Why do firms pay dividends? In Dividend Policy, Agency Costs, and Earned Equity  authors Harry DeAngelo, Linda DeAngelo, and René Stulz document that, for the 25 largest long-standing dividend payers in 2002, a decision to retain earnings instead of paying dividends would have resulted in firms with little or no long-term debt and enormous cash balances, far outstripping any reasonable estimate of their attractive investment opportunities. Had they not paid dividends, those firms would have had cash holdings of $1.8 trillion (51 percent of total assets), up from $160 billion (6 percent of assets), and $1.2 trillion in excess of their collective $600 billion in long-term debt. Paying dividends also prevented these firms from having significant agency problems -- the incremental costs and inherent conflicts of having managers make decisions for investors -- because the retention of earnings would have given managers command over an additional $1.6 trillion without access to better investment opportunities and with no additional monitoring.

Agency theory assumes that large-scale retention of earnings encourages behavior by managers that does not maximize shareholder value. Dividends, then, are a valuable financial tool for these firms because they help avoid asset/capital structures that give managers wide discretion to make value-reducing investments. The evidence presented in this paper uniformly and strongly supports this view of dividend policy.

This view also makes sense when one considers the rationale behind agency theory. Managers acquire control over corporate resources either from outside contributions of debt or equity capital, or from earnings retentions. From an agency perspective, one advantage of contributed capital is that it comes with additional monitoring, because rational suppliers of outside capital will not be forthcoming with funds at attractive prices if they believe that managers' policies merit low valuations.

Earned equity is not subject to the same ongoing, stringent discipline. Accordingly, potential agency problems are higher when a firm's capital is largely earned, since the more a firm is self-financed through retained earnings, the less it is subject to the ongoing discipline of capital markets.

Looking forward, firms with a greater demonstrated ability to self-finance most likely are also firms with a greater ability to internally fund projects that reduce stockholder wealth. Such potential waste is limited by ongoing distributions that reduce the cash resources under managerial control. A regular stream of dividends reduces the threat of agency problems that become increasingly serious as earned equity looms ever larger in the firm's capital structure.

For publicly traded industrials during 1973-2002, the proportion that paid dividends was high when the ratio of earned equity to total common equity (or to total assets) was high. It fell with declines in either ratio, coming close to zero when a firm had little or no earned equity. The authors consistently find a highly significant relationship between the decision to pay dividends and the ratio of earned equity to total equity (and to total assets), even after controlling for firm size, current and recent profitability, growth, leverage, cash balances, and dividend history.

The relationship between earned equity and the decision to pay dividends is significant economically as well as statistically, with the difference between high and low values of earned equity translating to a substantial difference in the probability of paying dividends. In fact, earned equity has an economically more important impact on the dividend decision than do profitability or growth, variables that are typically emphasized in the literature on empirical corporate payout. Overall, the results support the theory that firms pay dividends to mitigate the agency costs associated with the high cash/low debt capital structures that would eventually result if they did not pay dividends.

Les Picker

Abstract -----

Why do firms pay dividends? If they didn't their asset and capital structures would eventually become untenable as the earnings of successful firms outstrip their investment opportunities. Had they not paid dividends, the 25 largest long-standing 2002 dividend payers would have cash holdings of $1.8 trillion (51% of total assets), up from $160 billion (6% of assets), and $1.2 trillion in excess of their collective $600 billion in long-term debt. Their dividend payments prevented significant agency problems since the retention of earnings would have given managers command over an additional $1.6 trillion without access to better investment opportunities and with no additional monitoring. This logic suggests that firms with relatively high amounts of earned equity (retained earnings) are especially likely to pay dividends. Consistent with this view, the fraction of publicly traded industrial firms that pays dividends is high when the ratio of earned equity to total equity (total assets) is high, and falls with declines in this ratio, becoming near zero when a firm has little or no earned equity. We observe a highly significant relation between the decision to pay dividends and the ratio of earned equity to total equity or total assets,controlling for firm size, profitability, growth, leverage, cash balances, and dividend history. In our regressions, earned equity has an economically more important impact than does profitability or growth. Our evidence is consistent with the hypothesis that firms pay dividends to mitigate agency problems.