Estimating persistent and transient technical efficiency and their determinants in the presence of heterogeneity and endogeneity

We develop an estimation procedure that generates consistent estimates of the technology parameters, long-run (persistent) and short-run (transient) technical inefficiencies and the marginal effects of their determinants for the stochastic frontier model developed by Colombi et al. (2014, Journal of Productivity Analysis 42, 123) and Kumbhakar et al. (2014, Journal of Productivity Analysis 41, 321). Our approach accounts for three sources of potential endogeneity: (i) unobserved heterogeneity; (ii) simultaneity of input use with both types of technical efficiency; (iii) potential correlation of the noise term with the regressors. Using this approach we examine the effect of direct payments and farm size on the persistent and transient technical efficiency of French crop farms before and after the European Union's Common Agricultural Policy decoupling reform of 2003. Our results show that subsidy payments per hectare of utilised agricultural land had a significant positive effect on persistent technical efficiency and a significant negative effect on transient technical efficiency during the period before decoupling. For the period after the reform, the effect of subsidies is found to be significantly negative for persistent technical efficiency and insignificant for transient technical efficiency. The overall effect of subsidies on technical efficiency is found to be negative in both periods, albeit substantially lower in the period after decoupling. The effect of farm size on technical efficiency is found to be significant only for the period prior to the reform: it reduced persistent technical inefficiency but increased transient technical inefficiency during that period.