Until a decade or so ago, September was the best month on the calendar for gold. Since then, it has gotten worse.
Anindito Mukherjee/Bloomberg
font size
September on Wall Street is known for its strong monsoon crosscurrents. Some of them may even have strong enough statistical and theoretical support to justify betting on them.
Perhaps the most well-known truism about September is that it is a terrible month for stocks. What many may not know is that it is also considered the best month of the calendar for gold.
Consider here the performance of gold since December 31, 1974, the date when it became legal for US citizens to own gold. since then,
gold bar
In US dollars it posted an average gain of 1.8% in September, more than four times its average return of 0.4% in the other eleven months of the calendar. The exact opposite was the case for stocks:
Dow Jones Industrial Average
It lost an average of 1.0% in the month of September starting in 1975, compared to an average gain of 1.0% in all other months.
The Dow has been around for much longer than the period since 1975, of course, and September was the worst average performance over that long history as well. Since the index was created in 1896, September has produced an average loss of 1.1%, compared to an average gain of 0.8% in all other months. Moreover, September’s dismal relative performance has been remarkably consistent: its return relative to the other eleven months has been below the average in every decade since 1900 except for one.
Because of this, it is perhaps not surprising that the average difference in the Dow’s return between September and all other months is large at the 95% confidence level that statisticians often use when assessing whether a pattern is likely to be true.
Gold’s September record is less consistent. Until a decade or so ago, September was the best month on the calendar for gold. Since then, this has been the worst, and this latest experience weakens – though does not remove – the statistical argument in favor of gold being an excellent September performer. Over the entire period since 1975, this statistic was only marginally significant – at only a 90% confidence level, rather than the traditional 95% confidence level.
Why might stocks do badly in September?
No matter how strong the statistical case is in favor of the seasonal pattern, you should not bet on its persistence unless a reasonable theoretical case can be made for why it exists in the first place.
After years of insisting that such a case did not exist for stocks, a recent study published in the Journal of Quantitative and Financial Analysis changed my mind. Worthy “Seasonal Asset Allocation: Evidence from Mutual Fund Flows,” The study was conducted by Mark Kamstra of the University of York, Canada. Lisa Kramer of the University of Toronto; the late Morris Levy of the University of British Columbia; and Ross Wermers of the University of Maryland. And they find strong evidence of this Seasonal affective disorder (SAD) is the source of the poor performance of the stock market for the month of September.
SAD, of course, is a depressive mood disorder associated with the change of seasons. Few of us associate it with the month of September, as more people experience grief than during the winter months. But what affects the stock market is not the number of people who suffer from social anxiety disorder, but changes in that number. The largest monthly change in those with social anxiety disorder, according to psychiatric data, occurs between August and September.
The authors of this recent study correlated these monthly changes in SAD with the stock market by measuring cash flows in and out of mutual funds. After controlling for other potential factors that could also explain these outflows, the researchers found a high correlation between changes in the incidence of SAD and mutual fund outflows. The month with the largest net inflow is September.
Thus a strong statistical and theoretical argument can be made as to why September might be a below-average month for stocks. While this does not guarantee that the stock market will lose steam next September, it does increase the possibility of a decline.
Why might gold perform better in September?
There are reasons why gold did well in September, but they are weaker than the reasons why stocks did badly. The only academic study that looked at gold’s tendency to do well in September, at least to my knowledge, was conducted more than a decade ago by Dirk Bohr, professor of finance at the University of Western Australia. In his 2012 study entitled “Seasonality of Gold – Autumn Effect” He hypothesized that this trend could be caused by any of various factors, including the SAD factor which could explain the prospects for equity weakness in September. Another possible factor, he wrote, was the demand for jewelry in India ahead of Diwali, that country’s annual “festival of lights” that takes place sometime during October or November. (This year Diwali will be celebrated on November 12).
Bohr confirmed in his study that these hypotheses are just speculations. In a recent email, he also speculates that in recent years September may no longer be a good month for gold because enough investors became aware of the seasonal pattern and tried to take advantage of it – thus killing the goose that lays the golden egg. He wrote that by jumping the gun and buying gold in August instead of September, they had carried over September’s gains to August. Since 2010, August has been the best month in the gold calendar.
Bottom line: betting that gold will do well in September is much more temporary than betting that stocks will do poorly.
Corrections and amplifications
Diwali is the Hindu festival of lights. An earlier version of this column incorrectly gave its name as Denali.
Mark Hulbert is a regular contributor to Barron’s. for him Hulbert Reviews Tracks investment news releases that pay a flat fee to proofread. He can be reached at [email protected].
Write to [email protected]