|
|
The Impact of Dividend Initiation on the Information Content of Earnings Announcements and Returns Volatility P.C. Venkatesh Journal of Business April 1989
Executive Summary Dividends act as a substitute for quarterly earnings announcements. That is after the initiation of a dividend, the firm experiences lower price movements around earnings announcements. Thus, dividend do act as a signal. Moreover, after initiation overall volatility appears to fall.
Much research has looked at dividends as a means of signaling to shareholders. For example, Asquith and Mullins (1983) as well as Richardson, Sefcik, and Thompson (1986) interpret the positive stock price move on a dividend initiation as a confirmation that managers use the news as a means of signaling to investors. In this article, Venkatesh "examines another implication of this 'dividend signaling argument': are dividends and earnings announcements substitutes or compliments?" The paper begins off with a literature review. Specifically, "Richardson et al (1986) observe that dividend-clientele theories predict a shift in clienteles." Consistent with this they find an increase in volume around initiations. However, the authors suggest that this is due to signaling and NOT merely the clientele effect causing portfolio shake-ups. In this literature review, Venkatesh also cites the Marsh and Merton 1987 paper where they argue that firms change dividends for industry reasons. Shareholders also use earnings announcements in their valuation decisions. This has been shown by many authors: example Miller and Rock (1985 where dividends are residuals payments and hence any unexpected change is the result of a shock to earnings. Asquith and Mullins (1986) suggest that dividends MAY be a batter signal than earnings announcements due to managers' ability to manipulate earnings. The current paper is most like that of Haley and Palepu (1988) in which both dividends and earnings are examined. An important difference of the Haley-Palepu (1986) paper and this paper is that this paper looks at more than just the first dividend announcement but "the information-transmission mechanism; that is, investors learn from the initial dividend as well as from subsequent dividends."
Data: Venkatesh looks before and after the dividend initiations for CRSP firms (generally small) that that began paying dividends during the 1972-1983 period. Interesting notes:
Methodology: The author compares absolute price reactions in a predividend period to the post dividend period. The reason for using absolute returns is that the study is interested in looking at the information content. Must make some adjustment for positive/negative returns. Taking the absolute value is easier to interpret than looking at squared terms. Returns were defined as raw returns, but in footnote 7 it is also mentioned that market-adjusted (market model) were also investigated. Pre and post comparisons are done with cross-sectional distributions
Results: (Table 2) The basic finding is that the average price reaction to earnings announcements is smaller in the post-dividend period. Using the Wilcoxon signed rank test, the results are significant at the 1% level. This result is true regardless of whether the dividend announcement is before or after the earnings announcement. The author also looks at firm-specific characteristics such as where the stock trades, how long the stock has been traded, size. This is because previous research has shown that earnings announcements at large firms have lower informational content (Atiase 1985). In this paper however, neither size, trading history, nor market had any significant impact.
|
|
|
************************************************************** Due to several requests, the non-Financial Pages have been de-emphasized but remember Finance is supposed to be fun so I will be keeping these "fun" pages online. **************************************************************
Copyright 2002, 2003 for FinanceProfessor.com |