The recent outperformance of commodities versus equities has caused a positive re-evaluation of commodities by both retail and institutional investors. Since December 2001, the annualised performance of commodities as represented by the Goldman Sachs Commodity Index (GSCI) has been +26.2% while the annualised performance of equities as represented by the S&P 500 equity index has been +3.8%. See Figure 1.
Figure 1
Data Source: The Bloomberg.
The positive performance of the GSCI has
largely been due to the following two factors:
(1) adverse supply shocks resulting from
the aging energy infrastructure in the US
and Europe, and (2) expanding demand, particularly
from China. These factors may well continue
to boost the returns from investing in a
commodity futures programme.
As discussed in Till and Gunzberg (2005),
if an investor elects to invest in a commodity
index product, then that investor realises
that he or she will earn the inherent return
of the asset class, will be able to do so
cheaply, but will not be provided with any
downside risk protection. It will be the
responsibility of the investor to either
time their investments in commodity indices,
or create a properly balanced overall portfolio,
so as to avoid downside risk.
Instead, if the investor chooses an actively managed commodity programme, then that investor expects the potential downside of this investment to be carefully managed.
Risk Premia Strategies
Some active programmes obtain their returns through the careful exploitation of time-varying risk premia in the commodity markets. For example, Figure 2 illustrates an active portfolio's exposure to various commodity sectors and fixed income during the first eight months of 2004.
Figure 2
Exposures to Time-Varying Risk Premia in
the Commodity Markets
|
This graph illustrates Premia Capital's rolling exposures in energies, metals, US fixed income, livestock, and agriculture during the first eight months of 2004. More technically, the graph shows the conventional benchmarks that were most effective in jointly explaining Premia's daily return variance using an advanced returns-based-analysis technique. The benchmarks are
the Goldman Sachs (GS) Commodity sector
excess return (ER) indices and a Bloomberg
US fixed-income index. The graph's
y-axis is the fraction of R-squared
that can be attributed to a benchmark
exposure. This is also known as the
benchmark's variance component. The
middle chart shows each benchmark's
contribution to R-squared over the
whole history. Source: PRISM ANALYTICS,
http://www.prismanalytics.com. |
We have found that the most important element in an investment programme that exploits the risk premia available in the commodity futures markets is its risk management methodology. In the interests of brevity, we will focus on only one aspect of risk management: the avoidance of inadvertent concentration risk.
Avoidance of Inadvertent Concentration Risk
Ideally a commodity portfolio manager will attempt to create a portfolio of diversified commodity strategies in order to dampen risk. According to hedge fund manager Paul Touradji, "One of the best things about being a commodity manager is the natural internal diversification." "While even unrelated equities have a beta to the overall market, many commodities, such as sugar and aluminum, traditionally have no correlation at all," according to Teague (2004) in his interview with Touradji.
Seasonally Varying Correlations Due to the Weather
One might expect that the price of natural
gas should not be correlated to the
prices of either corn or hogs. But in reviewing
Figures 3, 4, and 5 from the summers of
1999, 2001, and 2005, one might question
that expectation.
Figure 3
Panel A
Panel B
Data Source: The Bloomberg.
Figure 4
Data Source: The Bloomberg.
Figure 5
Data Source: The Bloomberg.
It turns out that the prices of natural gas, corn, and hogs all depend on the outcome of weather in the US Midwest during the summer. An exceptional heat-wave in the Midwest can impair corn pollination prospects as well as stress the delivery of adequate natural gas supplies for peak air-conditioning demand. Also, the delivery of hogs to market can be curtailed during heat-waves because the animals may get too stressed during transit. Because each of these commodity markets have common reactions to the possibility of extreme heat, their prices tend to wax and wane at similar times during the summer.
If a commodity portfolio manager does not want to own too much risk to the outcome of summer Midwest weather, then it would be prudent for the manager to examine how the portfolio would do during times of both extreme and normal weather during the US summer, especially if the portfolio includes natural gas, corn and hogs.
Seasonally Varying Correlations Due to Chinese Holidays
One might expect that the price of crude
oil should not be correlated to the prices
of either soybeans or copper. But in reviewing
Figures 6 and 7 from this past spring, one
might question that expectation.
Figure 6
Panel A
Panel B
Data Source: The Bloomberg.
Source: Till (2005).
Figure 7
Panel A
Panel B
Data Source: The Bloomberg.
What might explain the common waxing and
waning of prices in crude oil, soybeans
and copper during this past spring? As Figure
8 summarises, China is now the number 1
or 2 consumer in a number of major commodities.
When one re-examines Figures 6 and 7 in
light of the Chinese holiday calendar, one
notes that the lulls in each commodity's
bull market have occurred around the time
of Chinese holidays in February and May
of this year, presumably when Chinese demand
was absent.
Figure 8
Impact of Chinese Demand: The Macro View
- 1.3 billion people
- World's No.1 consumer
of copper
- World's No.1 consumer
of steel
- World's No.1 consumer
of iron ore
- World's No.1 consumer
of soybeans
- World's No.2 consumer
of energy
- 50% of the world's
cranes are in China.
Source: Howell (2005).
If a commodity portfolio manager does not want to own too much risk to fluctuating Chinese demand, then it would be prudent for the manager to be careful in his or her risk capital allocation to the petroleum complex, industrial metals and soybeans.
Cautionary Note on Chinese Demand
In April 2004, following reports of a more stringent official policy towards industrial loans in China, both copper and platinum prices declined precipitously, as shown in Figure 9.
Figure 9
Platinum and Copper Prices During the Last
Half of April 2004
Data Source: The Bloomberg.
Source: Till and Eagleeye (2005).
Copper and platinum became the same trade during the latter half of April of 2004. If a manager had both of these trades in his or her portfolio, then that manager may have inadvertently doubled up on risk to China's intensive economic development.
The entrée of China as a dominant force in the commodity markets obviously has a number of implications for an active manager's risk-management and portfolio-construction methodologies. One consequence may be to view the world in terms of internationally-traded commodities versus purely domestically-traded commodities. A basket of internationally-traded commodities includes, for example, crude, copper, and soybeans while the diversifying basket of (US) domestically-traded commodities includes livestock and natural gas. The "domestic" basket would be expected to perform independently of the "international" basket, and presumably the performance of the "domestic" basket would not be impaired by a Chinese economic hard-landing. In that scenario, we might expect the "international" basket to potentially perform poorly as its components have thus far benefited from strong Chinese demand.
That said, taking into consideration the risk of a Chinese economic hard-landing is not meant to invalidate the case for commodity investing. We are only recognising that the history of developing economies is one of violent, unpredictable fluctuations around a long-term growth trend (such as occurred in the US during the 19th Century) so one might expect similar patterns during the ongoing development of China.
Conclusion
We conclude by noting that while the commodity markets provide a manager with ample opportunities for creating portfolios of diverse strategies, there are a number of challenges in doing so. In this article, we provide two examples of those challenges: (1) the correlations amongst commodities vary seasonally due to meaningful weather events, and (2) the entrée of China as a dominant force in the commodity markets has created new correlation footprints. The main implication of these observations is that risk management in the commodity markets is a very dynamic process.
A version of this article was previously published in the Fall 2005 issue of Commodities Now, http://www.commodities-now.com.
The authors wish to thank Galina Kalcheva for assistance in risk management research.
Chicago-based Premia Capital employs statistical techniques to detect opportunities in derivatives markets across asset classes. Premia Capital's focus, though, is on the (natural resources) commodity futures markets. The principals of Premia Capital also advise large investment firms on sophisticated risk-management techniques in periodic consulting engagements. http://www.premiacap.com.
References
Howell, Robert, Investment Seminar, Schroders Alternative Investments Group, Commodities, Gstaad, February 2005.
Teague, Solomon, "The Commodities 'Gladiator'," Risk Magazine, June 2004, p. 88.
Till, Hilary, "Risk Management in Commodity Futures Trading," Presentation at GAIM 2005 Conference, Lausanne, June 2005.
Till, Hilary and Joseph Eagleeye, "Commodities - Active Strategies for Enhanced Return," a chapter in the The Handbook of Inflation Hedging Investments (Edited by Robert Greer) McGraw Hill, Forthcoming 2005; and an article in The Journal of Wealth Management, Fall 2005, pp. 42-61.
Till, Hilary and Jodie Gunzberg, "Absolute Returns in Commodity (Natural Resource) Futures Investments," a chapter in Hedge Fund Investment Management (Edited by Izzy Nelken) Elsevier, Forthcoming 2005.
