The partial equilibrium technology rich global 15-region TIMES Integrated Assessment Model (TIAM) is used to assess climate policies in a very uncertain world. Stochastic optimizations are run with four possible climate sensitivities and two development rates, and full resolution of uncertainties in 2040. These assumptions are in line with those of the Energy Modeling Forum.

While a 3^oC target - for both the peak and the long term equilibrium temperatures - is achievable at moderate cost, the smallest achievable temperature increase is close to 1.9^oC, albeit at a very large cost. More severe temperature targets would require additional CO₂ abatement potential that is currently not yet seen as realistic.

The more detailed analysis of the optimal hedging strategy with a 2.5^oC target reveals that hydroelectricity, sequestration by forests and CH₄/N₂O reduction are optimal early robust actions. However, nuclear plants and capture and storage of CO₂ do not belong to robust abatement strategies. Moreover, the uncertainty on the GDP growth rates has very little impact on robust decisions. Finally, no perfect forecast strategy is able to reproduce the hedging strategy, hence the relevance of using stochastic programming.

Sensitivity analyses are undertaken on: the date of resolution of uncertainties, the exogenous radiative forcing, the very long term emissions, the price elasticities of demands, and nuclear development.