Investigating the energy efficiency determinants in EU countries by using multi-criteria decision analysis and the Tobit regression model

ABSTRACT This study investigates the effects of several contextual variables, including renewable energy intensity, capital stock per labor, natural resource rent, the share of imported energy in total energy consumption, the ratio of carbon emissions to GDP, population, and energy production on energy efficiency in EU countries. While Tobit regression is used to examine the effects of contextual variables on energy efficiency, Data Envelopment Analysis (DEA), Slack-Based Data Envelopment Analysis (SBM-DEA), and Technique for Order of Preference by Similarity to the Ideal Solution (TOPSIS) are used to calculate energy efficiency values of countries. Four different Tobit regression models are estimated as a function of the censored value for the energy efficiency series. The results show that renewable energy intensity, the ratio of carbon emissions to GDP, and population size have negative effects on energy efficiency. In contrast, regardless of the efficiency level, the share of imported energy in total energy consumption, total energy production, capital stock per labor, and technological progress have positive effects on energy efficiency.