Can you position yourself to snag better returns simply by owning stocks on Fridays? How about month’s end? Or around holidays, when investors might be jubilant headed into a long weekend? The answer, according to a study: maybe 60 years ago, but not anymore.
Equity returns around different days of the week or when the calendar flips are no better than any other times, according to research in the North American Journal of Economics and Finance, which analysed calendar anomalies for the Dow Jones Industrial Average between 1900 and 2018.
It used to be, some of this stuff worked. For six decades ending in the 1980s, for instance, Mondays posted reliably worse returns than other days, for example, and average returns on the day before a holiday tended to be higher for much of the century. But efficient markets prevailed and now all the fun ripples are history, according to the report.
The “‘golden age’ of calendar anomalies was in the middle of the 20th century,” wrote the group of researchers led by Alex Plastun, an economics professor at Sumy State University in Ukraine. Alas, they found, they’ve largely disappeared since the 1980s.
Though the authors ran their own analysis, they also combed through dozens of previous studies — some conducted on a variety of indexes and through different time periods — as reference points. For example, an examination of indexes in 19 countries from 1988 to 2000 — which found that nearly 90 percent of monthly returns were made in the four-day turn of the month — was compared with another stating end-of-month returns in the US exceeded those of the entire month. What the researchers ultimately found is that these effects started to fade and eventually vanished in the 2000s.
Similarly, average results on the last trading day before a holiday tended to be much higher between 1900 and the end of the 1970s. But, the holiday effect has been absent from US markets since then.
“The results of this study provide convincing evidence that the U.S. stock market evolved from being inefficient with a number of calendar anomalies, to being efficient such that it is impossible to find ‘holes’ in price dynamics that can generate exploitable profits,” the authors write.
But the same group subsequently explored what they called the “Halloween effect,” whereby investors adopted the sell-in-May stance and held cash until November in an attempt to avoid lower-than-average returns. The authors found that the phenomenon is real and present, with differences between November to April returns, compared with those from May to October, becoming more pronounced starting in the 1960s, though abating more recently. Even so, the effect can provide opportunities to build market-beating strategies, they said.
An analysis of S&P 500 returns over the last 90 years, broken up by those specific time periods, confirms that finding, according to the Bespoke Investment Group. While the market doesn’t typically go down during the six-month period that runs from May through October, returns are significantly worse than they are during the November through April period, Justin Walter, co-founder and managing partner at Bespoke, wrote in a recent note.
“Going all the way back to 1928, had you only owned the index from November through April each year, you’d be up 4,662 percent,” he wrote. “Had you only owned from May through October, you’d be up a measly 185 percent.” — Bloomberg.