This paper analyses the incentives to participate in and the stability of international climate coalitions. Using the integrated assessment model WITCH, the analysis of coalitions’ profitability and stability is performed under alternative assumptions concerning the pure rate of time preference, the social welfare aggregator and the extent of climate damages. We focus on the profitability, stability, and “potential stability” of a number of coalitions which are “potentially effective” in reducing emissions. We find that only the grand coalition under a specific sets of assumptions finds it optimal to stabilise GHG concentration below 550 ppm CO2-eq. However, the grand coalition is found not to be stable, not even “potentially stable” even through an adequate set of transfers. However, there exist potentially stable coalitions, but of smaller size, which are also potentially environmentally effective. Depending on the assumptions made, they could achieve up to 600 ppm CO2-eq. More ambitious targets lead to the collapse of the coalition.