This study re-examines the impact of the differential taxation of dividends
and capital gains on assets' prices. Our analysis shows that the time hori
zon used to define and measure the dividend period is a key issue when inte
rpreting the empirical results. Our results indicate that most of the retur
n variation previously attributed to dividends is not because of a cross-se
ctional variation in returns, but due to the time-series variation in retur
ns around the dividend payment. In light of the lack of cross-sectional ret
urn variation, interpreting the higher return around the dividend distribut
ion as a tax effect is problematic.