Sell in May and Go Away

|
 |  Includes: DIA, QQQ, SPY
by: Babak

Forget everything you know about the market. If you have a long enough time horizon, all you need is a calendar.

More specifically:

  1. Look at the calendar
  2. From May to October, stay out of the market (or go short).
  3. From November to April, buy or go long.

That’s it.

Forget technical analysis. Forget fundamental analysis; watching talking heads on TV; reading blogs (cough). Or anything else.

The only catch is that you have to have time on your side. Because each and every single instance is not going to go in your favor. But on average, you’ll be far ahead of a simple buy and hold strategy.

halloween indicator winter summer stock market seasonalityFor some strange reason, winter months are great for stock market returns and summer months are bad. This holds true across time as we go back decade after decade, it also holds true if we go across the globe and look at different stock markets in different countries.

This is known by a few names: Sell in May and go away, Halloween indicator, winter months good, summer months bad, etc. Theoretically this pattern should not exist. But it does. And it bothers a lot of economists. Most of them would rather not think about it so they just shoo away this and similar holes in the EMH.

Fact is that the Halloween indicator, like election year pattern, the January effect, 4 year cycle and many others, present not one but two challenges to EMH. One, they should not exist if the market is as economists theorize, “efficient”. Two, if they did exist, they shouldn’t persist, which they do. That is to say, if they do exist, which they shouldn’t in the first place, they certainly shouldn’t continue to exists year after year, decade after decade because theoretically, people would wise up and by taking advantage of it, remove it from occurring.

Neither of these is true, of course.

Probably the definitive report on this anomaly was published Ben Jacobsen and Sven Bouman. You can read their working paper from July 2001 in the free trading resource section (Reports & Articles).

They not only look at the pattern of stock market returns, they turn over every stone looking for an explanation. They can’t find any. But they do have some interesting things to say nonetheless:

Based on the old market saying Sell in May and go away (or the Halloween indicator), we find that there is a substantial difference between returns in the period May-October and the remainder of the year. In fact our evidence shows that while during the period November - April returns are large in most countries, average returns in the period May-October are not significantly different from zero and are often even negative.

We found that in many countries the Sell in May effect cannot be a January effect only.