Climate policy under sustainable discounted utilitarianism∗

Empirical evaluation of policies to mitigate climate change has been largely confined to the application of discounted utilitarianism (DU). DU is controversial, both due to the conditions through which it is justified and due to its consequences for climate policies, where the discounting of future utility gains from present abatement efforts makes it harder for such measures to justify their present costs. In this paper, we propose sustainable discounted utilitarianism (SDU) as an alternative principle for evaluation of climate policy. Unlike undiscounted utilitarianism, which always assigns zero relative weight to present utility, SDU is an axiomatically based criterion, which departs from DU by assigning zero weight to present utility if and only if the present is better off than the future. Using the DICE integrated assessment model to run risk analysis, we show that it is possible for the future to be worse off than the present along a ‘business as usual’ development path. Consequently SDU and DU differ, and willingness to pay for emissions reductions is (sometimes significantly) higher under SDU than under DU. Under SDU, stringent schedules of emissions reductions increase social welfare, even for a relatively high utility discount rate.