A pr 2 01 5 The Social Cost of Carbon with Economic and Climate Risks ∗

There is great uncertainty about future climate conditions and the appropriate policies for managing interactions between the climate and the economy. We develop a multidimensional computational model to examine how uncertainties and risks in the economic and climate systems affect the social cost of carbon (SCC)—that is, the present value of the marginal damage to economic output caused by carbon emissions. The SCC is substantially increased by economic and climate risks at both current and future times. Furthermore, the SCC is itself a stochastic process with significant variation; for example, the basic elements of risk incorporated into our model cause the SCC in 2100 to be, with significant probability, ten times what it would be without those risks. We have only imprecise information about what parameter values are best for approximating reality. To deal with this parametric uncertainty we perform extensive uncertainty quantification and show that these findings are robust for a wide range of alternative specifications. More generally, this work shows that large-scale computing can enable economists to examine substantially more complex and realistic models for the purposes of policy analysis.