Operationalizing climate targets under learning: An application of cost-risk analysis

Cost-Effectiveness Analysis (CEA) determines climate policies that reach a given climate target at minimum welfare losses. However, when applied to temperature targets under climate sensitivity uncertainty, decision-makers might be confronted with normatively unappealing negative expected values of future climate information or even infeasible solutions. To tackle these issues, Cost-Risk Analysis (CRA), that trades-off the costs for mitigating climate change against the risk of exceeding climate targets, has been proposed as an extension of CEA under uncertainty. Here we build on this proposition and develop an axiomatically sound CRA for the context of uncertainty and future learning. The main contributions of this paper are: (i) we show, that a risk-penalty function has to be non-concave to avoid counter-intuitive preferences, (ii) we introduce a universally applicable calibration of the cost-risk trade-off, and (iii) we implement the first application of CRA to a numerical integrated assessment model. We find that for a 2°-target in combination with a 66 % compliance level, the expected value of information in 2015 vs. 2075 is between 0.15 % and 0.66 % of consumption every year, and can reduce expected mitigation costs by about one third. (iv) Finally, we find that the relative importance of the economic over the risk-related contribution increases with the target probability of compliance