The Impact of Dividend and Earnings Announcements: A Reconciliation

Two extensive empirical studies have recently attempted to measure the impact of dividends on stock prices. One was a study by Watts' in the Journal of Business which concluded that there is little potential information in dividends. The second was an earlier article by Pettit2 which found that participants make considerable use of the information implicit in announcements of changes in dividend payments. Because of the diametrically opposite conclusions of these studies, and the resulting ambiguity in the size of the dividend effect, the questions raised merit further examination. To this end this note will evaluate the basis for the divergent findings. It should be kept in mind that the original rationale for the conveyance of information through dividend announcements was that reported earnings may not be an accurate reflection of real earnings since they are subject to random nonrecurring factors that cannot be specifically and exactly identified by the investing public. Since management may have greater insight than the rest of the market as to the level of present and future earning power, they may use dividend payments as the medium through which their expectations are conveyed. Ultimately, of course, it is an empirical question as to whether dividends convey new information over and above that conveyed by published earnings. After briefly summarizing the methodology used in the two studies in Section II, Section III explores for the potential differences in each and offers some empirical evidence to reconcile the differential results.