If you sold in May and went away, as the old stock market rhyming-strategy goes, then it is time to come back: The Halloween Indicator is flashing a buy signal.
The strategy calls for selling stocks at the start of May and then returning to the market after Halloween – for no other reason than historical returns point to poor performance for six months of the year, starting in May, followed by strong performance starting in November.
I have never heard of anyone actually using this strategy, but patterns are always fascinating. As Mark Hulbert points out, the Dow Jones industrial average has returned an average of just 1.7 per cent during the "summer" season, for data going back to 1896. The Dow has performed far better during the "winter" months, rising an average of 5.1 per cent.
The past two years have been more extreme. In 2011, the Dow fell 6.7 per cent in the summer and rose 10.5 per cent in the winter. In 2010, it rose 1 per cent in the summer but jumped 15.2 per cent in the winter.
But is 2012 shaping up as another win for the strategy? Mr. Hulbert sounds hopeful: The Dow fell 0.9 per cent between May and November, and turned in a triple-digit gain on Nov. 1.
"This is noteworthy, since patterns normally stop working once they become widely known," Mr. Hulbert said. "In the case of the Halloween Indicator, in contrast, it would appear that the pattern is strengthening."
Academics have even weighed in, and found that this pattern persists globally. Mr. Hulbert pointed to a new study from Ben Jacobsen, a finance professor at New Zealand's Massey University, who looked at the Halloween Indicator in 108 markets. Over the past 50 years, winter returns have been an average 6.3 per cent higher than summer returns.
Of course, investors probably have more on their minds than the joys of approaching winter. China and Europe hang over sentiment, as does the U.S. "fiscal cliff" which threatens the economy with spending cuts and tax increases.
As well, there is even a competing strategy facing investors right now: The Presidential cycle, which argues that the first and second years for a U.S. president coincide with dismal returns for the stock market.
Oh, but even that pattern should come with an asterisk. Ed Sollbach, portfolio strategist at Desjardins Securities, found that victorious Democratic incumbents bode well for U.S. stock returns. Since 1944, the first year of a second term for a Democratic president has brought returns of 20.3 per cent, on average.
In other words, this suggests "that 2013 should be a good year if President Obama wins," Mr. Sollbach said in a note.
David Berman