UBC Theses and Dissertations
An analysis of the cost of capital hypotheses Spence, John David
The cost of capital has received much theoretical and empirical study in recent years. Two contradictory views have emerged concerning the effect of capital structure on the cost of capital. Writers such as Modigliani and Miller maintain that the cost of capital is independent of the relative proportion of liabilities to owners' equity and depends only on the risk associated with the type of business the firm is in. The opposite view is taken by those writers who support what is known as the traditional view. These writers maintain that judicious use of debt can reduce the firm's cost of capital. The purpose of this paper is to determine which approach appears to be the more accurate in a real world situation. We first investigate the many difficulties associated with the empirical tests which have been applied to evaluate the two conflicting hypotheses. The problems associated with these tests lead us to reject them as a means of resolving the cost of capital controversy. Instead, we choose a theoretical approach. Based on suggestions made by Modigliani and Miller and by the traditional writers, we postulate the way in which debt and equity capitalization rates are expected to respond to increases in the amount of debt in the capital structure. The Modigliani and Miller hypothesis and the traditional hypothesis are studied in detail and computer models for each hypothesis are then developed. The hypothetically determined capitalization rates are used as independent variables in the models to develop relationships between the dependent variables and the debt-equity ratio. The response of the dependent variables to changes in leverage is studied to see if it represents rational investor behavior. Real world factors are introduced into the analysis. Effects of corporate income tax on both the Modigliani and Miller hypothesis and the traditional hypothesis are investigated. In addition, we study the way in which legal restrictions and limited personal liability may restrict the Modigliani and Miller arbitrage process. The three computer models used in the analysis are described in detail in the appendices. These models have been developed so as to be as flexible as possible. All parameters are specified by the user, so the models may be adapted to a variety of situations. Program listings and the output used in our analysis are also included.
Item Citations and Data