| The governments of the U.S. and Western Europe have traditionally chosen one of two means of controlling natural monopoly: public ownership or regulation of a private-owned firm. This dissertation seeks to explain theoretically the effects of ownership and control on the production decisions of the firm and to test the predictions of the theoretical model. The introduction contains a description of nationalization and privatization policies in the U.K., France and the U.S. to provide a context for the model. The first essay analyzes an optimal mechanism design problem in which the owners of the firm (either the government or the shareholders) contract with a manager who has private information about the firm's cost function and takes an unobservable action that affects the firm's costs. The model predicts that the privately-owned, regulated firm will use relatively more capital than is efficient, and the publicly-owned firm will use relatively more labor, because of differences in the objective functions and the operating constraints of the two types of owners. The second essay describes an empirical test of the implications of the model analyzed in the first essay, using data on U.S. water utilities. Maximum likelihood methods are used to estimate the econometric model, which includes a private information parameter not directly observed in the data. The null hypothesis that the parameters of the model are the same for publicly-owned and privately-owned, regulated firms can be rejected, but there is no clear evidence supporting the model's predictions of factor input biases. |
