Shadow prices and marginal abatement costs: Convex quantile regression approach

Marginal abatement cost (MAC) is a critically important concept for efficient environmental policy and management. In this paper we argue that most empirical studies using frontier estimation methods such as data envelopment analysis (DEA) over-estimate MACs. The first methodological contribution of to clarify the conceptual distinction between shadow price MAC in order analyze three sources upward bias due limited set options, inefficiency, noisy data. Our second develop novel approach based on convex quantile regression. Compared traditional methods, regression more robust choice direction vector, random noise, heteroscedasticity. Empirical application U.S. electric power plants demonstrates DEA may be serious problem real-world applications.