By blending seminal literature on non-spatial stochastic frontier models with key contributions to spatial econometrics we develop a spatial autoregressive (SAR) stochastic frontier model for panel data. The specification of the SAR frontier allows efficiency to vary over time and across the cross-sections. Efficiency is calculated from a composed error structure by assuming a half-normal distribution for inefficiency. The spatial frontier is estimated using maximum likelihood methods taking into account the endogenous SAR variable. We apply our spatial estimator to an aggregate production frontier for 41 European countries over the period 1990-2011. In the application section, the fitted SAR stochastic frontier specification is used to discuss, among other things, the asymmetry between efficiency spillovers to and from a country.
JEL: C23; C51; D24; E23