A Dynamic Analysis of Industrial Energy Efficiency and the Rebound Effect

Energy efficiency improvement (EEI) is known as a cost-effective measure to meet energy, climate change and sustainable development targets. However, behavioral responses to such improvements referred to as energy rebound effect may change the emission and energy saving gains expected from EEI. Despite broad consensus around the existence of energy rebound effect, significant divergence exists on how to measure this effect, which matters in order to set up realistic energy and climate policies. In this study, we propose a new approach to measure the energy rebound effect in both the short and the long run using a two-stage procedure, applied to a firm-level data set from Swedish manufacturing industry over the period 1997–2008. In the first stage, we use data envelopment analysis (DEA) in order to obtain energy efficiency scores. In the second stage, we estimate energy rebound effect using dynamic panel data regression model. We show that in the short run, partial rebound effects exist within all manufacturing sectors, meaning that the rebound effect decreases, but does not totally offset, the potential energy and emission savings expected from EEI. In the long run, our results suggest that rebound effects decrease within most of the sectors.