What do stochastic frontier cost functions tell us about inefficiency?

Abstract

Recent studies using the stochastic frontier approach have suggested high levels of inefficiency in nursing home and hospital facilities. This paper makes two observations. First, one could falsely estimate high levels of inefficiency if a statistical assumption--zero skewness of the random component of the cost residual--is violated. Second, an inefficient industry may be difficult to distinguish either statistically or visually from an industry free of inefficiency. If available, panel data yields more robust estimates of cost differences among nursing homes or hospitals.