Estimating the environmental efficiency of European cross-countries using a non-radial general directional distance function

This article deals with the evaluation of the environmental efficiency of 29 EU countries in the time period from 2000 to 2016. The classic version of the additive data envelopment (DEA) model is modified by the non-radial general directional distance function for distance measurement. The proposed model makes it possible to reduce the input variables (labor, capital, energy), to reduce the undesired output variables (greenhouse gas, GHG) and to increase desired output variables (gross domestic product, GDP) in the same time and non-proportionally. The results show the development of the average value of scaling factor beta, which measures the inefficiency, which has been decreasing from the average since 2004. Reducing inefficiency of work and reducing GHG emissions are a significant contributor to this. Also, this is related to the land mark environmental policy of reducing GHG emissions after 2005 in line with the Kyoto Protocol. Throughout the period of 17 years, according to GDP-GHG index, the environmental efficiency were 14 EU countries with best practice, but on the other hand, the least environmental efficient countries were Latvia, Poland and Romania.