EU climate and energy policy beyond 2020 : are additional targets and instruments for renewables economically reasonable?

The European Council has proposed to stick to a more ambitious GHG target but to scrap a binding RES target for the post-2020 period. This is in line with many existing assessments which demonstrate that additional RES policies impair the cost-effectiveness of addressing a single CO2 externality, and should therefore be abolished. Our analysis explores to what extent this reasoning holds in a secondbest setting with multiple externalities related to fossil and nuclear power generation and policy constraints. In this context, an additional RES policy may help to address externalities for which firstbest policy responses are not available. We use a fully integrated combination of two separate models the top-down, global macro-economic model E3MG and the bottom-up, global electricity sector model FTT:Power - to test this hypothesis. Our quantitative analysis confirms that pursuing an ambitious RES target may mitigate nuclear risks and at least partly also negative non-carbon externalities associated with the production, import and use of fossil fuels. In addition, we demonstrate that an additional RES target does not necessarily impair GDP and other macro-economic measures if rigid assumptions of purely rational behaviour of market participants and perfect market clearing are relaxed. Overall, our analysis thus demonstrates that RES policies implemented in addition to GHG policies are not per se welfare decreasing. There are plausible settings in which an additional RES policy may outperform a single GHG/ETS strategy. Due to the fact, however, that i) policies may have a multiplicity of impacts, ii) the size of these impacts is subject to uncertainties and iii) their valuation is contingent on individual preferences, an unambiguous, "objective" economic assessment is impossible. Thus, the eventual decision on the optimal choice and design of climate and energy policies can only be taken politically.