Disequilibrium Estimation of the Demand for Copper

In this paper we estimate the demand for refined copper in the United States, taking account of the fact that, during much of the sample period, copper supplies were rationed. The model employed here, which is closely related to the Tobit model, is much simpler than those models previously used for disequilibrium estimation. Our empirical results are consistent with institutional evidence on the existence of rationing and suggest that conventional estimates of the demand for copper, which implicitly assume that the market is always in equilibrium, are severely biased.