This paper undertakes an econometric investigation into the efficiency of commodity futures markets. Despite a considerable amount of empirical literature, there is no general consensus on whether or not the markets are efficient. The results of this study suggest that for certain commodities expected excess returns to futures speculation are non-zero, however, it is argued that these results do not necessarily imply that markets are inefficient, or that agents do not act rationally. The implications of the study for the cost of using the futures markets for hedging, and for the power of futures prices to forecast future spot prices, are also noted.