AN ECONOMETRIC STUDY OF DEMAND AND SUPPLY OF GASOLINE: DISEQUILIBRIUM ANALYSIS

The aims of this research are to formulate an energy model to examine the current energy situations and to explore the empirical acceptability of the model. More specifically, this study intends to develop an analytical framework for the U.S. gasoline market during the period 1960 to 1979.;In the empirical analysis, the assumption of disequilibrium gasoline market is tested to see whether the disequilibrium hypothesis is acceptable or not. Both two-stage-least-squares (2SLS) and maximum likelihood estimation (MLE) methods are used for estimating the model. In MLE the maximization of the likelihood function requires the use of a nonlinear maximization algorithm that can be used with a number of initial starting values for the parameters to ensure that a maximum has been reached. The techniques used in this study are the quadratic hill-climbing algorithm developed by Goldfeld and Quandt and the pattern search method by Davidon-Fletcher-Powell. It is empirically found that the hypothesis of a disequilibrium gasoline market during the 1960-1979 period is supported against the hypothesis of equilibrium.;Throughout the last two decades the U.S. gasoline market was in disequilibrium because of either imperfect price adjustment (in most of the 1960s and some periods in the 1970s) or price control by government (in some periods in the 1970s). Especially in the 1970s when the price controls were effective, it is theoretically found that the regulations were binding for some firms and nonbinding for the others depending upon each refining firm's accessibility to price controlled old (cheap) oil and base period profit margin. This has been resulted since the regulations were firm-specific. Three major policy changes--the Emergency Petroleum Allocation Act (EPAA), the Entitlement Program, and the Energy Policy and Conservation Act (EPCA)--are also surveyed based on the model developed.