Estimating emission coefficients and mass balances using economic data: A stochastic frontier approach

In this article, we propose a stochastic formulation of mass balances that impose physical constraints on production technologies. The estimation of the model involves a composed error term structure that is commonly applied in the literature on stochastic frontier analysis of productive efficiency. Moreover, we discuss how ordinary least squares and maximum likelihood methods can be used to estimate emission coefficients based on the proposed model and compare the results to estimates based on linear programming techniques from input–output analysis. In contrast to previous approaches, our model allows one to estimate the physical limitations to production possibilities in the presence of statistical noise and depends on substantially weaker data requirements. We apply our approach to estimate emission coefficients for sulfur dioxide and carbon dioxide using a sample of fossil‐fueled power plants in the United States.