Dynamic Efficiency Model: An Analysis of Efficiency and Deregulation in the U.S. Electricity Industry

The static production efficiency model and the dynamic duality model of intertemporal decision making using a parametric approach have been continuously developed but in separate directions. The parametric approach takes statistical noise into account, which can provide accurate measures in a stochastic environment. In this study the static shadow price approach and the dynamic duality model of intertemporal decision making are integrated to formulate theoretical and econometric models of dynamic efficiency in the presence of intertemporal cost minimizing firm behavior. The dynamic efficiency model measures the firms inefficiency and accounts for allocative and technical inefficiencies of net investment and variable inputs.A theoretically consistent method to measure the decomposition of dynamic total factor productivity growth in the presence of inefficiency is used to construct the dynamic total factor productivity growth that is adjusted for deviations from the long-run equilibrium within an adjustment-cost framework. This leads to the specific decomposition of the efficiency gain or loss as a source in explaining the growth.The dynamic efficiency model is implemented empirically using a panel data set of 72 U.S. major investor-owned electric utilities using fossil fuel-fired steam electric power generation during the time period of 1986 to 1999. Given a quadratic functional form to specify a value function of the dynamic programming equation, the dynamic efficiency model is applied to estimate the cost inefficiency of electric utilities and to decompose the cost inefficiency into allocative and technical inefficiencies of net investment and variable inputs, and to characterize the cost structure of the electric utilities under dynamic adjustment. The decomposition of the dynamic total factor productivity growth in the presence of inefficiency is estimated as a means to evaluate how different electric utilities participate in the deregulation of electricity generation. Finding that the electric utilities are perfectly technically efficient in net investment, the empirical results of this study are summarized as follows. First, electric utilities with relatively high technical inefficiency of variable input demand in states adopting a deregulation plan improve the performance of the utilities. The results indicate that the difference of allocative efficiencies of variable inputs and net investment by the group of electric utilities affected by the deregulation plan is not significant. Second, the estimates of the input price elasticities indicate the substitution possibilities among the inputs. Most electric utilities in this study are over-capitalized in the production and the estimates suggest the supporting evidence of increasing returns to scale in the production of the electricity industry. Finally, the results indicate that electric utilities located within states with the deregulation plan had lower average annual total factor productivity growth than those located outside of these states. The lower total factor productivity growth of electric utilities located within states with the deregulation plan results from the lower technological progress and combined scale effects but they have a higher combined efficiency effect than those located outside of these states. The results suggest that electric utilities located within states with the deregulation plan anticipated deregulation to a greater extent than those located outside these states.