Futures Prices in Supply Analysis

Abstract

Empirical work on agricultural supply functions typically makes use of product prices from past time periods. This approach has important advantage that it avoids problems created by simultaneous determination of supply and demand, since past prices are predetermined. But approach has important drawback that theory does not reveal exactly which past prices to use. The simplest procedure, use of last year's price, assumes farmers to be unduly naive in formation of expectations. The more sophisticated lagged-price procedures following work of Nerlove create econometric problems (Brandow, Nerlove 1958c, Griliches). Some are also liable to charge of theoretical ad-hockery (Griliches, p. 42ff). As an alternative approach to estimating supply elasticity, this paper is an attempt to exploit theoretically well-grounded hypothesis that price of a futures contract for next year's crop reflects estimate of next year's cash price.1 Since appropriate price for supply analysis is price expected by producers at time when production decisions are being made, a futures price at this time is a good candidate for a directly observable measure of product price in supply analysis. In context of crop supply, there are several problems to be faced in use of futures prices. First, the market's estimate as given by a futures price reflects expectations of nonfarm speculators as well as crop producers, and it reflects directly expectations only of those crop producers who themselves make futures transactions. Second, there is issue of which futures contract is most appropriate. Third, at what date should futures price be observed? With respect to first issue, use of a futures price can be justified by hypothesis of rational expectations as developed by John Muth. Under rational expectations, there is no reason for farmers to have different price expectations from futures speculators, nor for farmers who make no futures transactions to have expectations different from those who do. If price expectations of those out of futures market differ from futures price, there is great incentive for them to enter. Thus, those out of market likely have price expectations similar to market price of futures.2 The second issue should cause no serious problem so long as futures contract pertains to new crop. Of course, even old-crop cash prices are influenced by expectations concerning new crop. But cash-futures basis changes from year to year and secularly as cost of storage (which includes interest) changes. The present analysis uses first futures price after crop is in. The third problem is most difficult because it is not clear exactly when production decision is made. There may not be any preharvest date at which a farmer can be said to have made irrevoca-