Company dividend policies and decisions have been studied extensively and there are various theories explaining how management decides what, if anything, to pay out to shareholders. Factors influencing the dividend decision include the stability of earnings, dividend policy of competing firms, past dividend rates and the information signaling effect. We will consider the latter here.
The Logic of Information Signaling
Managers take dividend decisions seriously. After all, who doles out money without thinking? Accordingly, how much is paid out should tell us something about the company, such as how it is doing now and how the directors think it will do in the future. In other words, quite apart from the money itself, dividends arguably serve a partially independent signaling and informational function.
Companies and their accounting practices are often not particularly transparent, so investors are keen to find out what they can from all available sources. Therefore, analysts and active investors watch for news on dividends and for their part, directors and managers are surely aware that apart from rewarding shareholders, they are also transmitting a signal.
What the Research Tells Us
A seminal study by Merton Miller and Kevin Rock in 1985 stressed that dividend announcements reveal something about the future prospects of a company. Dividends can signal managerial perceptions of the condition of a firm and, in particular, how management thinks company earnings will develop over time. Indeed, earlier work had pointed to rises and falls in stock process in tandem with dividend announcements.
A 2006 article by Michelle Hanlon et al. confirms the value of dividends in conveying relevant information on future earnings. This information is also reflected in stock prices. Naturally, the size of the dividend is also significant, and not just the fact that there is one at all. Dividends vary from minimal to so high that the stock can be worth owning for this reason alone. In the simplest sense, dividend cuts are interpreted as bad news and vice versa. However, the reality is of course not quite so simple.
The Limits of the Research and of the Signaling Function
The research can be quite complex and there is a fair amount of inconsistency and contradiction in the literature. The methodology behind the research is not straightforward and is affected, for instance, by whether "loss firm-years" are included in the sample of companies investigated.
There can also be important differences between industries and sectors, both with respect to theory and practice. For example, one researcher demonstrated that dividend decisions made by commercial banks are quite different than those made by, say, manufacturing companies. The point is that commercial banks are monitored by a central one to ensure that capital reserves, loan loss reserves and so on, are all in order. Clearly, such regulation is reflected in dividend decisions.
Furthermore, it is not just the earnings position of the firm that matters, but the economic efficiency that underlies the earnings. There is empirical evidence to suggest a positive relationship between efficiency and dividend levels. Accordingly, earnings now and in the future may diverge quite substantially. A dividend may, thus, signal something about the present and/or the future and it is not always easy to tell the difference.
The conventional theory is that dividend changes should be followed by changes in profitability in the same direction. However, some major studies have found that this appealing logic does not apply with any certainty. What seems more likely, therefore, is that dividend changes are merely associated with changes in risk and future earnings growth or profitability, but one cannot rely on a clear, ambivalent signaling function in a specific direction.
This may be disappointing, but, after all, sure things in the field of investment are rare. Mostly, there are indications, tendencies and trends. The markets really aren't perfect.
However, There Is Still News You Can Use
However, the trend can be your friend and so can the dividend signals. A study published in 2005 from Brav et al. surveyed financial executives and found that two thirds regard earnings stability as an important factor influencing investment decisions.
In the same year, Caskey and Hamilton found that firms accused by the SEC of financial accounting fraud paid generous dividends when actually committing the fraud, but less and less often generally. Therefore, there were some signals there all right, but they were not very easy to interpret up front.
The lesson for investors is that dividend decisions certainly transmit information and signals above and beyond the mere fact that money will be paid to shareholders. However, the information needs to be interpreted with caution and in context. It is also important for investors to use information from other sources in combination with that conveyed by dividends.
The Bottom Line
The exact information value of dividends remains something of an open issue. The mechanisms at work and the conversion mechanisms of financial and economic realities into dividend policies and decisions are complex and neither fully researched nor completely transparent. All the same, this is undoubtedly a factor that analysts and investors should observe. However, they also need to observe the ongoing controversy on the subject. As the saying goes, "watch this space."