The economic meaning of Data Envelopment Analysis: A ‘behavioral’ perspective

We reconsider the motivation of Data Envelopment Analysis (DEA), the non-parametric technique that is widely employed for analyzing productive efficiency in academia, the private sector and the public sector. We first argue that the conventional engineering motivation of DEA can be problematic since it often builds on unverifiable production axioms. We then provide a dual viewpoint and highlight the ‘behavioral’ interpretation of DEA models. We start from a specification of the production objectives while imposing minimal structure on the production possibilities, and construct tools to meaningfully quantify deviations of observed producer behavior from optimizing behavior. This brings to light the economic meaning of DEA, provides guidelines for selecting the appropriate model in practical research settings, and prepares the ground for instituting new DEA models. We also provide an empirical application that demonstrates the practical relevance of our arguments. We hope that our insights will contribute to the further dissemination of DEA, and stimulate public sector applications of DEA that build on its behavioral interpretation.