Climate policy without intertemporal dictatorship: Chichilnisky criterion versus classical utilitarianism in dice

Unlike discounting and the damage function, the social welfare function has not received so much attention in the debate on climate economics. An important challenge has been to combine efficiency and equity considerations in a single social welfare framework. The Chichilnisky criterion is one way to resolve this. We consider its implementation in the climate-economy model Dynamic Integrated Climate-Economy (DICE), and compare results for different damage functions, namely the standard one in DICE and the one proposed by Weitzman implying potential large climate damages at high temperature increases. We calculate optimal climate policy for different parameter settings and compare the results with those under the green golden rule (only final utility matters) and classical utilitarianism (no discounting). Optimal emission abatement trajectories turn out to be very different between standard discounted utilitarianism, classical utilitarianism and Chichilnisky specifications. The results are very sensitive to the damage function, the climate sensitivity parameter and especially the "Chichilnisky weight" given to utility of generations in the far future. We discuss conditions and reasons for preferring either classical utilitarianism or the Chichilnisky criterion, and conclude that a critical factor is the time horizon used in climate policy analysis. Adopting sustainable preferences as formalized by the Chichilnisky criterion in climate policy analysis has the advantage that the very long-term implications of greenhouse gases in the atmosphere on the environment and human welfare are not downplayed.