International greenhouse gas emissions when global warming is a stochastic process

This paper integrates two different but crucial aspects of global warming: free riding and uncertainty. Although certainty equivalence does not hold, an analytical solution is obtained if 'clean up' (e.g. reforestation) is possible. Assuming realistically irreversible emissions, numerical methods are necessary to obtain the value function and the implied strategies, which are shown to be unique (for the calibrated parameters). Although uncertainty stresses the importance of conservation, the incentives to free ride are almost the same. Beyond the particular application, the paper shows how to analyze stochastic, dynamic games. Copyright (C) 2004 John Wiley Sons, Ltd.