U.S. stock prices and the dot.com-bubble: Can dividend policy rescue the efficient market hypothesis?

Abstract

This paper thoroughly integrates speculative bubbles to corporate finance literature by focusing on dividend policy issues. More specifically, we examine the importance of dividend policy when testing for speculative bubbles in the S&P 500 equity index on a dataset spanning 1871 to 2014. Given the phenomenon of dividend smoothing, in particular in the U.S., we question the usefulness of observed dividend payments as fundamental factor in testing for bubbles. Circumventing dividend smoothing, we construct hypothetical dividend payouts which are based on reported corporate earnings instead. The empirical evidence presented here indeed suggests that the dividend policy of firms affects testing for speculative bubbles. While the dot.com-bubble—commonly seen as the prime example for a stock price bubble not only in the NASDAQ but also in other, broader equity indices—is detected with the observed dividend series as fundamental factor, this is not necessarily the case with our adjusted dividend time series. Some of our results argue against a speculative price bubble in the broader U.S. stock market in the late 1990s.