The Capital Gain Lock-in Effect and Return Seasonality

Abstract

This paper conducts an empirical analysis of whether stock return seasonality is related to the capital gain lock-in effect. In the spirit of a recent paper by Ayers, Lefanowicz and Robinson (2003), it tests two hypotheses that are implied by the asset pricing theory developed by Klein (2001). The first hypothesis is that in the month of January, stocks with losses should have large returns and stocks with gains should have relatively poor returns. The second is that in the last few months of the year this effect is reversed, i.e., stocks with losses should have low returns compared to stocks with gains. Both hypotheses are well supported, in the precise way expected, in tests on the cross-section of stock returns for January and for the last few months of the year. These results also provide an alternate explanation of the disposition effect.