Accounting for global warming risks: Resource management under event uncertainty

Optimal management of atmospheric pollution is discussed with a special emphasis on the uncertainty concerning the occurrence of undesirable events associated with the greenhouse effect. The uncertainty considered here stems from our ignorance of the exact pollution level required to trigger the event rather than from the genuinely stochastic nature of the processes involved. Taking atmospheric pollution level as the state variable, it is found that uncertainty implies the existence of an equilibrium interval, within which the emission rate of the greenhouse gases should be kept equal to the natural removal rate of these gases. Processes initiated outside the equilibrium interval must converge monotonically to its nearest endpoint. The determination of the interval requires no knowledge of the optimal policy. In contrast, ignoring event occurrence risk implies a single equilibrium level, attracting the optimal process from any initial level.