Sell in May, but when should you buy?

Commentary: Stock market has a cure for the summertime blues

By John Prestbo

NEW YORK (MarketWatch) — One problem with that catchy old adage, “Sell in May and go away,” is that it doesn’t tell you when to come back.

The answer, it turns out, is October.

At least, that’s the message from an analysis of the Dow Jones Industrial Average over its 117-year history. (The Dow’s birthday is May 26.) A few other possibly helpful details are revealed as well.

The first step was to calculate the average market returns (dividends excluded) over rolling six-month periods. Because of the market’s long-term tendency to rise, the average returns are positive in all 12 periods.

Sure enough, the lowest return is for the May-October period. Here are the results:

Jan-June	Feb-July	March-Aug	April-Sept	May-Oct	June-Nov
3.03%	3.51%	5.05%	3.03%	1.97%	2.99%
July-Dec	Aug-Jan	Sept-Feb	Oct-March	Nov-April	Dec-May
4.06%	3.44%	2.11%	4.12%	5.01%	3.94%

One might quibble as to whether selling in May should be followed by buying again in July, as the six-month period beginning then sports a none-too-shabby average return of 4.06%, which ranks fifth highest. But the returns recede again after that. Read more: Sell in June before the market's swoon.

Clearly, though, history favors closing the summer house and returning to the market salt mine by October — not only for the 4.12% average return in the six months starting then but also for the attractive returns in subsequent periods.

These rolling six-month spans shed some insight on the variations in annual results. Simply add the average returns in each column above, and you get:

Jan-Dec	Feb-Jan	March-Feb	April-March	May-April	June-May
7.09%	6.95%	7.16%	7.15%	6.98%	6.93%

(Note that the average returns are the same even if the starting and stopping months are reversed.)

Again, the adage holds up. Twelve-month periods beginning in May and June have two of the three lowest returns. Taken together, these 12-month returns average 7.05%. That is a baseline to keep in mind as you shape your expectations.

The second step in the analysis was to compute monthly returns for the Dow. These are the results:

Month	Average
January	1.02%
February	-0.19%
March	0.83%
April	1.21%
May	-0.09%
June	0.35%
July	1.53%
August	1.19%
September	-1.21%
October	0.20%
November	0.86%
December	1.22%

Selling in May — which has the second-worst monthly average — doesn’t look quite so wise in this context because presumably you would miss the month with the best average, July. On the other hand, you also would skip the year’s worst month, September.

Another way to look at it is that selling in May and returning in October would deprive you of 1.77 percentage points of return from the average year’s total of 6.92% (the sum of all 12 average months). That leaves you with an average annual return of 5.15%.

If you have the feeling that these sets of average returns don’t square with your recent experience, you are right. Average returns for the 21st century are in some instances significantly different from the full-history versions.

For example, the May-October period gets much worse, switching from a positive 1.97% to a negative 1.62%. October-March, by contrast, strengthens from 4.12% to 4.8%, but it is only one of two periods that get better. Ten of the 12 six-month spans have lower average returns, and the annual average shrinks to 3.43% from 7.05%.

Similarly, the monthly average returns are rejiggered. Eight get worse, with three of those switching from positive to negative. October, November, March and April returns are fattened by an average of 90 basis points each.

Statistically, the 12 full years of this century are too small a sample to support conclusions of any value. But they do show that the market can’t be depended upon to deliver average results consistently. Indeed, the monthly returns span more than 7,000 basis points, which provides plenty of room for times of un-average behavior.

Still, more than a century of data demonstrates that at least this market maxim hangs on for a good reason. It’s just that you don’t know until it’s too late whether this is the year to follow it. Read more: 4 reasons why 'sell in May' won't play out this year.

John Prestbo is retired as editor and executive director of Dow Jones Indexes, now part of S&P Dow Jones Indices, in which Dow Jones & Co., publisher of MarketWatch, holds a small interest. Mr. Prestbo also is an adviser to MarketGrader Capital, which scores stocks on the basis of fundamental factors and chooses components of the Barron’s 400 Index. 

source

please give me comments thanks