|
|
Sell in May and go away? It's never that
simple
Heydon Traub, Boston Business
Journal, May 24, 2002
Various writers and market researchers have noted the striking difference in
returns over history between the six- month period from November to April
(the "good" period), and the six months from May to October (the "bad"
period). The facts show that nearly all of the stock market's gains take
place from November to April. Conversely, returns are close to zero
(certainly less than a money market fund) from May to October.
Some of the reasons behind this may relate to patterns in bonuses, taxes and
savings vehicles. For those who receive part of their compensation as
bonuses, most are paid anywhere from December to March. This is a result of
companies basing bonuses on the calendar-year results, or in some cases in
the form of holiday bonuses. This creates a large cash flow for these
individuals, and since more of bonus money is likely to be saved than salary
payments, a large portion of this is saved, with a sizable amount invested
in stocks or stock funds.
Taxes also favor this time period. Mutual funds often have fiscal years
ending in October. There is an incentive to rid the portfolios of losers
before the Oct. 31 year-end report. In addition, funds and individuals have
an incentive to take their losses for tax purposes toward year-end, which
also creates some selling pressure on stocks in the fourth quarter. Both of
these events create some build-up of cash that often gets re-invested into
stocks around year-end.
Savings plans also cause above-average investment in the first few months of
the year. New employees often have to wait until January to enter a 401(k)
plan, meaning the percent of people saving via a 401(k) plan is highest in
January and tails off during the year. In addition, there is a set dollar
amount each year for a given individual in 401(k) plans that people will hit
as the year progresses, leading to fewer "savers" as the year goes on.
Lastly, IRAs have a deadline of April 15, meaning that most people put
contributions into these from January to mid-April. All of these also lead
to higher stock purchases early in the year, helping to push prices up early
in the year.
So what exactly are the numbers? Keppler Asset Management has done research
on this topic globally. For the U.S. stock market, the average return for
the "good" period is 7.5 percent vs. 1.2 percent for the "bad" period. This
is based on returns from 1969-2001. In order to compare to non-U.S. stock
markets, we only show results since 1969. However, other researchers have
shown similarly startling results for the U.S. when returns go back to 1940.
Looking at markets overseas, the same pattern persists. Looking at Morgan
Stanley Capital International's (MSCI) World Index, a commonly used
benchmark for global investments, "good" period returns are 8.4 percent
while "bad" return periods are actually negative (minus 0.4 percent).
Interestingly, this phenomenon (where "good" period returns exceeded "bad"
period) applied to all 18 markets where MSCI had index returns back to 1969.
As implied by the spread for the World index, the United States was far from
the most notable. Italy won that prize with a spread above 16 percent (14.5
percent vs. 2.2 percent). Denmark had the smallest differential with "good"
at 6.8 percent and "bad" at 5.1 percent.
Moving back to the market that most readers have the greatest interest in --
the United States -- Ned Davis Research takes this thesis to another level
of detail. Its research indicates the optimal "bad" period (based on
history) lasts from the sixth trading day of June to the fifth to last
trading day of October.
The "good" period is the rest of the year. Using these time frames, the
"good" period would have had investments going up by 134 times ($1 becomes
$134) from 1942 to 2001. Amazingly, the bad period would have had a return
of minus 3 percent, meaning $1 becomes $0.97.
Now some people in our business call this data mining, i.e. finding data
patterns that explain the past.
The question then becomes: Will these patterns persist? There are some valid
explanations for what happened and these structures/incentives are still in
place (timing of bonuses, savings, etc.). This at least puts some likelihood
that the pattern will persist.
So what is an investor to do with this information? A key point to note is
that we are either early in the "bad" period or about to enter it. Should
you sell all your stocks? Well, we wouldn't suggest that, as there is no
guarantee this pattern will persist. There certainly have been years where
the market did well in the "bad" period. In addition, for taxable
investments, there are reasons not to dart in and out of stocks taking
capital gains, or worse short-term gains. And a final reason not to sell all
and then re-invest later in the year are the costs associated with it.
If you own mutual funds with sales loads, making these calendar switches
would be a very bad idea. Even if you own individual stocks it can become
costly to move in and out so frequently, locking in definite costs, for only
potentially better prices when you buy back in.
But given where we are in the cycle, it may make sense not to rush into
investing idle cash at the moment. And if you have stocks or funds you are
thinking of selling anyway, this is one reason to do so now instead of
waiting until the end of year. Based on history, don't expect big stock
gains over the next five months.
|
|
