The competitive firm with simultaneous price and output uncertainty

Analysis of the competitive firm with simultaneous price and output uncertainty has been very limited, and those analyses that exist are generally conducted in the context of hedging. Yet, because of incomplete forward or futures markets, many firms do not have the opportunity to hedge. This dissertation attempts to fill this gap in the literature by undertaking a study of a competitive firm facing simultaneous price and output uncertainty that does not have the opportunity to hedge any of its output.; The firm is assumed to be risk-averse and to maximize the expected utility of profits. Output uncertainty is modeled in two alternative forms: additive and multiplicative. After laying out both models, the optimal output with both price and output uncertainty is compared with the optimal output which would result if one or both sources of uncertainty is absent. We find that, for both models, in any of the latter cases optimal output is higher than when both types of uncertainty exist.; Next, comparative statics results are derived for both models.{09}In each case, a general set of results is obtained and, in addition, results are obtained for some specific simple cases. In the analysis, the Arrow-Pratt measures of risk aversion are supplemented by the Ross measures of risk aversion. While the results display more indeterminacies than for either price or output uncertainty alone, they are intuitively plausible and do not provide a basis for choosing between the additive or multiplicative form of output uncertainty.; Consequently, an empirical analysis is conducted in order to investigate this issue. Estimating equations are derived for both models by exploiting the properties of the indirect expected utility function and making use of the envelope theorem. The models are estimated using a standard and readily available data set. A series of hypotheses is tested for the additive model, and the additive and multiplicative models are compared. We conclude that the data are more effectively modeled by the additive model. Hence, both on theoretical and empirical grounds, we conclude that there is no justification for neglecting the additive model in favor of the multiplicative model.