Ebook Do Dividends Signal Earnings? The Case Of Omitted Dividends
The payment and size of dividends have long been matters of debate in corporate finance. Under conditions of symmetric information and taxes, dividends have been dubbed a puzzle [Black (1976)]. Several authors model dividend policy under the assumption that information is distributed asymmetrically between managers and investors. Bhattacharya (1979, 1980) argues that firms pay dividends because dividends signal the private information of managers and thus help market participants value the firm. Ambarish, John and Williams (1987) suggest that high value firms choose investment and dividends jointly to separate themselves from low value firms. In other words, dividends are not a residual payment as implied by classical finance theory. John and Williams (1985) and Ambarish, John and Williams (1987) predict a positive association between dividends and stock prices. John and Nachman (1987) describe dividends as a “coarse signal of earnings.” Miller and Rock (1985) argue that once the investment decision of a firm is made, unanticipated dividends signal changes in earnings and cash flows. These models differ in the details of their assumptions and approach, but reach the same broad conclusion: firms pay dividends to convey information to investors that cannot be conveyed costlessly and credibly in other ways.
Empirical evidence supports the signaling function of dividends. Asquith and Mullins (1983) find that the initiation of dividends has a significant positive impact on the firm’ s stock price. They interpret their evidence as consistent with the signaling hypothesis in that managers use dividends to communicate private information to investors, and investors react favorably. Richardson, Sefcik and Thompson (1986) report similar evidence.
Healey and Palepu (1988) investigate the impact of dividend initiations and omissions on stock prices along with changes in earnings. They find that initiations are associated with positive stock price reactions and omissions with negative stock price reactions. They also document that stock price reactions to dividend announcements are positively related to subsequent earnings changes. Ofer and Siegel (1987) report that security analysts revise their earnings forecasts in response to unexpected dividend changes. The greater the unexpected dividend change, the greater is the forecast revision.
Venkatesh (1989) examines the effect of dividend initiations on the information content of earnings announcements. He reports that the average price reaction to announcements of earnings is smaller after the firm starts paying dividends than before. Venkatesh interprets the result as consistent with dividend signaling. He also documents that the variability of stock returns is lower in the period after dividend initiation; the reduction primarily consists of lower residual risk rather than lower correlation with the market. Venkatesh suggests that after dividends begin, investors accord less importance to non-dividend information that previously could have caused price reactions. Thus, the availability of superior information from dividend announcements may result in lower variability of returns. Venkatesh concludes, “This evidence supports the notion that investors view dividends as an information-transmission mechanism.”
Howe and Lin (1992) report an inverse relationship between bid-ask spreads and dividend yields across stocks. They interpret the relationship as supportive of dividend signaling. In this paper, we investigate the effect of dividend omissions on stock return volatility. If Venkatesh’s interpretation of decreases in volatility following dividend initiations is correct, then we should observe increases in volatility following omissions because of the removal of the dividend as an information source. We also examine the effect of dividend omissions on security-analyst forecasts of earnings. If analysts use dividend information when they estimate earnings, a lack of dividends should make their estimates less precise. Less precise estimation should result in greater disagreement among analysts. Hence, we measure the precision of estimates as the variance of forecasts across different analysts.
Download
PDF Ebook Do Dividends Signal Earnings? The Case Of Omitted Dividends
Posted in :
