. But not over the summer months.

“Overall, the 62,962 monthly observations over 323 years show a strong Halloween effect,” Zhang and Jacobsen report. “The effect seems remarkably robust with returns on average 4% higher during November-April period than during May-October.”

In 65 countries they had substantial data showing not only stock market monthly returns but also the interest you could have earned each month by keeping your money in a deposit account or short-term bills. “In none of the 65 countries for which we have total returns and short term interest rates available—with the exception of Mauritius—can we reject a ‘Sell in May’ effect based on our new test.” In other words, only in Mauritius did it definitely make sense to keep your money in the market all year.

This Halloween effect “seems to defy economic gravity,” note Zhang and Jacobsen. It hasn’t disappeared or even gone into reverse since it was first written about extensively about 20 years ago.

It has been confirmed by others, such as economics professors Thomas Degenhardt and Benjamin Auer of Germany’s University of Leipzig, who did a methodical review of all the academic studies into the Halloween Effect and confirmed that the findings were “robust.”

Meanwhile, more recent research has found, for example, that “quality” stocks have tended to do better in the summer months and “junk” stocks in the winter: Another way of saying that you have traditionally got the best rewards for taking on risk from November to May.

And I’ve pointed out here before, the summer months certainly seem like a time of volatility.

If you’d sold out of the S&P 500 SPX, +2.46% —or an index fund that tracks it like the SPDR S&P 500 ETF SPY, +2.38% – on the first of May this year you could have bought it back for up to 13% less by earlier this month.

Why would markets typically be so lousy during the summer? Or so good during the winter? No one seems to know.

I always wonder if investors during September and October are somehow haunted by the ghosts of 1929 (and, for that matter, 1987).

But the scariest time of the year has traditionally been the best time to ramp up your exposure to the stock market. Spooky, but true.