When will this stock-market boom end? Here's a creepy day to put on your calendar.

By Brett Arends

Wall Street is again proving that investors seem to make all their money during the winter months

Up and down the country, financial "experts" are telling anyone who will listen that the stock market is efficient and you can't beat it.

Business school professors are telling it to their students (for $100,000).

Financial advisers are telling it to their clients.

Gurus are telling it to the Nobel Prize committee, which has over the years lavishly rewarded many apostles of this so-called efficient markets hypothesis.

The only problem? Somebody forgot to tell the stock market. Again.

It is an astonishing but undeniable fact that, once again, Wall Street is proving that you seem to make all your money during the winter months - starting around Halloween and, traditionally, ending at the beginning of May.

It makes no logical sense. It defies every tenet of the efficient market hypothesis. It shouldn't happen. But it does.

The current stock-market boom began on Monday, Oct. 30, when, after months of sluggishness and weeks of sliding, the S&P 500 SPY suddenly jumped more than 1%. Since then, the benchmark index has risen an astonishing 25% and change.

During the preceding six months? Bupkis. From the last day of April through the last day of October 2023, the S&P 500 went from 4,169 to 4,194 - a gain of half a percent. And all of those gains came on the final day, Oct. 31.

It was a similar story a year ago. The tough 2022 stock-market slide ended just two weeks before Halloween, on Oct. 14. From Oct. 31, 2022, through the end of April 2023, the S&P 500 rose 7.7%. Then, as noted above, it stalled - until the next Halloween.

This isn't merely a matter of coincidence or a few anecdotes. And naturally there are exceptions, like the pandemic years of 2020 and 2021. But finance professors Cherry Zhang and Ben Jacobsen have run all the available historical data on every stock market in the world, going back in some cases to the 1600s.

"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," they report. "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."

Their conclusion: The Halloween effect is real and pervasive. Average stock-market returns during the winter months have been much higher than those in the summer months.

"The effect seems remarkably robust with returns on average 4% higher during November-April period than during May-October," they write. In markets where they also had data on short-term interest rates, they found that on average, you didn't get paid much, if anything, to take on stock-market risk over the summer months. "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," they write. "Summer risk premiums are not only not significantly positive, they are in most cases not even marginally positive. In 45 countries the excess returns during summer have been negative, and in 7 significantly so." On average, investors during the summer earned 1.1% less than they would have earned just holding cash. "Only in the winter months do we find evidence of a positive risk return relation," they add.

Nobody knows why this should be the case. One strongly held theory is that investors - I am not making this up - are affected by the shortening of the days in the Northern Hemisphere after June 21. This subconsciously makes them gloomier and more cautious going into the fall, driving down stock prices. Why this should cause a rally in late October, two months before the winter solstice, is anyone's guess. Why this should also cause stock prices to fall in the Southern Hemisphere is another challenge, although you could argue plausibly that markets in places like Johannesburg and Sydney have generally danced to a global stock-market tune orchestrated in New York, Tokyo and the major European capitals. Who knows?

Other ideas include the past role of the harvest calendar - which involved cash calls in farm states in September and October - and long, long memories of the infamous Wall Street crash of 1929, which began in October.

But it raises the question of how long the current surge can last without at the very least pausing for breath. Markets rarely go up or down in a straight line. And there are plenty of signs of ominous complacency or euphoria, such as the new records for bitcoin (BTCUSD) and Nvidia (NVDA) and the widespread indifference to inflation. Expecting the rally to run out of steam over the next seven weeks, before a long spring and summer lull, may make no logical sense. But it would fit the calendar.

My own assumption is that this "sell in May" effect, or Halloween effect, will go dramatically into reverse the year I finally give in and follow it. The year I dump my stocks on the last day of April, you can expect a summer boom for the ages.

Meanwhile, anyone worrying about how far this market will rise or when the boom will end now has a date to put on their calendar. It's no worse than anyone else's guess, though that's another thing they won't tell you.

-Brett Arends

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

03-16-24 1722ET

Copyright (c) 2024 Dow Jones & Company, Inc.