By now, just about everyone is aware of the market’s seasonal trends. Just as investors now know to “sell in May and go away,” so too have we been trained to buy in November and expect gains in the final two months of the year. But when, exactly, is the time to buy?
First, let’s look at the market’s recent results in November and December. In the past 10 years, the S&P 500 has averaged gains of 1.66% and 1.24%, respectively, in November and December, with an average two-month return of 2.90%. In the 20-year interval (1991-2010), the November and December returns are 1.28% and 1.99%, with a two-month average of 3.27%. These are small numbers on paper, but on an annualized basis they are comfortably ahead of the actual returns in most of the individual calendar years.
Digging a little deeper, we can see where these returns are concentrated. The data show that the best returns within the November-December period have occurred on Thanksgiving week, and in the week before the week in which Christmas falls. Christmas week itself also has been very positive on a 20-year basis, but this is largely due to the strong returns registered from 1991 through 2000.
The table below shows the strength of the returns in these and other periods within the years’ final two months:
While potentially informative, this table comes with one caveat — the 10-year Thanksgiving week return is skewed by the large rally of nearly 11% that occurred in 2008. This brings up the average, but it doesn’t change the fact that the S&P has finished in the green in that week in eight of the past 10 years.
With regard to December, the week that precedes the holiday week has been one of the strongest segments of the two-month interval. In both the 10- and 20-year periods, the return for this one-week period has far exceeded the two-month return on an annualized basis. The odds have been firmly on the bulls’ side in this week, with the market having finished in positive territory in nine of the past 10, and 16 of the past 20, calendar years.
One interesting aspect of the historical data is changes that have occurred over time. In the first half of the 20-year period, Thanksgiving week was relatively quiet, while Christmas week brought nice gains. In the past 10 years, that trend has reversed. At the same time, though, November has gradually moved ahead of December as being the better of the two months.
Naturally, there is a tremendous dispersion of returns over the course of each year. The period from Halloween to Thanksgiving week is particularly notable for its wide range of results, which is scattered from -17.42% in 2008 to 7.44% in 2001. As a result, consider this article a guide for refining entry and exit points rather than a recommendation to go long in any particular period based on historical data alone.
Even more important, the historical results show that even with the market having come so far in the past month, the seasonal winds are firmly at its back. And this comes at a time in which the Europe crisis has been taken off the table (at least for now), the economy appears to be steadying, earnings are robust and the Fed is exceptionally supportive. This indicates that the inevitable pause we’re likely to see in the next week or two will represent an opportunity to buy the dip rather than run for cover.