We address whether environmental regulation has potentially adverse effects on firms' economic performance, using the EU ETS as a case study. This is a concern considering that inclusion in the EU ETS imposes additional costs by forcing firms to cover all emissions by allowances.
As a measure of the economic performance, we estimate cost efficiencies and their determinants using a stochastic cost frontier (SCF) analysis. Firms' efficiency is measured relative to an efficiency frontier, controlling for observable determinants such as input use, ETS membership or export status. However, this efficiency frontier may differ across industries and treatment status. To enable the comparison of cost efficiencies across industries and treatment groups, we estimate a meta frontier, which envelopes the individual frontiers, and which (by assumption) is independent of the treatment.
Based on this model, we then analyze the causal impact of the EU ETS on the cost (in)efficiencies based on a difference-in-differences (DiD) framework as well as using semi-parametric matching. This means that we compare the change in the cost efficiency for ETS firms relative to the change in emissions of firms in the same industries, but which are not subject to the EU ETS.
The results of industry-specific stochastic cost frontier estimations reveal that the average firm cost efficiency in the period 2003-2014 varies between 55% to 8\% at the level of the 2-digit industry classification. This suggests a considerable cost saving potential in all industries of the German manufacturing sector. We find that, depending on the industry, total costs decrease by about 0.5 to 3% per year. We find that firms covered by the EU ETS are less cost-efficient than unregulated firms, and this result emerges from both the SCF and the DiD analyses. The results of meta frontier analysis suggest that in the German manufacturing sector the EU ETS has significantly reduced cost-efficiency in all three phases, by around 15%. Our sub-sample study suggests that effects vary considerably across industries. The weakest effect is found in the first phase of the EU ETS, whereas the strongest is found in the third phase with a reduction of around 25%. We find stronger negative effects on efficiency for firms that make considerable investments into R&D, suggesting that an over-investment in emissions abatement may drive our results. Given the recent increase in allowance prices, this investment may pay out after all.