The paper discusses the coupling of non-linear non-convex damage costs due to climate change with a cost-efficiency analysis based on a technical-economic linear programming model like MARKAL and studies the implications for the computation of cooperative and non-cooperative solutions.

Our empirical analysis of climate damages based on different world emissions levels and paths prove a) that the dependency of damages on the trajectory of emissions may be neglected, so that the only relevant variables are the cumulative emissions in each country, and b) that a linear relationship links regional damages and cumulative global emissions.

Based on these results, cooperative and non-cooperative equilibria can be much more easily calculated by solving local optimization problems in a case where international trade effects of GHG policies are neglected: given the linearity of damage functions, each country chooses its non-cooperative strategy by considering only the part of its own damage cost due to its own emissions; in the cooperative case, each country takes into account its contribution to the damages done to all countries.

Of course, any cost-benefit conclusion that will be produced by this approach is fully dependent on the damage functions. Also, this approach may be extended to the case where trade effects are modeled.