RDP 9211: Dividends and Taxation: A Preliminary Investigation 2. The Determinants of Dividend Policy

One of the most influential works on dividend policy is by Miller and Modigliani (1961) who argued that, given a firm's investment decision, its dividend policy is irrelevant to its market valuation. The value of the firm is solely determined by the stream of future earnings. The division of earnings between dividends and retained earnings will not affect this value as long as the firm's investment policy is unaltered. There is no unique optimum dividend payout ratio. In the Modigliani/Miller environment dividends should not be paid out at all in a tax regime that discriminates against dividend income and all earnings should be retained when capital gains are taxed favourably. Of course, in practice, this is rarely the case. Other reasons have to be sought to explain the payment of dividends:

(i) Signalling

Signalling theory argues that managers (insiders) have better information about the firm's prospects than shareholders (outsiders), but the latter have an interest in receiving as much information as possible on the firm in order to determine their portfolio allocation.^[1] Dividends convey a signal about a firms' present and future cash flows from investments.

(ii) Uncertainty

Gordon (1962) argues that dividends expected in the near future are less risky than those expected over a longer horizon. Risk averse shareholders will discount expected future dividends at a higher rate to compensate for the higher risk. Therefore investors will not be indifferent between the payment of current dividends and the retention of earnings. Crockett (1988) makes a similar point: if stockholders perceive the stream of cash flows generated by an investment as more uncertain than current dividend payments, they will discount it at a higher rate.

(iii) Agency costs

In agency cost models (Jensen and Meckling (1976)) the payment of dividends restricts the actions of management and so reduces costs from possible control problems caused by the separation of ownership and management.

(iv) Issue and Transaction costs

Modigliani and Miller argue that if a firm wishes to pay a higher dividend without changing its investment and borrowing plans, it may finance the dividend payment by issuing equity. In practice, this will require the company to meet the issue costs. Similarly, shareholders who require current income will be required to sell a portion of their share portfolio and will also be subject to transaction costs.

(v) Taxation

The dividend irrelevance theorem holds in a world without taxes. Differential taxes on dividends and capital gains alter the preference of individual investors for receiving income in one form or the other. This market imperfection is the main focus of this paper. Section 3 examines the effects of tax changes on individuals' preferences in the context of the switch in Australia from a classic double taxation system to a dividend imputation system.

Although the ideas presented above provide some rationale for the observed widespread payment of dividends, there is still considerable debate as to the appropriate framework for thinking about dividend policy. Without a strong theoretical foundation underlying dividend policy, it is not surprising then, that the literature provides little guidance as to the specific factors that managers consider when establishing a dividend policy. There is some agreement that firms have target payout ratios, but are reluctant to change dividends when earnings change, believing that shareholders prefer a steady stream of dividends. However there is little agreement as to the factors that determine target payout ratios. In this paper we focus on two factors that appear to be important in Australia: cash flows and tax policy. Cash flow determines the long-run capacity of firms to pay dividends and the tax regime influences the preferences of shareholders as to the form of their returns. Other factors that may also be important are a firm's liquidity, funds requirements, access to capital markets, costs of securing external finance, income and capital gearing, and industry norms. Section 5 examines the significance of taxation, cashflow and other possible determinants of dividend payments.

Footnote

See Poterba and Summers (1984), Bhattacharya (1979), Miller and Rock (1985), Ofer and Thakor (1987) and John and Williams (1985). [1]