بازنگری پازل سود سهام: آیا همه قطعات در حال حاضر مناسب است؟

This paper revisits the dividend puzzle, described here as questions about the relevance of dividend policy and how managers should (and do) determine dividend policy. We examine theoretical and empirical research on dividends and share repurchases because they are the principal mechanisms by which corporations disburse cash to their shareholders. Despite a voluminous amount of study, researchers still do not have all the answers to the dividend puzzle. However, we are closer to its resolution. We also do not have definitive answers to why managers choose one method of cash distribution over the other. Solving the dividend puzzle may depend on understanding the effects of various market imperfections or frictions. Because various imperfections affect firms differently, dividend policy may vary substantially from one firm to another. Models that consider the competing frictions on a firm-specific basis offer promise for resolving the dividend puzzle.

A main principle of corporate finance is that managers should make decisions that lead to maximizing the wealth of their shareholders as reflected in share price. Much debate exists about the role, if any, of dividend decisions on share prices. Both academics and corporate managers continue to disagree about whether the value of a firm is independent of its dividend policy. Despite exhaustive theoretical and empirical analysis to explain their pervasive presence, dividends remain one of the thorniest puzzles in corporate finance. Miller and Modigliani (1961), hereafter called M&M, first posed this puzzle in their classic paper. They provide a compelling and widely accepted argument for dividend irrelevance if some well-defined conditions are met. M&M frame their analysis in the context of a perfect capital market with rational investors. Their premise is that valuation depends only upon the productivity of the firm's assets and not the form of payout. The irrelevance argument implies that no matter how much care managers take in choosing a dividend policy for their firm, the chosen policy has no beneficial impact on shareholder wealth. Thus, all dividend policies are equivalent. Early research by Black and Scholes (1974), Miller (1986), and Miller & Scholes, 1978 and Miller & Scholes, 1982 support the dividend irrelevance argument. Once we leave M&M's idealized world of economic theory and enter the real world, the issue of dividend irrelevance becomes more debatable. Such market imperfections as differential tax rates, information asymmetries between insiders and outsiders, conflicts of interest between managers and shareholders, transaction costs, flotation costs, and irrational investor behavior might make the dividend decision relevant. Researchers responded to M&M's dividend policy's irrelevance conclusion by offering many competing hypotheses about why companies pay dividends and why investors pay attention to dividends—the “dividend puzzle.” In assessing the contributions provided by researchers, Black (1976) concluded, “The harder we look at the dividend picture, the more it seems like a puzzle, with pieces that just don't fit together.” Several years later, Feldstein and Green (1983, p. 17) echoed Black's sentiments by stating, “The nearly universal policy of paying substantial dividends is the primary puzzle in the economics of corporate finance.” Miller (1986) also recognized that the observed preference for cash dividends is one of the “soft spots in the current body of theory.” The profusion of theories led Ang (1987, p. 55) to observe, “Thus, we have moved from a position of not enough good reasons to explain why dividends are paid to one of too many.” These observations suggest that solving the dividend puzzle has been neither simple nor obvious. This paper revisits the dividend puzzle in order to determine whether all of the pieces now fit. Because the dividend literature is large, we limit our review to some salient theories and evidence.3 We focus on dividends and share repurchases because they are the principal mechanisms by which corporations disburse cash to their shareholders. Because some people argue that companies should buy back their shares rather than pay or raise dividends, we examine share repurchases as an alternative mechanism for distributing cash to shareholders. The structure of the paper is as follows. Section 1 discusses several theories on why companies pay dividends and summarizes some empirical evidence on each explanation. Section 2 provides survey evidence on dividend policy. In Section 3, we summarize various explanations for share repurchases and associated evidence. Section 4 presents the results of survey research on share repurchases. In Section 5, we examine the firm's choice between paying cash dividends and repurchasing common stock. Section 6 describes the trends involving dividends and share repurchases. Section 7 contains our conclusions about the dividend puzzle.

Despite a voluminous amount of research, we still do not have all the answers to the dividend puzzle. We also do not have definitive answers to why managers choose one method of cash distribution over the other. Although dividends and share repurchases are similar in many ways, they are not perfect substitutes. For example, both are costly signals, but tax and signaling effects differ between the two methods of distributing cash. Developing models that describe the choice between paying dividends and repurchasing stock continues to be a fertile area for further research. While not fully solving the dividend puzzle, theoretical and empirical studies over the past four decades have provided additional puzzle pieces that move us closer in the direction of resolution. In reality, there is probably some truth to all of the explanations of why corporations pay dividends or repurchase stock at least for some firms. Although evidence shows that fewer corporations are paying dividends, a firm's distribution policy still matters because it can affect shareholder wealth. Solving the dividend puzzle appears to rest with better understanding the effects of various market imperfections or frictions. Despite some mixed results, the evidence suggests that managers can use dividends and repurchases as useful tools to create value for shareholders due to various market imperfections. Concentrating on one piece of the puzzle at a time (i.e., a single imperfection) fails to provide a satisfactory resolution because the puzzle contains multiple pieces. A promising approach to resolving the dividend puzzle involves combining the various pieces and understanding their interactions. Lease et al. (2000, p. 179) conclude, We believe that the lack of empirical support for a particular dividend policy theory is the result of problems in quantitatively measuring market frictions and the statistical complications in dealing with the myriad interactive imperfections that likely affect individual firms differentially. In other words, since each firm faces a combination of potentially different market frictions with varying levels of relevance, the optimal dividend policy for each firm may be unique. If each firm has a uniquely optimal dividend policy, we should not be surprised that significant statistical generalizations still elude researchers. A problem with most existing theories is that they fail to consider the potentially complex interactions among the various imperfections. Another problem is that each theory typically takes a “one-size-fits-all” approach by trying to generalize the findings. Not surprisingly, Frankfurter and Wood (1997, p. 31) conclude, dividend policy “… cannot be modeled mathematically and uniformly for all firms at all times.” Because various imperfections affect firms differently, dividend policy may vary substantially from one firm to another. Thus, not all firms should have the same distribution policy. Researchers have identified the key factors that tend to affect the dividend policies of firms in a real-world setting. Clearly, some factors may be unimportant to some firms and overlap and/or interact with other factors. We have listed the key factors below. 1. Market imperfections or frictions: taxes, asymmetric information (signaling), agency costs, transaction costs, and flotation costs. 2. Behavioral considerations: irrational behavior of investors, behavioral needs of shareholders, and habitual behavior of firms. 3. Firm characteristics: profitability, investment opportunities, size, availability of cash, and anticipated future earnings and cash flows. 4. Managerial preferences: desire to smooth dividends and to avoid later dividend reductions. Researchers have also identified the key factors that tend to affect firm decisions to repurchase shares of common stock. Some of these factors are similar to those affecting the dividend policy decision (e.g., lack of profitable investment opportunities and signaling), while others are not (e.g., change ownership structure and use of shares to compensate managers or employees). Once again, some factors may be unimportant to some firms, and they may overlap and/or interact with other factors. To deal with multiple factors affecting dividend policy, Lease et al. (2000) propose a competing frictions model that includes all previously discussed market imperfections, except investor irrationality. They illustrate how a firm's dividend payout policy changes as a function of its life cycle. Based on the interactions of the market frictions, they show an implied dividend policy for five stages of the dividend life cycle: start-up, IPO, rapid growth, maturity, and decline. Their model suggests that a firm should pay no dividends during the start-up and IPO stages but should pay a low, growing, and generous dividend during the three latter stages, respectively. Although the model requires testing, it provides a framework for potentially bringing together key pieces of the dividend puzzle. We believe there are two relevant questions related to the dividend puzzle: (1) Do we have all the pieces of the dividend puzzle? and (2) Do all the pieces of the dividend puzzle now fit? Our view is that researchers have identified all the key pieces of the dividend puzzle but need to focus their attention on developing firm-specific dividend models. If this puzzle were a jigsaw puzzle, different firms use different combinations of puzzle pieces to form different pictures of the firm. This results because each firm has different characteristics, managers, and stockholders. One firm may focus on puzzle pieces that together form an Andrew Wyeth-like picture of the firm and another firm may select puzzle pieces that form a Pablo Picasso-like picture of the firm. Is one of them correct and the other incorrect? Not necessarily. Each policy may be appropriate for each firm. Conversely, both firms may possibly need a Wyeth-like structure in which case the value of the Picasso-like firm is lower than necessary. If this view is correct, dividends can matter. However, because no single model specifies which dividend policy results in the highest value of a firm, neither managers nor investors know with certainty if a specific firm's dividend policy is optimal. Therefore, a firm's dividend policy may be viewed as being acceptable within a range. After identifying this range, managers should clearly communicate their firm's dividend policy so that the market can incorporate such information into its valuation process. Only when a firm deviates from this range of acceptable dividend policy, all else equal, should share prices react negatively to a substantial degree.