A tution reflects all available information at any point in time such that only the arrival of new unanticipated information leads to a change in price. However, the notion of an efficient market is often confused with the hypothesis that forward prices represent unbiased estimates of future spot prices. Bias found in forward prices is not inconsistent with certain types of market efficiency. Dusak (1973) has applied the capital asset pricing model to holding futures market contracts and has shown that investors can earn a "risk premium" proportional to the contracts' contribution to entire portfolio risk. That is, the presence of real systematic risk associated with futures contracts earns the investor a risk premium paid through a bias in futures prices, regardless of the information used in deciding to invest.
Authors such as Leuthold (19741 Just and Rausser (1981), and Martin and Garcia (1981) have specifically investigated the question of bias in live cattle futures prices. Leuthold (1974) finds that futures price quotes four to eight months prior to maturity are biased predictors of spot cattle prices. He argues that by observing spot cattle prices one could obtain a more accurate expectation of eventual spot prices four to eight months forward. Just and Rausser (1981) also conclude that live cattle futures prices are better forecasts for a short-term rather than long-term horizon; they even outperform some commercially available forecasting services. Martin and Garcia (1981) extended Leuthold's analysis using more recent data to argue that live cattle futures prices have deteriorated as spot market predictors even as the market has matured. They also find some seasonality in the predictive accuracy of the market. AU three of these papers seem to agree that live cattle futures prices provide biased forecasts beyond four months prior to maturity. The major shortcoming of this conclusion is that if one knew the bias was constant or varied in some systematic way, this information could be used to create accurate price expectations using This article has benefited from discussions with Swarnjit Arora and Thomas B. Fomby. The author alone is responsible for any remaining errors.