“Sell in May and then walk away,” is what traders of the Halloween Strategy say, referring to an investing technique in which an investor sells stocks before May 1 and refrains from investing again until October 31 in order to increase gains. Sound silly? There’s actually a lot of merit to it. So throw away everything you thought you knew about the financial markets – technical and fundamental analysis, risk management and value investing – because there are other forces at play.

The Halloween Indicator has been spooking investors since 2006

This so-called Halloween Indicator is based on the premise that most capital gains in the stock market are made between Halloween and May 1. This gives investors another six months to invest in other assets or simply refrain from trading the markets all together.

We’ve known for quite a while that the US stock market performs better during the Northern Hemisphere’s winter than in summer. This intriguing discovery was first linked to Halloween – at least scientifically – in 2006 by two researchers in New Zealand, Ben Jacobsen and Nuttawat Visaltanachoti. Their results were fairly conclusive: based on a sample from 1926 to 2006, the difference in summer and winter returns is statistically significant in more than two-thirds of all sectors and industries studied. The cut-off point separating these two intervals was none other than October 31, Halloween.[1]

Forces of evil – or good?

It appears as if the forces of evil that abound during Halloween are a blessing on equities. Take a look at a practical example: Over the past 50 years the S&P 500 Index has gained an average of 6.6% between Halloween and May, compared with just 0.8% in the other six months. Do we really have to ask in which half-year you’d rather be investing?

Just by looking at the numbers, it’s abundantly clear that the Halloween Indicator isn’t just happenstance – it’s too eerily consistent to be one. Just read the way the researchers describe this strange effect:

“Recent evidence has revealed a strange anomaly known as the Halloween indicator… To date, the anomaly has survived all the usual controls and robustness checks and it seems unrelated to any other anomaly identified in the market efficiency literature” (p. 1).

While the authors provide alternative explanations for why this might be the case – seasonal affective disorder in traders, changes in risk aversion – they weren’t entirely convinced.

“We still know little about the anomaly,” they said.

The eerie exceptions

As one might expect, the researchers found some exceptions to the rule. Using Fidelity Select mutual funds, they identified five sectors that, on average, performed just as well in the six months before Halloween than in the six months after it. These are: select consumer staples, select leisure, multimedia, retailing and utilities – all sectors related to consumer consumption.[2]

Giving it a shot

If you decide to give the Halloween Indicator a try, you may consider selling sectors related to production rather than consumer consumption, as the latter perform much better during the summer. So eat up your candy, wear your favourite costume and visit a few haunted houses while you’re at it because the next six months after Halloween may just set you up comfortably for the rest of the year.

[1] Ben Jacobsen and Nuttawat Visaltanachoti (May 8, 2006). “The Halloween Effect in US Sectors.” The Financial Review.

[2] Mark Hulbert (Oct. 2, 2015). “Opinion: Those who sold in May are looking to return to stock market.” Market Watch.