Estimating Variable Returns to Scale Production Frontiers with Alternative Stochastic Assumptions

A stochastic production frontier model is formulated within the generalized production function framework popularized by Zellner and Revankar (1969) and Zellner and Ryu (1998). This framework is convenient for parsimonious modeling of a production function with variable returns to scale specified as a function of output. Two alternatives for introducing the stochastic inefficiency term and the stochastic error are considered, one where they are appended to the existing equation for the production relationship and one where the existing equation is solved for the log of output before the stochastic terms are added. The latter alternative is novel, but it is needed to preserve the usual definition of firm efficiency. The two alternative stochastic assumptions are considered in conjunction with two returns to scale functions, making a total of four models that are considered. A Bayesian framework for estimating all four models is described. The techniques are applied to USDA state-level data on agricultural output and four inputs. Posterior distributions for all parameters, firm efficiencies and the efficiency rankings of firms are obtained. The sensitivity of the results to the returns to scale specification and to the stochastic specification is examined.