The welfare properties of climate targets

Two approaches are predominant in climate models: cost-benefit and cost-effectiveness analysis. On the one hand, cost-benefit analysis maximises welfare, finding a trade-off between climate damages and emission abatement costs. On the other hand, cost-effectiveness analysis minimises abatement costs, omits damages but adds a climate constraint, such as a radiative forcing constraint, a temperature constraint or a cumulative emissions constraint. These constraints can be applied from today onwards or only from 2100 onwards, allowing to overshoot the target before 2100. We analyse the impacts of these different constraints on optimal carbon prices, emissions and welfare. To do so, we fit a model with abatement costs, capital repurposing costs (stranded assets) and technological change on IPCC and NGFS scenarios. The welfare-maximizing scenario reaching 1.5°C in 2100 has almost no net negative emissions at the end of the century (-2GtCO2/y). A constraint on cumulative emissions has the best welfare properties, followed by a temperature constraint with overshoot. A forcing constraint with overshoot has insufficient early abatement, leading to a substantial welfare loss of $29 Trillion, spread out over the century. As to the paths reaching 2°C, all cost-effectiveness analysis abate too late, but the welfare impact of this dynamic inefficiency is milder. Again, a forcing constraint with overshoot scores worst.