A comparative analysis of the trading behavior of the participants in the first three phases of the EU emissions trading system

The European Union emissions trading systems EU ETS in the third phase (2013-2020) differs significantly from the first phase. The emissions cap is now EU-wide in place of national caps, whereas auctioning is the default method for allocating allowances instead of the free allocations. The aim of this paper is to empirically quantify the changes in the allowance trading behavior between the first, the second and the third phase of the EU ETS. European Union transaction log data from the first and the second trading phase is matched with the years 2013-2014 from the third period. The question to be answered here is whether, holding constant other factors, trading behavior has changed. In a first step, by descriptive statistics, differences of the trading behavior between the first and the third trading phase are identified by comparing firms with an allocation surplus with firms with no allocation surplus. In a second step, binary choice regression models with panel data are used to figure out the determinants of the propensity to engage in trading. The increase in trading participation between these three phases is mainly based on a reduction of the allocation cap, the auctioning of a part of the allowances instead of free allocation and learning effects. In a third step the opportunity costs of those firms that do not sell their allocation surplus are quantified by directly using the data of the European Union Transaction Log linked with annual average transaction price data. This approach is novel in the literature. The results show that the overall opportunity costs of non-trading are decreasing between the first and the third phase