Stochastic Production Frontier and Technical Inefficiency: Evidence from Tunisian Manufacturing Firms∗

In this paper, a stochastic frontier production model is estimated on panel data, and technical inefficiency indices are computed for Tunisian manufacturing Þrms over the 1983-1993 time period. The most commonly used one-sided distributions of the inefficiency er-ror term are speciÞed, namely the truncated normal, the half-normal and the exponential distributions. A generalized version of the half-normal, which does not embody the zero-mean restriction, is also ex-plored. For each distribution, the likelihood function and the coun-terpart of Jondrow et al. (1982) estimator of technical efficiency are explicitly stated. Based on our data set, formal tests lead to a strong rejection of the zero-mean restriction embodied in the half normal distribution. Our main conclusion is that the degree of measured in-efficiency is very sensitive to the postulated assumptions about the distribution of the one-sided error term. The estimated inefficiency ∗ We are acknowledged to Tahar Abdessalem, Jean Pierre Florens, Farouk Kriaa, Mohamed Salah Rejeb, Rached Bouaziz and Jean Charles Rochet for helpful comments and suggestions. We are grateful to the Secrétariat dEtat à la Recherche ScientiÞque en Tunisie for the Þnancial support