|
|
Bear hibernation: A method behind calendar
myth
Heydon Traub, Boston Business Journal,
November 21, 2003
It's the time of year when the bears hibernate -- or at least when they
normally should. And I am not talking bears that live in the woods. I am
talking about stock market bears -- bears who expect the market to fall.
Historically, we have just begun what is the best series of months for stock
returns.
The best consecutive six months of the year have been November to April.
According to studies by Ned Davis Research (which created all the data I'm
referring to here), since 1900 the S&P 500 Index has had average six-month
returns of 4.24 percent from November to April, excluding dividends. By
contrast, the worst period was May to October, with a 2.28 percent average
return.
And don't think that things have changed in recent years.
The case is even stronger since 1982, when the longest-running bull market
began. Over those 21 years, the best period, the November-April time frame,
averaged returns of 8.52 percent while the worst months, May to October,
returned the same disappointing 2.28 percent.
Don't despair that we are well into November and you have missed the boat.
The best three-month period is just about here. Since 1900, the best
three-month span in terms of returns period is December to February, which
has seen average returns of 2.75 percent. Conversely, September to November
had returns of only 0.23 percent. And again in recent years (since 1982),
those same "good" three months provided gains of 4.79 percent, while the
"bad" three months (August to October) actually generated losses (minus 0.22
percent).
And to complete the picture, the best month for returns has always been
January. Since 1900, January's average return is 1.74 percent for just that
month. October and September are both negative, with October bringing up the
rear with a 0.27 percent decline.
Interestingly, since 1982, February is the best month, with gains of 1.77
percent, with November (1.62 percent) and January (1.57 percent) right
behind it. September and October continue to be the worst, and both are
still negative.
Statistical accidents are not that unusual. Myths like the Super Bowl theory
show great correlation between original NFC teams winning it all and the
stock market going up that year. And although it seems to have worked again
this year, it seems a strange coincidence, and I wouldn't suggest laying
down money on this idea.
But there are valid reasons this seasonality has occurred and may continue
to work. Bonuses are typically paid between December and March, providing
money for long-term savings that often finds its way to stocks. Investors
(both mutual funds and individuals) have incentive to sell their losers as
year-end approaches for taking tax losses or for "window dressing" (some
mutual funds don't want to show losing bets on their year-end reports). This
money raised is not likely all reinvested in stocks right away, but probably
most finds its way back to stocks over the following months.
And 401(k) retirement plans often allow new employees to enroll in January,
creating the month with the largest number of participants in the plans. The
numbers decline as the year progresses and employees turn over.
In addition, some hit the contribution limit during the year, meaning that
the aggregate amount invested each month declines during the year.
Lastly, IRAs have an April 15 deadline, meaning savings are invested during
the first few months of the year.
This activity is supported by a look at the flows into domestic mutual
funds. Since 1978, the month with the largest average inflows is April ($8.1
billion). This coincides with the deadline of April 15 for IRAs. The next
best month is January ($7.4 billion), likely supported by both increases in
401(k) participants and bonuses.
There are seasonal patterns elsewhere, as well. Global equities exhibit
comparably strong returns for similar months. Looking at global stock
returns relative to global bond returns, the best six months are also
November to April (based on returns since 1984). Global stock returns
actually lag global bonds in June, August and September. And we see only 0.1
percent outperformance by stocks in July and October.
And we see some seasonality in the dollar as well. The patterns are less
clear as well as less explainable, but the dollar tends to rise in January.
This strength also occurs against the two other major currencies, the yen
and euro. Since 1972, the dollar has risen over 60 percent of the time in
January.
So the lesson to take from this is clear. If you have long-term savings that
may find its way to the stock market, now is a good time to invest rather
than waiting until next year. And since the dollar tends to rise early in
the year, you may be better off investing in domestic stocks in the near
term.
The key is not to wait until the bears come out of hibernation in April, as
you will likely have missed some favorable returns.
|
|
