Researchers have, typically, used data on variables such as dividend payout ratio and divided yields to investigate theories of earnings payout. In this paper, the authors suggest that, at least in some cases, it may be more appropriate to look at earnings retention rather than the more common dividend variables. As an example, the tests of the Investment Opportunities hypothesis in the context of deregulation in the utilities industry is taken. Prior to deregulation, cash generated by the business was largely returned to the stockholders; whereas, subsequent to deregulation, firms have been free to acquire and merge with other firms, as well as to make investments that were either prohibited or frowned upon in the days of regulation. Under these circumstances, the Investment Opportunities hypothesis predicts that firms will reduce their dividend payouts in order to pay for these investments. The impact of deregulation in the utilities industry on firms' dividend policies is investigated. While the evidence from dividend-based quantities is ambiguous, it is found that firms have clearly increased their earnings retention.

Following the publication of Franco Modigliani and Merton Miller's two seminal papers on dividend policy, there has been a considerable amount of research into what determines firms' dividend policies. In this paper, we suggest that these theories may be better investigated by looking at firms' earnings retention policies rather than their dividend policies. We use the example of the Investment Opportunities theory of dividend policy to prove our point. This theory has not been given much attention in the literature. However, it is often mentioned in practice-oriented textbooks, such as Damodaran (2001, 1998), that if a firm has positive-NPV investment opportunities, it should retain earnings; else it should pay them out. This is a simple statement of the theory, and is implied by the Pecking Order Hypothesis (Myers, 1984); it can also be inferred from Smith and Watts (1992). However, direct tests of this inherently reasonable maxim are lacking.

Our examination of the Investment Opportunities Hypothesis focuses on the recently deregulated electric utilities industry. Until 1992, utilities were regulated and restricted in their investment choices. In 1992, with the passage of the Energy Policy Act, utilities were freed from conforming to such restrictions. Investment Opportunities expanded, not only because the onerous restrictions of the PUHCA(Public Utility Holding Company Act were diluted, but also because, with the end of rate regulation, there was no need to worry about assets being declared stranded. Furthermore, since costs had been increasing in the pre-1992 era, regulator-approved rates were always playing catch-up and did not constitute a sufficient return. Release from rate-regulation, therefore, provides an increased incentive to invest. Subsequently, in 1996, the Federal Electric Regulatory Commission (FERC) relaxed some of the standards that it used to evaluate mergers. This provided additional incentives to invest both within the industry and outside it. Consequently, the Investment Opportunities hypothesis predicts that the dividend payout of such deregulated firms would drop. We show that the evidence from the dividend payout is not very clear. However, earnings retention policies of these firms tell a much clearer story.