Parametric and nonparametric measures of productive performance: The case of private manufacturing in Egypt
Doctor of Philosophy
This dissertation investigates the productive performance of the private manufacturing sector of Egypt, particularly within the context of economic reforms undertaken since the early 1990s. Beginning in 1991, the Egyptian government implemented reforms to achieve macroeconomic stabilization and to promote private sector development. We utilize complementary methodologies to analyze the growth of total factor productivity (TFP) and its decomposition into efficiency and technical change. We use nonparametric methods, including the Malmquist and Tornqvist approaches to measuring TFP, along with stochastic frontier methods to undertake the analysis. It is well known that the stochastic frontier approaches have the advantage of allowing statistical noise in the measurement of efficiency while the nonparametric methods do not. However, we use recently developed bootstrapping methods to study the sensitivity of the nonparametric indices to sampling variation of the frontier they measure. The analysis is based on a newly developed panel of 28 three-digit sectors from 1987/88 to 1995/96. Our analysis indicates that a considerable increase in technical progress is due to the loosening of impediments to competitive forces. We find that growth in total factor productivity would be higher if growth in technical efficiency were higher. Cost savings are found as a result of the increase in total factor productivity. We also use a generalized cost function to study the impact of the economic environment on all sectors. This approach, which subsumes the 'naive' neoclassical cost function as a special case, allows us to assess the impact of institutional constraints on the structure of production, including factor demands, total cost and scale economies. Our findings indicate the presence of substantial distortions in relative prices, and hence on cost efficiency, due to the binding institutional and regulatory constraints. In addition, we find that these inefficiencies are magnified among large firms. We find improvements in efficiency as a result of reforms initiated to remove the constraints.