Exhaustible Resource Models: The Value of Information

This paper investigates the effect of future uncertainties on current policy decisions relating to resource depletion. The analysis is based on an extension of the Hotelling-Nordhaus exhaustible resource model. It is assumed that existing extraction capacity will be gradually replaced as the more expensive backstop technologies become economically competitive. Under plausible conditions, it is found that short-run investment and pricing decisions are insensitive to distant future uncertainties. Hence the expected value of perfect information is low. This indicates that deterministic scenarios may provide a good approximation to a stochastic problem, and this may be used to simplify the work of policy analysis. A numerical model of the U.S. economy is employed to illustrate these ideas.