A new tool-resampled efficiency-helps create optimal allocations.

Commodities are often characterized as a defensive asset class. They bear a low to negative correlation to traditional asset classes like stocks and bonds, which are closely correlated to each other and tend to be more sensitive to the movements of financial indicators. In an unanticipated inflationary environment, commodities tend to outperform traditional assets because investors are inclined to move away from paper assets and turn toward tangibles such as commodities.

Commodities can be used as a diversifier to improve the risk and return profiles of different portfolios. Selecting the right commodities for an investor based on his or her risk tolerance and return preference, however, is not always an easy task. But a new tool-resampled efficiency-patented in 1999 can make the process easier. It helps create optimal portfolio allocations, not only with traditional assets like stocks and bonds but also with commodities. Although resampled efficiency is one of many optimization tools, Nobel laureate Harry Markowitz found that it produces more diversified and stable portfolio returns than his own mean-variance portfolio model, a pioneering development in Modern Portfolio Theory.

Investors can gain exposure to commodities by investing in commodity index or individual commodity futures contracts. A third option is mutual funds.

Futures contracts for the Goldman Sachs Commodity Index (GSCI) and Dow Jones AIG Commodity Index (DJ-AIGCI) trade on the Chicago Mercantile Exchange (CME) and the Chicago Board of Trade respectively. DJ-AIGCI also has TRAKRS, an exchange-traded fund, on CME. Although Standard and Poor's Commodity Index futures have not been listed on the New York Board of Trade (NYBOT) since July 2003, swap contracts are available based on the index.

The Commodity Research Bureau index (CRB) is an important benchmark index and can be invested in through exchange-traded futures. However, it's not ideal for investment purposes. Anthony Scamardella, managing director of marketing at NYBOT, says the commodities in the CRB are equally weighted, and that does not justify economic realities. In most other indexes, weightings vary from one commodity to another, aiming to reflect their economic or commercial importance.

Commodities can be broadly classified into three major sectors-agricultural, energy and metals-with associated futures contracts traded on different exchanges. Table 1 displays factors that have price implications on different commodity classes.

Each commodity or commodity class is governed by a different set of supply and demand fundamentals and responds differently to the same event and factor. An idiosyncratic event may have a substantial price implication on individual commodity markets, but the resulting impacts are diluted when adding all commodities together in an index.

Because of these varying responses, a uniform strategy to long or short a basket of different commodities may not be appropriate for achieving optimal returns. The strategy may be right for some commodities within the index but not for others. Investment savvy is called for, and trading strategies should be flexible and individually tailored for each commodity.

"Each of the three sectors, being agricultural, metals and energy, is governed by different production cycles and economic demand and supply factors, hence prices behave differently. The rhythm of economic timings for each commodity is different. Investments play an instrumental role in each economic cycle," says Jim Steel, director of research at Refco Group Ltd. LLC in New York.

The length of a production cycle varies from sector to sector. Metals tends to have the longest production cycle, due to the lengthy and capital-intensive mining process. Being sensitive to economic cycles, metals are prone to experience a contraction in a sluggish economy because of lower demands for construction. Energy has the shortest cycle. Demand for both energy and metals tend to be cyclical. Energy products like gasoline and heating oil are also sensitive to weather conditions.

In contrast, production of agricultural products tends to be seasonal and demand is noncyclical. A high degree of government intervention in the agricultural sector very often includes heavy subsidies.

How To Invest

Institutional investors still dominate commodity trading, particularly trading in individual commodities. However, retail investors also may invest in individual commodities through a broker or commodity trading advisor (CTA). The legal definition of a CTA is any person who, for compensation or profit, directly or indirectly advises others on buying or selling commodity futures or option contracts. A CTA must be a member of the National Futures Association (NFA) and must register with the Commodity Futures Trading Commission (CFTC), which regulates the U.S. futures industry. NFA screens firms and individuals wishing to conduct business with the investing public.

However, a CTA can offer services only to an eligible qualified person, defined as someone who,has a net worth of more than $1 million or , for two succeeding years, an income of more than $200,000 a year or a joint income of more than $300,000 a year.

Another way to invest is through two commodity mutual funds, PIMCO Commodity Real Return Strategy Fund and Oppenheimer Real Asset Fund. They are widely available to individual investors and are considered true commodity plays. PIMCO has swap positions based on DJ-AIGCI as an underlying index and Oppenheimer tracks the performance of GSCI. Both have positions in fixed-income securities, ensuring liquidity and creating a cushion without significantly affecting overall returns.Several exchange-traded funds based on commodity indexes are being planned.

Defining Commodity Allocations

Optimization tools help managers assess which assets or combination of assets will generate more superior returns. A recent technological development-resampled efficiency- has simplified creating optimal asset allocations and satisfying investment objectives. As mentioned earlier, resampled efficiency is considered an improvement over the classical mean-variance efficiency theory developed by Markowitz. Resampled efficiency, devised by the Michaud family at Boston-based New Frontier Advisors LLC, adopts numerous simulations to identify portfolios that are statistically similar to those on the MV efficient frontier and to define a more realistic "efficient region" that is considered a "true" efficient frontier.

As opposed to a point, the region offers a greater number of optimal portfolios, forming an efficient region at any risk level. Resampled efficiency also offers a rebalancing test that significantly reduces trading costs, and it leads to higher returns. In a December, 2003 issue of Pensions & Investments, Markowitz tested the resampled efficient optimization model and found that it produces more diversified and stable portfolio returns than those produced by MV optimization.

Table 2 illustrates the results of resampled efficient optimization of 17 individual commodities and three investable commodity indexes from February 28, 1994, to April 30, 2004. Six optimized portfolios are created, ranging from A to F. Different asset allocations are shown across the risk spectrum. The bottom two rows of the table show the mean returns and standard deviations of each optimized portfolio. The last two columns on the far right display the mean returns and standard deviations of each stand-alone asset. Commodity indexes are selected to represent a passive vehicle. Given futures contracts have expiry, returns on the individual contracts are calculated under the assumptions that underlying contracts are rolled on the last trading days. A buy-and-hold strategy is also assumed for both index futures and individual commodity futures for the entire period.

Analysis Of The Results

The resampled results provide a range of different asset mixes across the entire risk spectrum. As indicated in Portfolio A, commodity indexes (DJ-AIGCI and SPCI), dominate the allocations for 33% and 13.6% respectively at the lower end of the risk spectrum. That particular allocation gives a return of 8.7% and a standard deviation of 6.9%. The relatively lower risk level can be explained by the diversified nature of the indexes. As investors become willing to take on more risk for a higher return, they adopt commodity portfolios at the higher end of the risk spectrum.

As illustrated in Portfolio F, the total holding in commodity indexes drops to only 1%, represented by GSCI. Natural gas dominates the portfolio representing 50.5%. As a stand-alone investment, natural gas has the highest mean and standard deviation out of the 17 individual commodities. Despite the volatility, it is the individual commodities that drive the returns as their allocations increase within a given portfolio. The cross section and the gradation of changes in the optimized asset allocations can also be shown in a decomposition map. Another observation is that Portfolio C and the stand-alone SPCI investment share a similar level of returns whereas the standard deviation of Portfolio C is almost 30% lower due primarily to its diversified nature.

Conclusion

Although commodity indexes offer the benefits of diversification, it is individual commodities that drive the overall returns. Available commodity products can be used to complement different asset classes, or simply used together to take advantage of arbitrage opportunities. Given the unique characteristic of low to negative correlations to stocks and bonds, and the behavior of moving away from paper assets and turning towards hard assets during unanticipated inflationary periods, direct commodity investment is well placed to be a part of every investment portfolio.

Pauline Lam is a New York City-based consultant specializing in commodity investment research and alternative investments. She can be reached at 212 829-7189 or [email protected].