Estimating technical efficiency under unobserved heterogeneity with stochastic frontier models: application to northern German dairy farms

This article uses panel data estimation techniques to examinetechnical efficiency of individual dairy farms in northern Germany.To the extent that agricultural production is characterisedby heterogeneous production conditions, estimation techniquesthat do not account for unobserved heterogeneity produce biasedefficiency estimates. We therefore estimate a number of conventionalpanel data models and Greene's recently proposed ‘true’random-effects model, as well as an extension of the model toascertain the effects of different specification on the productionfunction and efficiency estimates. Our results appear to supporttheoretical expectations and previous findings according towhich a specification that is both time-variant and also controlscorrelations between unobserved heterogeneity and the explanatoryvariables avoids heterogeneity bias and thus ensures consistentefficiency estimates.