Seasonal patterns in individual stock returns

New insights on monthly stock return patterns. A strategy which selects stocks based on their historical same-calendar-month returns has previously been found to earn an annualized average return of 13%.¹ In other words, in January investors should have the stocks which had the highest returns in past Januaries, in February investors should have the stocks with the highest return in past Februaries, and so on. Unfortunately, this strategy involves a very high turnover, since every month one needs to hold a completely different portfolio.

In a new working paper,² the same authors show that the same-month momentum effect is offset by a reversal effect in subsequent months, e.g. stocks that had the highest returns in past Januaries tend to underperform over the next February to December period. Although this reversal effect may be a bit less strong before costs, it has the benefit of involving considerably less turnover.

In our own research we have been able to replicate these results, making the strategy a potential candidate for inclusion in our models. Like with all calendar-based strategies the economic rationale is a bit of a concern. Likely suspects such as earnings announcements or dividend payments, which tend to occur in the same month every year, do not appear to be the source.

Seasonal patterns in individual stock returns

From the field

Footnotes

Let's keep the conversation going