Estimating A Policy Model Of U.S. Coal Supply

It is becoming evident that traditional methods of econometric analysis are inadequate in estimating-the long-run supply of minerals. Econometrics is limited to extracting estimates from observations of the theory predicts that the least cost deposits will be exploited first.1 The estimation of parameters from past data then must lead to biased projections of the future. Mining conditions in the future will be worse than in the past and it is the impact of the future conditions on cost that we need to know. In this paper we present a methodology for dealing with this problem. We estimate a "cumulative cost" function that yields the long-run marginal cost of mining as a function of cumulative output produced over time.2 This is relatively easy to do for coal mining in the United States where exploration for new deposits is not a major factor in supply. The first step of the procedure relates cost to the rate of output and to the geological conditions of the mine. The second step combines geological information on remaining deposits with the cost function to yield the cumulative cost function. A central theme of both these steps is the interaction of economics with geology. The best deposits are selected first and in each step of the estimation allowance must be made for this fact. The problem of depletion is part of the general problem of structural change. The change through depletion in the nature of the exploitable mineral stock is only one factor that calls for an analysis sensitive to the changing circumstances of the industry. The factors impinging most heavily on the coal industry today are the strict environmental regulations legislated