Temperature targets, deep uncertainty and extreme events in the design of optimal climate policy

Abstract

We study optimal climate policy consistent with the constraint that average global temperature remains below 1.5 ∘ C relative to pre-industrial levels. We consider a holistic representation of uncertainty including traditional risk, deep uncertainty and stochastic arrivals of climate-related disasters. Using robust control methods, we derive optimal emission and carbon tax paths and calculate when temperature exceeds the target in the absence of the constraint. We show that policy under deep uncertainty requires strong action now relative to pure risk but the policy stringency is reversed later. Preliminary estimates suggest that the COVID-19 impact on attainment of the temperature target is negligible.