There has, however, been a bit of revival of the January Effect in re-
cent years. Since 2000 the return on small stocks has risen to 1.68 versus
0.21 percent for large stocks. This is not as big as the historical advantage,
but it is far larger than that in the 1990s. Perhaps the poor performance of
the January Effect during that decade caused traders to ignore the phe-
nomenon, and the anomaly has resurfaced. We all await further data.LARGE MONTHLY RETURNS
There are other seasonal patterns to stock returns besides the January Ef-
fect. The monthly returns on the Dow Industrials and S&P 500 Index are
displayed in Figure 18-2. December has been the best month since World
War II for both indexes, but only the fifth-best month since 1885. In strik-
ing contrast, August, which was the best month for the past 116 years, is
actually the second-to-worst month since World War II for the Dow and
third-worst month for the S&P 500 Index. Since the end of World War II
there has been really no evidence of the “summer rally” that used to be
much trumpeted by brokers and investment advisors.
These monthly patterns of returns have a worldwide reach. Al-
though January is a good month in the United States, it is an excellent
month in most foreign countries. The January returns for the 20 countries
covered by the Morgan Stanley Capital Market Index are shown in Figure
18-3. In every country, January returns are greater than average and con-
stitute nearly one-quarter of the annual stock returns abroad. Investor en-
thusiasm in January also seems to infect the neighboring months of
December and February. Well over one-half of all returns outside the
United States occur in the three months of December through February.^7THE SEPTEMBER EFFECT
Summer months have good returns, but after the summer holidays,
watch out! September is by far the worst month of the year, and in the
United States, it is the only month to have a negative return including
reinvested dividends. September is followed closely by October, which,
as Chapter 16 indicated already, has a disproportionate percentage of
crashes.CHAPTER 18 Calendar Anomalies 311
(^7) The data presented in Figure 18-3 are from a value-weighted stock index calculated on large
stocks. As noted previously, there is evidence that smaller stocks experience even higher January re-
turns, so the January returns shown in Figure 18-3 are probably much lower than those that can be
gained in the average stock.