Since February ended on a positive note for the S&P 500, we’ve come across a number of stories discussing S&P 500 performance during years in which both January and February closed in positive territory. In nearly all of the stories, it’s mentioned that the S&P 500 has never ended in negative territory for the year when this has occurred. There have been 26 instances since 1945 where both January and February exhibited positive performance. Based on our calculations, excluding dividends and using only price appreciation, there have actually been 25 positive years, and one very slightly negative year. On a price appreciation basis, 2011 ended essentially flat, down 0.03%. Still, it’s an impressive record overall.
Most of the stories, however, failed to discuss the nature of the performance for the remainder of the year after the January/February positive run. Was performance run of the mill relative to other years for the other months? Were most gains for the entire year achieved during January and February? Should investors feel comfortable investing after January/February runs?
Looking at performance for the final 10 months during these January/February episodes, we found that performance for the remainder of the year was exceptionally strong, especially compared to the final 10 months of years in which either January or February (or both) came in negative. Let’s move on to the data.
As mentioned, there have been 26 years since 1945 in which both January and February were positive. Here is a table showing the years of occurrence, the cumulative performance for January and February in those years, and the cumulative performance for the 10 months following in those years:
Year
|
Jan to Feb Return
|
Mar to Dec Return
|
YR Return (Simple Return)
|
2013
|
6.20%
|
????
|
????
|
2012
|
8.59%
|
4.43%
|
13.406%
|
2011
|
5.53%
|
-5.25%
|
-0.003%
|
2006
|
2.59%
|
10.75%
|
13.619%
|
2004
|
2.97%
|
5.85%
|
8.993%
|
1998
|
8.13%
|
17.14%
|
26.669%
|
1997
|
6.76%
|
22.71%
|
31.008%
|
1996
|
3.98%
|
15.66%
|
20.264%
|
1995
|
6.12%
|
26.37%
|
34.111%
|
1993
|
1.76%
|
5.20%
|
7.055%
|
1991
|
11.16%
|
13.63%
|
26.307%
|
1988
|
8.39%
|
3.70%
|
12.401%
|
1987
|
17.36%
|
-13.06%
|
2.028%
|
1986
|
7.40%
|
6.72%
|
14.620%
|
1985
|
8.34%
|
16.61%
|
26.333%
|
1983
|
5.28%
|
11.39%
|
17.271%
|
1975
|
19.01%
|
10.54%
|
31.549%
|
1972
|
4.39%
|
10.77%
|
15.633%
|
1971
|
4.99%
|
5.52%
|
10.787%
|
1967
|
8.03%
|
11.17%
|
20.092%
|
1964
|
3.71%
|
8.93%
|
12.970%
|
1961
|
9.17%
|
12.78%
|
23.129%
|
1955
|
2.17%
|
23.72%
|
26.404%
|
1954
|
5.40%
|
37.59%
|
45.022%
|
1951
|
6.71%
|
9.04%
|
16.349%
|
1950
|
2.56%
|
18.64%
|
21.680%
|
1945
|
7.68%
|
21.40%
|
30.723%
|
The first fact of note from the table: of the 26 years exhibiting positive performance in both January and February, there were only two instances in which the S&P 500 posted a negative performance number for the final ten months, 2011 and 1987. The summer and fall months of 2011 were negatively affected by the European debt crisis and the sovereign ratings downgrade for US debt. Of course, 1987 was the year of the October stock market crash. Even with the devastating crash, the S&P 500 only posted a negative 13% number for the final 10 months, relatively tame compared to many other episodes in market history. Moreover, 1987 produced one of our favorite fun facts: the S&P 500 was actually up for full year 1987 despite the crash.
Now for some summary statistics that show the strength of the cumulative performance for the final 10 months, both in years with positive performance in both January and February and years without:
Jan/Feb Both Positive
|
Jan/Feb (other)
|
|
Average
|
12.00%
|
4.57%
|
Median
|
10.97%
|
4.04%
|
St. Dev
|
10.08%
|
16.18%
|
Min
|
-13.06%
|
-32.12%
|
Max
|
37.59%
|
51.70%
|
As you can see above, average and median performance was much higher over the final 10 months for “positive” years compared to years with either a negative Jan. or Feb. Dispersion or volatility of the returns is also much lower as indicated by the standard deviation, minimum, and maximum summary statistics.
Is there a statistical relationship between strong January/February returns and returns the remainder of the year? There’s a very weak statistical relationship. The R-squared between them is a paltry 0.08. Correlation is -0.298 indicating that there is a negative statistical relationship between the two (i.e. higher cumulative Jan/Feb returns mean lower Mar to Dec returns), but again the strength of the relationship is weak. In layman’s language, just because the first two months are positive in a big way doesn’t mean the final 10 months are going to come in below average relative to the other 25 years, or vice versa.
What can we take away from the exercise? There appear to be reasons for optimism the rest of this year. Cumulative performance for January and February this year came in at 6.2%, slightly below, but close to in-line with the average over the other 26 episodes of 6.8%. There are certainly going to be bumps in the road over the course of any year, but the historical record for performance in years with positive January and February performance is consistently strong. March is off to a good start. Let’s hope the remainder of 2013 doesn’t prove to be the very rare exception.
