A martingale characterization of the price of a nonrenewable resource with decisions involving uncertainty (Q1064957)

This paper presents a general dynamic model of decisions regarding a nonrenewable resource (such as a natural energy resource) under uncertainty. The decisions involve determining optimal consumption and exploration rates over time. The uncertainty is about the time of occurrence of a significant event such as possible resource exhaustion, a new stock discovery or a development of a producible substitute. Optimal stochastic control of the resource level process yields a stochastic process of the discounted imputed price of the resource. Necessary and sufficient conditions are identified in order for this process to be a martingale, a submartingale or a supermartingale. The results imply a complete characterization of conditions under which the resource price is expected to rise at a rate equal to, slower than or faster than the rate of discounting, which is a problem of some economic interest. The model and results are then specialized to three cases of uncertainty studied separately in the literature, yielding a review of most of the known and some new results. One shortcoming of the paper is that only one type of an uncertain event can be considered and it can occur only once; a more realistic model would simultaneously permit multiple occurrences of various kinds of uncertain events. Another limitation of the model is that it does not incorporate any learning by the decision maker over time.