Asset Allocation: Where Commodities Fit in Your Portfolio
Book: Commodities: Markets, Performance, and Strategies
Editors: H. Kent Baker, Greg Filbeck, Jeffrey H. Harris
Publisher: Oxford University Press, 2018
ISBN: 9780190656010
The Rise (and Stumble) of Commodities as an Asset Class
Chapter 21, written by Claudio Boido of the University of Siena, asks a straightforward question: should commodities be in your portfolio, and if so, how much?
The story starts after the dot-com bubble popped in the early 2000s. Stock markets collapsed, and asset managers started looking for alternatives. Commodities seemed perfect. They had historically offered returns comparable to stocks, they were negatively correlated with stocks and bonds, and they were positively correlated with inflation. Billions of dollars poured into commodity markets.
The most popular approach was investing in a commodity index. The S&P Goldman Sachs Commodity Index (GSCI) and the Dow Jones-UBS Commodity Index were the go-to choices, both heavily weighted toward energy.
Then the 2007-2008 financial crisis happened, and the picture changed.
What Changed After the Crisis
Between 1959 and 2004, commodities looked like a dream asset class. The risk premium was 5.23 percent. Correlations with stocks were near zero (0.05). Correlations with bonds were negative (-0.14). Everything you want from a diversifier.
Between 2005 and 2014, those numbers shifted dramatically:
| Asset | 1959-2004 | 2005-2014 |
|---|---|---|
| Stocks | 0.05 | 0.52 |
| Bonds | -0.14 | 0.05 |
| Inflation | 0.02 | 0.25 |
The correlation with stocks jumped from 0.05 to 0.52. The negative correlation with bonds disappeared. The risk premium dropped to 3.67 percent. The diversification case for commodities got weaker.
What happened? The financial crisis itself increased correlations across all asset classes. China’s industrial expansion drove commodity demand and linked commodity prices to global growth. Monetary policy by the Federal Reserve and European Central Bank pushed interest rates toward zero, creating unusual market conditions. And the financialization of commodities meant that more financial participants were trading commodities for portfolio reasons rather than for hedging physical exposure.
Between December 2004 and June 2015, the S&P GSCI commodity index returned -4.6 percent annually. Compare that to the S&P 500 at +7.4 percent and the Barclays U.S. Aggregate Bond Index at +4.5 percent. Commodities went from being a strong performer to a drag.
The Performance Numbers
Looking at monthly data between 1970 and 2009, the GSCI had a monthly return of 0.979 percent with a standard deviation of 5.763 percent. The S&P 500 had a 0.669 percent monthly return with lower volatility (4.497 percent).
Energy was the most volatile sub-sector with a 1.180 percent monthly return and a massive 9.315 percent standard deviation. Precious metals had a relatively high standard deviation (6.692 percent) even though many investors think of gold as a safe haven.
One interesting detail: commodities show positive skewness. That means positive returns are more common than negative returns. This is the opposite of stocks, which tend to have negative skewness (big drops are more common than big jumps).
How Much Should You Allocate?
This depends on who you ask. Investor preference data from Norrish (2015) shows that the average allocation to commodities was just 0.24 percent. Facebook’s market capitalization alone was larger than the total market value of commodity investments at that time. So in practice, most investors barely touch commodities.
But research by Idzorek (2006) suggests much higher optimal allocations:
- Conservative portfolios: 9.1 percent
- Moderate portfolios: 21.9 percent
- Aggressive portfolios: 23.2 percent
That is a massive gap between what academic models suggest and what investors actually do.
Three Approaches to Commodity Allocation
1. Strategic Asset Allocation (SAA)
This is the long-term, set-it-and-forget-it approach. Investors decide on target weights and use passive products (commodity index funds or ETFs) to get exposure. Performance is benchmarked to an index like the GSCI or Bloomberg Commodity Index.
The problem: these benchmark indexes have design flaws. They weight commodities based on production and liquidity statistics, which leads to some issues. Natural gas accounts for more than 11 percent of the Bloomberg Commodity Index, for example, which might be too much. The S&P GSCI and Bloomberg Commodity Index have high volatility (24 and 19 percent respectively) and include 22 to 24 commodities. You could probably get similar volatility with fewer commodities if you chose them more carefully.
2. Tactical Asset Allocation (TAA)
Active managers adjust commodity exposure over time based on market conditions. They use technical analysis, momentum signals, carry indicators, and other factors to time their commodity trades.
Some research supports this for short-run performance, especially in high-volume and low-open-interest contracts. But contrarian commodity strategies tend to be unprofitable over longer horizons. TAA works in the short term but is hard to sustain.
3. Factor-Based Asset Allocation
This is the newer approach that gained traction after the financial crisis. Instead of just buying a broad commodity index, you build a portfolio around specific factors:
Value investing: Buy undervalued commodities and sell overvalued ones. In commodities, you can measure value by comparing the current price to the five-year lagged spot price.
Momentum: Buy commodities that have been going up and sell ones that have been going down. Research by Erb and Harvey (2006) and Miffre and Rallis (2007) found that momentum strategies are profitable for holding periods up to nine months.
Carry: Buy higher-yielding (more backwardated) commodities and sell lower-yielding ones. This strategy borrows from the carry trade concept in currencies.
Monetary policy: Adjust commodity weights based on central bank actions. During restrictive monetary policy phases, increase the weight in commodities. During expansions, reduce it.
Bhardwaj and Dunsby (2014) found that a diversified portfolio of commodity futures using these factor approaches produces higher risk-adjusted returns. Different commodity sectors behave differently during economic expansions and recessions. Industrial metals suffer during recessions but gain during expansions. Grains are insensitive to the economy and perform well during stock market crashes. Energy rises during high inflation.
Ways to Get Commodity Exposure
The chapter lists four main vehicles:
Physical commodities: Buy gold bars or silver coins. Simple in concept, but low liquidity, high storage and insurance costs. Only practical for precious metals and only for conservative investors worried about extreme scenarios.
Equity of commodity companies: Buy shares of mining or oil companies. This correlates with the stock market, which partially defeats the purpose of commodity diversification.
Derivative contracts: Futures, options, swaps, and commodity-linked notes. Efficient way to get exposure, but requires understanding of margin, rolling costs, and counterparty risk.
Mutual funds and ETFs: Easiest entry point for retail investors. Lower fees (especially ETFs). But futures-based ETFs face rolling costs that can create a tracking error. The ETF might not actually track the commodity price very well.
The Financialization Problem
One of the bigger themes in this chapter is the “financialization” of commodity markets. Before the early 2000s, commodity markets were dominated by producers, consumers, and speculators who understood the physical markets. Then financial institutions poured in, and the character of the market changed.
Some researchers worry that financialization has caused price movements that do not reflect actual supply and demand of physical commodities. Oil went from $65 per barrel in June 2007 to $145 in July 2008 before crashing to $31 in December 2008. Similar wild swings hit cotton, gold, copper, and coffee.
But other researchers push back. Irwin, Sanders, and Merrin (2009) argue that calling commodity index investing “speculation” is wrong. And Tang and Xiong (2012) found that the diversification benefits of commodities were stronger in 2000 and weakened as more index money flowed in.
My Take
This chapter paints a complicated picture. The case for commodities in a portfolio used to be clean and simple: low correlation, inflation hedge, equity-like returns. That case has gotten weaker since the financial crisis.
But it has not disappeared entirely. The factor-based approach is the most promising path forward. Instead of buying a broad index and hoping for the best, you can target specific factors (momentum, carry, value) that have shown persistent returns across time periods.
Warren Buffett’s advice in this chapter is worth noting: diversification helps those who do not know the financial markets, while it is less relevant for investors who know more. If you understand commodities well enough to pick the right factors and time your exposure, you can still benefit. If you are just buying the GSCI and hoping for the old diversification benefits, you might be disappointed.
The question the chapter leaves open is whether the pre-2005 commodity era was the norm and the post-2005 era is a temporary deviation, or whether financialization has permanently changed the game. Nobody knows for sure. But the data suggests that blind commodity indexing is probably not the answer anymore.
Previous post: Commodity Trading Mistakes and Catastrophic Blowups