Free Allocation and the Endowment Effect in Cap-and-Trade Systems: Evidence from the European Electricity Sector

Independence of installation-level emissions from endowments of allowances allocated for free constitutes a necessary condition for the cost-effectiveness of a cap-and-trade system. A causal relationship between allocations and emissions suggests the presence of an endowment effect induced by free allocation and indicates a loss in cost effectiveness. The issue is relevant to the EU's Emissions Trading System (EU ETS), where a large share of the total allocation occurs for free. This paper tests for the presence of an endowment effect among European electricity sector plants as regulated under the EU ETS by evaluating whether growth in plant-level emissions of power generators changed due to a switch from free allocation to full auctioning. To overcome the endogeneity of allocations I exploit a natural experiment inducing exogenous variation in the allocation of allowances to power producers. While electricity producers located in EU-15 countries were subject to full auctioning starting in 2013, free allocation continued under the so-called 10c rule in eight member states. I apply a matched difference-in-differences research design to a unique EU-wide plant-level dataset of emissions and technical characteristics, constructing a synthetic control group. I find no evidence of a general endowment effect. However, there is some evidence in favor of an endowment effect for a sub-sample of small emitters.