What drives expected commodity returns?
It is tempting to think that stocks belonging to one industry, say pharmaceuticals, have higher expected returns than stocks belonging to a different industry, say automotive, because of differences in consumer demand, raw material prices, etc. Similarly, within an industry one stock may have higher expected returns than another, because the company has a competitive advantage or serves different consumer segments.
This paves the way for what is known as fundamental analysis. For commodity markets it is equally tempting to analyze expected commodity (futures) returns based on demand and supply factors for specific commodities, such as new industrial applications of certain commodities or temporary shortages.
More than two decades of research into stock markets has shown that expected stock returns are related to stock characteristics such as Size (of market equity), Book-to-Market ratios (also known as Value), Momentum, Sales-to-Price ratios, Earnings-to-Price ratios, Long term price reversals, etc. Sorting stocks on these types of characteristics turns out to give a useful indicator for differences in expected stock returns, unrelated to the industry to which the stock belongs. Moreover, the cross-section of expected stock returns appears to be conveniently summarized by only four stock characteristics or risk factors, namely Market risk, Size, Book-to-Market ratio, and Momentum.
In a recent paper, we show that a similar characterization of expected returns also holds in commodity futures markets. First, building on recent research, the cross-section of expected commodity futures returns does not so much depend on the sector to which the commodity belongs (i.e., agricultural, metals, energy, …) but much more on commodity characteristics such as the futures Basis (the difference between the futures and spot price), Momentum, Volatility, Inflation, Hedging pressure, and Liquidity. We show that these characteristics play a different role in the expected returns on short-maturity futures contracts (referred to as spot premia) than in the expected returns on spreading strategies (combining long and short positions in contracts on the same commodity but with different maturities, referred to as term premia). Depending on the characteristic on which we sort, commodity spot premia (based on short-maturity contracts) vary between 5% and 14% per annum. Term premia (based on spreading strategies) are smaller, and vary between 1% and 3% per annum.
As in the stock market, we find that also in the commodity markets, the cross-section of expected returns based on the various characteristics can be conveniently summarized by a few factors only. Spot premia can be characterized by one Basis-factor only, where the spread on the High-Basis versus the Low-Basis commodities is between 8% and 14% per year. Two additional Basis-factors are needed to characterize the variation in term premia, where the High-Basis and Low-Basis factors differ by a spread of 0.6% to 1.8% per year.
For asset managers seeking commodity exposure in their investment portfolio, the Basis as a characteristic is therefore a useful starting point to structure their commodity portfolio.
An Anatomy of Commodity Futures Risk Premia, Szymanowska, de Roon, Nijman, van den Goorbergh (2013)