Uncertainty and the Value of Information with Stochastic Losses from Global Warming

We present an uncertainty analysis conducted using CETA-R, a model in which the costs of climate change are specified as Risks of large losses. In this analysis, we assume that three key parameters may each take on "high" or "low" values, leading to eight possible states of the world. We then explore optimal policies when the state of the world is known, and under uncertainty. Also, we estimate the benefits of resolving uncertainty earlier. We find that the optimal policy under uncertainty is similar to the policy that is optimal when each of the key parameters is at its low value. We also find that the value of immediate uncertainty resolution rises sharply as the alternative to immediate resolution is increasingly delayed resolution.