Environmental Efficiency Measurement with Translog Distance Functions: A Parametric Approach

We use a flexible parametric hyperbolic distance function to estimate environmental efficiency when some outputs are undesirable. Cuesta and Zofio (J. Prod. Analysis (2005), 31-48) introduced this distance function specification in conventional input-output space to estimate technical efficiency within a stochastic frontier context. We extend their approach to accommodate undesirable outputs and to estimate environmental efficiency within a stochastic frontier context. This provides a parametric counterpart to Färe et al.’s popular nonparametric environmental efficiency measures (Rev. Econ. Stat. 75 (1989), 90-98). The distance function model is applied to a panel of U.S. electricity generating units that produce marketed electricity and non-marketed SO2 emissions.