It has been well documented that commodities have had low correlations with other asset classes like stocks and bonds. Perhaps a more powerful statistic, though less well known, is that since 1970,  four times that the S&P 500 posted a negative return for the year, the S&P GSCI posted negative as well.  The years when both indices were down were times of aggregate demand destruction like in the 1980s recession, the tech bubble burst, and the financial crisis.  However, during most years, there are fundamental drivers that caused a difference in return patterns between commodities and financial assets.

 

Although the prices of commodity futures contracts are directly linked to the expected future spot (cash) price of the underlying commodity, the expectations are generally wrong.  Supply and demand are the economic drivers of commodity prices, and while there are certain fundamental indicators that are more reliable, many events are unpredictable. 

 

For example, if a cattleman and a meatpacker agree today that the price of meat will be 72 cents per pound six months from now, they might not be correct. This is because something unexpected can occur before delivery such as a mad cow scare, which might drive the price down, or an Atkins diet craze, which might cause a price increase [Greer, Robert J., 2006, PIMCO. The handbook of inflation hedging investments].

 

Since sometimes the estimated price will be too low and sometimes the estimated price will be too high, the prices may even out over long periods of time; however, from month to month, there will be a variance of expected prices to actual prices and hence the term “expectational variance”. This is a main source of commodity index returns which will contribute to a pattern of returns that look different than that of stocks or bonds. 

 

Expectational variance, or market surprises, may also be responsible for the different return patterns between sectors. Each commodity sector responds to a unique supply/demand model where key economic factors that change our expectation about the price of energy are different from the factors that change our expectation about agriculture or precious metals.  For example, a drought may affect agriculture but not metals.

 

Not only has this been reflected through the historically low correlations between the commodity sectors as shown in Table 1, but is supported by the same type of bold metric that defines the asset class.

 

Since 1984, the first common year where each of the five sectors index returns were available, there were only five times when all five sectors have annual returns in the same direction.  Moreover, in four of those years, the returns across all sectors were positive whereas in only one year, all five sectors were negative, as shown in Table 2.

 

While the long-term result of expectational variance could be a zero-sum game from the impacts of both negative and positive market surprises, for commodities, there tend to be more positive surprises than negative surprises. Oftentimes, there are supply shocks such as weather events, political pressures, or accidents that disrupt the quantity of the commodities brought to market, resulting in price spikes. However, it is much rarer to experience demand shocks, or even supply shocks, with the opposite effect of causing an oversupply like what is happening from the new technology in the natural gas market.

 

Given there are low correlations between different commodity sectors, an opportunity arises from investing in across the markets.  An investor can take advantage of this fact by building a commodity portfolio of all the sectors with the same method that Harry Markowitz demonstrated was the efficient way to construct a portfolio of uncorrelated assets. This method of portfolio construction can be designed to have weightings that force it to buy what goes down and sell what goes up. This rebalancing can generate a source of return to a commodity portfolio using sectors that are not highly correlated with each other.

 

As investors get comfortable with using commodities in their portfolios and are ready to move beyond the simple yet important diversification and inflation protection of a broad index, sector indices may be used to express a view. Investors can study the distinct economics or idiosyncratic risks that drive each sector to perform separately but should also consider macro risks that can create inadvertent correlations amongst seemingly uncorrelated positions. [Till, Hilary and Gunzberg, Jodie, 2006, Hedge fund and investment management] With proper risk management investors can create alpha opportunities that can be implemented using sector indices.