Using Dividend Discount Models to Estimate Expected Returns

Abstract

Because stock returns are so noisy that historical average returns cannot be measured with precision, analysts often turn to dividend discount models to estimate the expected return on equity and the equity risk premium. In recent years, concern has arisen that models based on dividend yield and dividend growth may produce misleading estimates of expected returns because of the impact of share repurchases. As a result, Damodaran [2013], among others, suggests replacing the dividend yield with the “total yield,” defined as the sum of the dividend yield and the repurchase yield. This short article addresses the extent to which the total yield model is, in fact, superior to the dividend growth model for estimating expected returns, and concludes that, overall, the dividend growth model is the better choice.