The Long-Run Performance Following Dividend Initiations and Resumptions: Underreaction or Product of Chance?
Posted: 30 Aug 2001
We examine the long-term stock performance following the initiation and resumption of stock dividends during the period from 1927 to 1998. We show that the post-announcement abnormal returns are significantly positive for equally weighted calendar time portfolios, but become insignificant when the portfolios are value-weighted. Moreover, the equally weighted results are confined to firms who initiate or resume dividends during the period from 1964 to 1998. We find no evidence of abnormal returns prior to 1964. We provide a rational explanation for the post-1964 positive price drift, after observing significant declines in the loadings of the three Fama and French risk factors during the post-announcement period. Cross-sectionally, we find that firm-specific price drifts are negatively related to contemporaneous changes in the three factor loadings: firms with the highest reduction in loadings are also the ones with the most positive simultaneous abnormal performance. In general, for any random sample, declines in risk factor loadings are associated with positive abnormal returns, due to a combination of unexpected strong firm performance and reductions in the corporate cost of equity. This suggests that the positive price drift observed in this paper may be a sample-specific result of chance.
Our paper makes an important methodological contribution: we caution future researchers of long-term anomalies to be aware of situations in which the price drift is in the opposite direction of any concurrent post-announcement changes in the loadings of the three Fama-French risk factors. In such instances, the chance explanation is quite plausible and must be carefully considered as a viable alternative to the more common behavioral interpretation.
Keywords: Long-term event studies, market efficiency, dividend policy
JEL Classification: G14
Suggested Citation: Suggested Citation