Why individual investors want dividends
Introduction
Our understanding of dividend policy depends on the behavior of individual investors, from the early work of Miller and Modigliani (1961) and Gordon (1961) to the more recent behavioral finance theories. Many empirical papers have documented corporate dividend policy and payments, and have related the policies in various ways to the theories based on the behavior of individual investors. While there appears to be a general agreement that investors like dividends, there has been no empirical study of why individual investors want dividends. We fill that gap by asking individual investors about their attitude towards dividends.
Miller and Modigliani (1961) show that individuals can undo management's decisions on dividend policy in a perfect and complete capital market by either reinvesting dividends or selling off stock, making dividend policy irrelevant. In the United States until recently, as well as in most other countries, dividends have been taxed more heavily than capital gains. The irrelevance theorem in combination with the unfavorable taxation of dividends makes dividends a puzzle. Brealey and Myers (2003) consider the dividend controversy to be one of the “10 unsolved problems in finance.”
Fama and French (2001) find that the proportion of U.S. firms paying cash dividends has fallen from 66.5% in 1978 to 20.8% in 1999.1 Grullon and Michaely (2002) show that firms have gradually substituted repurchases for dividends.2 These findings might be a response to the dividend puzzle. Dividends have gained renewed attention recently. On May 23, 2003, the U.S. Congress passed a “tax relief” bill that includes a major change in taxation of investments.3 Capital gains and dividends will now be taxed equally at a top rate of 15%, eliminating the tax penalty on dividends. During the period after the bill was proposed and before it passed, Microsoft announced that it would start paying dividends for the first time in its 28-year history. Technology companies such as Cisco and Oracle stated that if dividend taxes were eliminated, they would start paying dividends. This has led to a renewed interest in why investors want dividends.
The modern literature on dividend policy has been strongly dominated by economic modeling approaches, both in developing hypotheses and in empirical investigation of dividend policy.4 Brennan (1970) provides an after-tax model of dividend valuation and the Capital Asset Pricing Model that other researchers use. Notably, Black and Scholes (1974) find no evidence of a dividend effect and Litzenberger and Ramaswamy (1982) find a significant effect. Many other papers grapple with this conflict and related hypotheses, but the field of finance has not yet reached a consensus on the effect of dividend policy on firm value. Even though many papers appear later than Black (1976), his belief is still the current opinion (p. 5): “Why do corporations pay dividends? Why do investors pay attention to dividends?… I claim that the answers to these questions are not obvious at all. The harder we look at the dividend picture, the more it seems like a puzzle, with pieces that just do not fit together.”
To shed more light on the dividend puzzle, we surveyed a unique Dutch panel of ordinary families who answer questions on personal finance and consumption matters weekly via an organized website/e-mail link. Since this voluntary panel is accustomed to completing questionnaires and most of the panel members will respond, many of the difficulties of survey research are avoided. Furthermore, a demographic profile of the panel members is available, which allows us to better understand the survey responses and test the dividend theories more fully. To the extent that the characteristics of Western investors are similar, we expect our results to be relevant for other Western countries.
We do not include institutional investors in our survey. We are testing the theories developed over more than 40 years relating to individual investor decisions. If institutional investors are acting in place of their clients, then their portfolio decisions will reflect the preferences of their clients. This is particularly true for managers of investment funds, since the income flows directly to the beneficial owners. Our survey looks at individual investors who hold shares directly and/or through investment funds. The indirect holdings through pension plans are not represented.
Conducting research in the Netherlands on dividend preferences has a special advantage, because the new Dutch tax system does not tax dividends and capital gains differently, while the old (pre-2001) system taxed dividends more heavily than capital gains. This tax environment provides us with an excellent setting to test dividend theories by isolating the tax effect on dividends from other considerations. In this regard our results should be more informative than past U.S. survey results, and also provide a preview of investor preferences in the United States under the new tax law.5
In some previous studies, including Lintner (1956), Baker et al., 1985, Baker et al., 2002, and De Jong et al. (2003), researchers used questionnaires to find out why companies pay dividends. A particularly interesting paper in this vein is Brav et al. (in press), who have surveyed 384 CFOs and treasurers of mostly U.S. companies, and conducted 23 in-depth interviews with executives on their payout policy. They find that financial executives believe that retail investors have a strong preference for dividends, despite the tax disadvantage. Another interesting finding is that financial executives believe that dividends convey management's confidence about the future. However, managers say that they do not pay dividends as a costly signal to convey their firm's true value. Therefore Brav et al. (in press) conclude that they only find modest support for the signaling hypotheses. Their survey finds little support for the agency theories of Easterbrook (1984) and Jensen (1986).
The aim of our paper is to determine what individual investors believe about dividend policy. This question is also examined by Brav et al. (in press), but they do so indirectly by asking financial managers for the view of the investors. The managers in their questionnaire believe that individual investors want dividends, but the reasons are not clear. Therefore Brav et al. (in press) conclude, “At this point we can only speculate about what causes individual investors to prefer dividends”. In this paper we try to fill this gap.
We start by summarizing the various hypotheses and conjectures that have been advanced about investor beliefs and behavior regarding corporate dividend policy. We test their validity as descriptions of investor behavior with direct survey research conducted on a sample of Dutch household members who constitute a voluntary panel that answers personal survey questionnaires on family financial and consumer matters every week. From a sample of 2723 household members we received 555 usable responses from members who hold or recently held common shares and/or investment funds.
Our results unambiguously indicate that individual investors are not indifferent to dividends. The mean score on the question whether they want dividends was answered with an average score of 4.98 on a scale that ranges from 1 (=I do not want dividends) to 7 (=I want dividends), with 4 being the neutral score. 60.5% of the respondents indicate a score of 5 or above, while only 12.3% answer 3 or below. Both the mean and the median score on this question are significantly different from 4 at the 1% level. There is a strong confirmation of the signaling role of dividends. Our results are inconsistent with both the free cash flow theory of Jensen (1986), and the agency theory of Easterbrook (1984). All these results are fairly consistent with Brav et al. (in press).
The results further indicate that transaction costs are an important reason for individuals to like dividends. Investors appear to view dividends as saving transaction costs when they do not reinvest the dividends in the same stocks, e.g. when they want to consume, deposit in a bank account, or reinvest the dividends in a different security. These results are stronger for relatively old, low income and less-educated investors. The uncertainty resolution theory of Gordon, 1961, Gordon, 1962 is not confirmed. Investors seem to consider dividend-paying stocks to be more risky than non-dividend-paying stocks. Overall, investors do not believe that dividend-paying firms are less likely to manipulate earnings, although the notion that dividends work as a guarantee of earnings quality finds some support among older investors.
As one of our most striking results, we find that investors consume significantly less from dividend income than from regular income. Therefore, the behavioral finance theory of Shefrin and Statman (1984) is not confirmed for cash dividends. On the other hand, this theory is confirmed for stock dividends. In case companies cannot pay out a cash dividend, investors prefer the companies to “pay” a stock dividend, rather than no dividends at all, even though in principle stock dividends are no more than stock splits.
Finally, we find that, apart from older and low-income investors, individual investors do not tend to consume a large part of their dividends. Rather, they re-invest their dividend income. This raises some doubt as to whether a reduction or elimination of dividend taxes will stimulate the economy.
The Table A1 summarizes the results of our survey.
2 Theories on why investors want cash dividends, 3 Theories on why investors want stock dividends summarize the theories and hypotheses in the literature on cash and stock dividends, respectively. Section 4 describes the research method and data collection. Section 5 describes survey results and discusses how they relate to our hypotheses. Section 6 concludes.
Section snippets
Theories on why investors want cash dividends
a Miller and Modigliani (1961) show that in a perfect and complete capital market the dividend policy of a firm does not affect its value. A stockholder can replicate any desired stream of payments by purchasing and selling equity. b Transaction costs An investor who wants to receive a regular income from her security holdings has a choice between buying dividend-paying stocks and cashing in the dividends, and buying non-dividend-paying stocks and regularly selling part of her portfolio. For a small individualThe Miller and Modigliani (1961) dividend irrelevance theory
Theories on why investors want stock dividends
k Stock dividends as small stock splits An issue that is closely related to that of cash dividends is the question of why some companies “pay” stock dividends. As every standard textbook in Finance teaches us, stock dividends are nothing more than a small stock split. DeBondt and Thaler (1995) refer to stock dividends as one of the big anomalies in finance. l Transaction costs Stock dividends may have an advantage over cash dividends because they may carry lower transaction costs. This is the case if the ultimate goal of the investor is to re-invest the
Survey methods and CentER Panel
We surveyed individual investors to test the theories discussed in the previous two sections. Survey-based research is becoming more common in finance. Surveys complement research based on large samples and clinical studies, particularly for a question like dividend policy where the beliefs of investors are the basis for most of the theoretical models. Graham and Harvey (2001) argue that large-sample studies often have weaknesses related to variable specification and the inability to ask
Overview of survey respondents
The questionnaire based on the theories discussed in 2 Theories on why investors want cash dividends, 3 Theories on why investors want stock dividends was presented to the 2723 members of the panel of CentERdata on the weekend of October 4, 2002.20
Summary and conclusions
In this paper we have tried to contribute to the solution of the dividend puzzle. Most of the finance theory on dividend policy starts with the behavior of shareholders. The empirical finance literature on this topic either studies share price reactions or surveys corporate executives for their opinions. No one has asked individual investors why they want to receive dividends. In this paper we have tried to fill this gap by submitting a questionnaire on cash and stock dividends to a Dutch
Acknowledgements
The authors thank Marcel Das and Mariëlle Klerks for their help with the CentERpanel survey. Furthermore they are grateful to Susan Belden, Abe de Jong, George Frankfurter, David Hirshleifer, Rez Kabir, Skander Lazrak, Elizabeth Maynes, Moshe Milevsky, Terrance Odean, Lynnette Purda, Sara Reiter, Gordon Roberts, Pauline Shum, Yulia Veld-Merkoulova, Marno Verbeek, participants at the APFA-conference in Hamburg (August 2002), the Multinational Finance Society Conference in Montreal (June 2003),
References (47)
- et al.
Payout policy
- et al.
A survey of behavioral finance
- et al.
The effects of dividend yield and dividend policy on common stock prices and returns
Journal of Financial Economics
(1974) - et al.
Financial decision making in markets and firms: a behavioral perspective
- et al.
The dividend and share repurchase policies of Canadian firms: empirical evidence based on an alternative research design
International Review of Financial Analysis
(2003) - et al.
Subjective measures of household preferences and financial decisions
Journal of Economic Psychology
(1999) - et al.
Disappearing dividends: changing firm characteristics or lower propensity to pay?
Journal of Financial Economics
(2001) - et al.
Dividend policy theories and their empirical tests
International Review of Financial Analysis
(2002) - et al.
The theory and practice of corporate finance: evidence from the field
Journal of Financial Economics
(2001) Corporate governance, dividend payout policy, and the interrelation between dividends, and capital investment
Journal of Banking and Finance
(2003)