Optimal capital structure and dividend policy in modern corporate finance

Capital Structure

Modigliani and Miller argued that capital structure is irrelevant, all other things being equal, but in the real world those other things are never equal. The factors that are ruled out of MM are neutral taxes, no capital market frictions, symmetric access to credit markets, and that firm finance policy reveals no information. Normally, arguments against the irrelevance of capital structure are based on these factors that MM assumed away (Villamil, n.d.). In the U.S., taxes on dividends are very different from the taxes paid on loan interest. There are transaction and bankruptcy costs; firms cannot borrow and lend at the same rate, and financial policy does reveal information. As such, MM does not hold in the real world, and this implies that capital structure does matter.

That capital structure does matter implies that for every firm there is an optimal capital structure. What that structure might be, however, is dependent on a wide range of factors, many firm-specific and some highly subjective. This results in firms having a wide range of capital structures, even within the same industry, and that might give the impression to the casual observer than capital structure does not matter. Yet, there are in all likelihood some very good reasons for the differences between capital structures of different firms.

The objectives of the firm and the personal taste of management -- especially its risk aversion -- are key elements that contribute to the choice of capital structure. In general, debt has a lower cost than equity, but it increases the risk to the company because it represents an obligation that must be paid from the firm's cash flows before it accrues to the shareholders. In contrast, equity is less risky but is more costly. All firms must make a tradeoff between the two, and firms that have many similarities will choose divergent courses. Consider the examples of Mattel and Clorox. Both firms deal in consumer products and they have relatively stable revenues and earnings that have been sustained for years. Yet, Mattel maintains a capital structure that is weighted to just over 50% debt. Clorox is financed almost entirely by debt. There are a few explanations for this disparity. Clorox pays lower taxes as a result of its policy, and it has a lower cost of capital. In addition, the company's ROE is high (actually, in three of the past five years it had negative equity). Mattel's strategy is more costly, but less risky. It pays more in taxes, and has a higher cost of capital, but does not face situations where it struggles to make debt payments or has negative equity book value.

Bradley, Jarrell and Kim (1984) argued that the question of optimal capital structure comes down to "an empirical issue as to whether or not the various leverage-related costs are economically significant enough to influence the costs of borrowing." This argument states that if the firm's degree of leverage is so high that its credit quality has declined (thereby increasing the cost of debt), then the optimal capital structure lies in a more balanced approach. The implication of this is that perhaps Clorox does not see its cost of borrowing increase despite its very high degree of leverage, or does not see it increase to the point where it eclipses the cost of equity. In addition, it also says that Clorox management has a high degree of risk tolerance as any deviation from the company's revenue or income stability would create a liquidity crisis. For Mattel, higher leverage would increase its borrowing costs. Additionally, Mattel's cash flow is very seasonal, whereas Clorox's cash flow is spread throughout the year, and that alone could account for the difference in risk tolerance between the executives of these firms.

Another wrinkle in the discussion is the issue of dividends. Apple, for example, just declared its first dividend since 1995 (Goldman, 2012). Even without the dividend, the company's stock price cruised well north of $550, blowing out the Gordon Growth Model in the process. The current dividend, at $2.65 per share, remains relatively small compared with both the firm's EPS and its stock price. The increasingly speculative nature of the stock market has all but ruled out the need for dividends, at least in the short run. There are good reasons for this belief in capital gains as the sole source of wealth from equity ownership. The first is that capital markets are exceptionally liquid and with online trading the selling costs are low. For a company like Apple, even the derivatives market is fairly liquid, and any investor can program in a stop loss trade. Thus, investors have the power to limit their downside risk. In addition, globalization has allowed for companies to bathe in the glory of neverending growth in easily-accessed foreign markets. A company like Apple does not even have a large presence in some foreign markets, meaning that growth opportunities are near limitless. As long as it is conceivable that stocks can go up forever, there will be investors willing to take that bet.

In the long-run, of course, there are limits to the amount of new markets that can be opened up, and not all companies go up forever. Investors who are essentially gambling on capital gains are hurt badly when markets collapse, as in 2000-2001 or the more recent recession. Firms that want to attract investors may find that they do need to offer a dividend, in order to at least mitigate the declines in their stock price. That the potential income from capital gains is so much higher might convince investors otherwise, as long as they think the market is only going to go up, but when the market goes south firms with healthy dividends are more likely to outperform.

MM assumed away market signals from capital structure, but they do exist. Apple's dividend announcement was taken by the market as a sign that the company was beginning to shift away from growth mode and into a position as a stable provider of consumer goods, much the way Microsoft did several years earlier. Clorox pays out a great dividend - $2.25 on diluted EPS excluding extraordinary items of $2.06. That signals to the market, along with its high debt, that it knows it has a very stable business. Investors see Mattel with a more conservative capital structure and see a company with some volatility in its revenues and earnings. Firms with no dividends at all are sending the message that they are strictly growth plays. Dividend policy, therefore, does matter, at least from a market signal perspective.