Measuring Eco-efficiency Using the Stochastic Frontier Analysis Approach

The concept of eco-efficiency has been receiving increasing attention in recent years in the literature on the environmental impact of economic activity. Eco-efficiency compares economic results derived from the production of goods and services with aggregate measures of the environmental impacts (or ‘pressures’) generated by the production process. The literature to date has exclusively used the Data Envelopment Analysis (DEA) approach to construct this index of environmental pressures, and determinants of eco-efficiency have typically been incorporated by carrying out bootstrapped truncated regressions in a second stage. We advocate the use of a Stochastic Frontier Analysis (SFA) approach to measuring eco-efficiency. In addition to dealing with measurement errors in the data, the stochastic frontier model we propose allows determinants of eco-efficiency to be incorporated in a one stage. Another advantage of our model is that it permits an analysis of the potential substitutability between environmental pressures. We provide an empirical application of our model to data on a sample of Spanish dairy farms which was used in a previous study of the determinants eco-efficiency that employed DEA-based truncated regression techniques and that serves as a useful benchmark for comparison.