Blog - Investing Notes

January 1, 2010 - Backwards January Barometer?

Does the performance of the U.S. stock market in January relate to the performance of the market during the prior year, perhaps due to tax-related effects? Using monthly closes for the S&P 500 Index from December 1949 (using the January 1950 open) through January 2009 (60 years), we find that...

The following scatter plot relates the return for the S&P 500 Index during January (December close to January close) to the return for the immediately preceding year (December close to December close) over the entire sample period. The Pearson correlation between the two series is 0.00, as is the R-squared statistic, indicating no relationship.

Might there be some non-linearity in the relationship that shows up in a ranking?

The following table summarizes the average January return, standard deviation of January returns and percentage of winning Januaries following the 20 worst, 20 middle and 20 best years for the S&P 500 Index. Results suggest that Januaries after bad years tend to be better (and more variable) than Januaries after good years, but the subsamples in this analysis are very small.

A more granular visualization conveys a sense of the reliability of this finding.

The final chart recasts the data in the above scatter plot by ordering the S&P 500 Index annual returns from lowest (-38.5% in 2008) to highest (+45.0% in 1954) over the sample period. The horizontal axis is therefore not time-sequential. This view of the data suggests that January return does not vary systematically with prior-year returns.

In summary, evidence from simple tests suggests no reliable (exploitable) relationship between January return and prior-year return for the U.S. stock market.

For related research, see Blog Synthesis: Calendar Effects.