Are Reinvested Earnings a Substitute for Dividends?

Elliot Goldberg,Eliezer Goldberg
DOI: https://doi.org/10.2139/ssrn.3244048
2018-01-01
SSRN Electronic Journal
Abstract:Modigliani and Miller (M&M) proposed that investors forgo dividends, leaving the money available for reinvestment as retained earnings. This recommendation takes two parts: Proposition III, i.e., a dividend has no impact on market value, and Proposition IV, i.e., that financial policy is of no consequence to investors, and retention or distribution of earnings is purely a management decision (Stiglitz, 1974). Ross & Westerfeld (Ross, Westerfield, & Jaffe, 2010) amplify the M&M proposal by saying that companies should not give up projects that are net present value (NPV)-positive in order to distribute dividends. The purpose of this paper is to analyze non-dividend-distributing companies for positive NPV. I picked companies from the Standard and Poor’s 500 Index and studied their earnings for Internal Rate of Return (IRR) and NPV by using probability indices over the course of ten-year periods. The null hypothesis is that Modigliani and Miller were correct and “the division of the stream between cash dividends and retained earnings in any period is a mere detail.” The conclusion of the paper, however, is that in most cases companies are not successful in generating consistent IRR, or positive Profitability Index, and the null hypothesis should therefore be rejected. The paper also tabulates the assumptions needed for the Dividend Irrelevancy theory to function as published by M&M over 30 years ago. It makes the theory appear more mathematical fata morgana than applied finance. I would like to thank my colleague Jerome Balsam for his suggestions. The conclusion of the paper are, of course my own.
What problem does this paper attempt to address?