Energy Efficiency in Swedish Industry - A Firm-level Data Envelopment Analysis

This paper assesses energy efficiency in Swedish industry. Using unique firm-level panel data covering the years 2001-2008, the efficiency estimates are obtained for firms in 14 industrial sectors by using data envelopment analysis (DEA). The analysis accounts for multi-output technologies where undesirable outputs are produced alongside with the desirable output. The results show that there was potential to improve energy efficiency in all the sectors and relatively large energy inefficiencies existed in small energy-use industries in the sample period. Also, we assess how the EU ETS, the carbon dioxide (CO2) tax and the energy tax affect energy efficiency by conducting a second-stage regression analysis. To obtain consistent estimates for the regression model, we apply a modified, input-oriented version of the double bootstrap procedure of Simar and Wilson (Journal of Econometrics 136(1):31-64, 2007). The results of the regression analysis reveal that the EU ETS and the CO2 tax did not have significant influences on energy efficiency in the sample period. However, the energy tax had a positive relation with the energy efficiency.