April 26th, 2010
Can you believe it? It’s that time of the year again. I’m not talking about spring flowers, baseball or senior proms. I’m referring to “sell in May and go away“. Friday will be the last trading day of what is supposed to be the best time of the year for stocks, the six months of November-April, and the beginning of the worst, the six months of May-October. Are you getting anxious?
Here’s what CNN Money had to say on May 1, 2008:
The old ‘sell in May’ strategy says that if you invest in the S&P 500 or the Dow industrials during the ‘best six months’ (November through April) and then switch into bonds during the ‘worst six months’ (May through October), you’ll end up with better returns than if you did the reverse. The strategy doesn’t always work, but has been one of the more reliable indicators going back to 1950.
For example, $10,000 invested in the Dow during the ‘best’ period and switched to bonds during the ‘worst’ period in every year since 1950 would result in a compounded $578,413 return, according to the Almanac. The same $10,000 invested in reverse would leave you with a compounded $341.
Perhaps, but I did my own study using the monthly closings of the S&P 500 since 1939, that’s 71 years, and discovered the following:
- May-October (supposedly the bearish period):
- actually was positive more than it was negative: 47 out of 71 times, or 66%.
- Only 9 times did the market end October more than 10% below where it ended in April.
- The greatest declines were in the recent crash of 2008 (-30.08%) and 1946 (-20.90%).
- November-April period (supposedly the bullish period):
- was only slightly more positive being up 50 out of 71 times, or 70.4% of the time.
- The market ended April lower than October by more than 10% seven times, or nearly as often as the bearish period.
- The market declined more than 15% four times during what should be the bullish period.
But here’s the most startling statistic. In 44 out of 70 years, or 62%, what the market did during the 6-month period ending April, it also did in the 6-month period ending the following November. Regardless of whether it closed higher or lower in April than the previous October, the odds were (62% of the time) the trend would continue and it would do the same by the following November.
If the market closes this Friday approximately the same level as it did today, it will have closed 16.97% higher than the close last October. There’s a 62% likelihood that the market will be higher next October than the level it closes at on Friday. It may not be much and it doesn’t mean that there won’t be a correction in the interim but, if it did occur, then there’s a better than 50-50 chance that by the end of October it will have been recouped.
These findings are not dissimilar from those that I found when researching the January effect (see “Dissecting the ‘January Effect’). In it, I wrote “Over the past century, the market has ended the year up 70% of the time so it is also likely that January will be up and, if January is up, then the there’s a good chance that the first week will be up, too.” We find the “go away in May” mantra to be similar.
(click here to see the spreadsheet)