Inventories and the short-run dynamics of commodity prices

I examine the behavior of inventories and their role in the short-run dynamics of commodity production and price. Competitive producers of a storable commodity react to price changes by balancing costs of changing production with costs of changing inventory holdings. To determine these costs, I estimate a structural model of production, sales, and storage for copper, heating oil, and lumber. I then examine the implications of these costs for inventory behavior, and for the behavior of spot and futures prices. I find that inventories may serve to smooth production during periods of low or normal prices, but during periods of temporarily high prices inventories have a more important role in facilitating production and deliver scheduling and avoiding stockouts. This paper differs from earlier studies of inventory behavior in three respects. First, I focus on homogeneous and highly fungible commodities. This helps avoid aggregation problems, simplifies the meaning of marginal convenience yield, and allows the use of direct measures of units produced, rather than inferences from dollar sales. Second, I estimate Euler equations, and allow marginal convenience yield to be a convex function of inventories. This is more realistic, and better explains the value of storage and its role in the dynamics of price. Third, I use futures prices to directly measure marginal convenience yield. This produces tighter estimates of the parameters of the convenience yield function.Supported by M.I.T.'s Center for Energy Policy Research. Supported by the National Science Foundation