Using Data Envelopment Analysis to Assess the Relative Efficiency of Different Climate Policy Portfolios

Within the political, scientific and economic debate on climate change, the process of evaluating climate policies ex-ante, during and/or ex-post their lifetime, is receiving increasing attention from international institutions and organisations. The task becomes particularly challenging when the aim is to evaluate strategies or policies from a sustainability perspective. The three pillars of sustainability should then be jointly considered in the evaluation process, thus enabling a comparison of the social, the environmental and the economic dimensions of the policy's impact. This is commonly done in a qualitative manner and is often based on subjective procedures. The present paper discusses a data-based, quantitative methodology to assess the relative performances of different climate policies, when long term economic, social and environmental impacts of the policy are considered. The methodology computes competitive advantages as well as relative efficiencies of climate policies and is here presented through an application to a sample of eleven global climate policies, considered as plausible for the near future. The proposed procedure is based on Data Envelopment Analysis (DEA), a technique commonly employed in evaluating the relative efficiency of a set of decision making units. We consider here two possible applications of DEA. In the first, DEA is applied coupled with Cost-Benefit Analysis (CBA) in order to evaluate the comparative advantages of policies when accounting for social and environmental impacts, as well as net economic benefits. In the second, DEA is applied to compute a relative efficiency score, which accounts for environmental and social benefits and costs interpreted as outputs and inputs. Although the choice of the model used to simulate future economic and environmental implications of each policy (in the present paper we use the FEEM RICE model), as well as the choice of indicators for costs and benefits, represent both arbitrary decisions, the methodology presented is shown to represent a practical tool to be flexibly adopted by decision makers in the phase of policy design