The authors analyze the risk premium in commodity futures markets by deconstructing them into two primary risk factors: the spot premium, which is related to the risk in the underlying commodity, and the term premium, which is related to the changes in basis. By using various trading strategies, they are able to isolate the different risk premiums and find that they typically have opposite signs and are highly predictive. Their findings suggest that spot and term premiums are affected in different ways from forecasting variables that were previously recognized.

Not requiring an initial investment, futures contracts are zero-cost securities, and as such, the expected returns consist of only a risk premium. Because of the size of the market and the wide range of market participants, it is important for investors to understand these risks because futures contracts are used quite extensively to mitigate risk. Similar to bonds, commodity futures have a term structure component; at the same time, they have some stock-like characteristics that vary depending on the sector in which they belong.

The authors’ work provides three insights. First, they decompose commodity futures expected returns into spot premiums and term premiums. This separation is important because it is likely the result of different risk factors. Second, the authors show that differences in expected returns on various trading strategies are the result of such characteristics as basis, volatility, and momentum, as well as other instruments having a greater impact than just which sector they are in. Third, the analysis reveals that a limited number of factors can describe the cross-sectional variation in commodity returns, similar to the well-known Fama–French factors. In the authors’ analysis of commodity futures, one basis factor can explain the spot premium and two additional basis factors can explain the term premium.

The authors’ work suggests that the predictive power of valuation ratios, such as the dividend yield for stocks, carry trade for foreign exchange, forward premium for bonds, and the ratio of house price to rent for real estate, also applies to commodities markets. Standard asset pricing tests reveal that the cross-sectional patterns in a spot premium can be primarily attributed to a single basis factor, a factor that is long the highest-basis commodity futures and short the lowest-basis commodity futures. The term premium is explained by two factors: the highest- and lowest-basis commodity futures portfolios.

The authors examine bimonthly data from the Commodity Research Bureau covering 21 commodity futures contracts beginning March 1986 to December 2010. They divide the data into seven categories: energy, metals, grains, meats, oilseeds, industrial materials, and softs (the softs category includes coffee, orange juice, and cocoa). These markets were chosen because of their broad exposure to various commodities and relatively large trading volumes. The authors use as many as four different maturity dates ranging from two months to eight months, and consistent with other studies, they use the closest month futures contract as the spot contract because spot contracts tend to be illiquid. The authors build 21 equally weighted indices using the contracts as well as 7 equally weighted “sector-maturity” indices, taking the simple average of log returns. Sorting on the futures’ basis, inflation, liquidity, momentum, and volatility results in a significant spot premium of 5% to 14% in the high-minus-low portfolios. For these portfolios, the term premium is between 1% and 3%.

The authors’ use of almost 25 years of data covering 21 commodity futures markets and their use of rigorous analysis provides valuable information. The examination of the breakdown of a commodity futures risk premium gives additional insight into the separate risk factors, which, in turn, gives investors additional information and possible ways to manage risks when using these instruments for hedging.