No abstract

No abstract

he question of a risk premium in the futures markets has been the object of T numerous studies throughout the years. Still, the empirical evidence is inconclusive. Tests which rely on various hedging models or on the consumption CAPM (capital asset pricing model) generally warrant the existence of a risk premium while studies based on the CAPM support the contention of no risk premium. Part of the problem lies in the absence of a legitimate model of futures prices. In this article we attempt to circumvent this problem by following an approach similar to the one developed by Mishkin (l981,1982a,b) in the context of the bond market. No attempt is made to test a particular theory or model. Instead, assuming rational market participants, estimates of the risk premium are generated from unrestricted regressions including pertinent variables selected from recent equilibrium and hedging models.Using this methodology, we find evidence of nonzero cyclical risk premiums in the Chicago corn, wheat, and oats markets for the period 1970111 to 1981IV. Furthermore, these risk premiums are shown to be different across markets. We also provide an interpretation of the behavior of the risk premiums during this period. The results generally support the view that risk premiums are allocated between long and short positions rather than on a speculator-hedger basis. We begin our discussion by reviewing the empirical and theoretical studies published in this area. We then briefly present the methodology used for the tests. The empirical analysis follows. The final section develops the implications of the results. Jacques Raynauld is

No abstract

No abstract