A Stochastic Frontier Model with short-run and long-run inefficiency random effects

This paper presents a new stochastic frontier model for panel data. The model takes into account firm unobservable heterogeneity and short-run and long-run sources of inefficiency. Each of these features is modeled by a specific random effect. In this way, firms’ latent heterogeneity is not wrongly modeled as inefficiency, and it is possible to disentangle a time-persistent component from the total inefficiency. Under reasonable assumptions, we show that the closed-skew normal distribution allows us to derive both the log-likelihood function of the model and the posterior expected values of the random effects. The new model is compared with nested models by analyzing the efficiency of firms belonging to different sectors.