he pricing of futures contracts relative to their underlying cash assets via no-T arbitrage relations has been a subject of extensive theoretical and empirical research. Recent studies of arbitrage-enforced relative futures-cash pricing restrictions by for Treasury bill futures,' and by ; Cornell and French (1983); for stock index futures, report varying degrees of apparently significant deviations from fair cost-of-carry pricing. These results are of both practical and academic interest. First, the failure of such elementary no-arbitrage relations suggests that systematic market inefficiencies exist? implying simple riskless strategies through which portfolio managers may improve returns. Second, these deviations from no-arbitrage relative pricing suggest that excess risk is borne by hedgers who must close out their positions prior to contract expiration.
While many studies have examined the pricing restrictions implied by arbitrage activity, little attention has been given to the implications of arbitrage activity for the determination of trading volume in both cash and futures markets. In particular, Financial support from the L. Glucksman Institute for Research in Securities Markets is gratefully acknowledged. Glucksman Fellow Laurie Hodges. U Fi Wen and John Young provided thoughtful and capable research assistance. Interactive Data Corporation kindly supplied access to their data. Larry Hams, Joel Hasbrouck, William Silber and palticipants of a GBA Finance Workshcp provided useful criticisms of an earlier draft. The views expressed here are solely those of the author and do not necessarily represent the views of the Federal Reserve Bank of Philadelphia or of the Federal Reserve System. 'See also the earliest studies of Treasury bill futures arbitrage relations (Poole, 1978; Capozza and Comell, 'An alternative equilibrium interpretation of the evidence is that a somewhat deeper modeling effort may be and Lang and Rasche. 1979).
necessary to uncover the hidden costs and impediments to arbitrage.