London, 10 August 2011: Reuters

Commodity futures have again failed to offer a safe haven from turmoil in global equity markets and other asset classes. The past week's sell-off across commodity markets likely marks the end of the road for "first generation" indices such as the Standard and Poor's Goldman Sachs Commodity Index which have been marketed to institutions as offering a passive long-term risk premium similar to equities as well as portfolio diversification.

Broad-based commodity indices have failed to outperform equities in 2011 or in the last five years and returns are tightly correlated with equities, undercutting the argument for using commodity indices as a source of diversification. The arguments for treating commodities as a separate asset class, set out in 2004 by Gary Gorton and Geert Rouwenhost in "Facts and Fantasies about Commodity Futures" have proved illusory when put into practice.

In the language of economists schooled in the Lucas critique, the sources of previous returns are not "deep structural" and evaporated when too many investors tried to chase them. Investors will continue to seek substantial exposure to commodity derivatives as part of their portfolios but increasingly in the form of second-generation enhanced indices and third generation dynamic indices, as the first generation are gradually abandoned as perennially underperforming.

In second and third generation indices, returns are due to the skill of the index designers and managers rather than intrinsic to the derivatives themselves. Commodities are not really a separate asset class but a sub-species of the hedge fund and active asset management business.

In future, the greatest returns will come to investors employing the most skilled and knowledgeable experts in commodity and cross-asset fundamentals, able to tap the most detailed information on physical flows and superior forecasting and trading. The popularity of buy, hold and roll strategies will fade.

ASSET CLASS FAIL

The plunge in almost all commodity prices over the last week, with the exception of precious metals, has crystallised problems with the performance of the "asset class" that have been evident for some time. The cross-market plunge has wiped out all the gains recorded by the GSCI and its main sub-components this year.

The GSCI total return index closed down 4.9 percent on Monday compared with the end of 2010 while the GSCI non-energy total return index was down 4.6 percent. In comparison, the Standard and Poor's 500 was off 10 percent. But once timing differences for market closing are taken into account declines are broadly similar. First generation indices have outperformed the equity market this year, but not by very much, and not enough to justify treatment as a separate asset class.

Chart 1 shows the performance of equities as well as broadbaskets of all commodities and non-energy futures since the start of 2011. While there is a certain amount of noise, and daily correlations between commodities and equities are only 40 percent, the three indices clearly show a high degree of synchronisation over slightly longer time horizons.

http://graphics.thomsonreuters.com/ce/COMMODS-EQUITIES.pdf

Chart 2 shows the same three indices over a longerperiod dating back to 2006. Over the medium term, GSCI and component returns appear even more strongly correlated with U.S. equities. If anything the GSCI has underperformed the S&P 500, mostly because of the contango problem.

Few institutional investors will be willing to invest in products that offer equity-like performance but underperform the main equity indices in the medium term. One interesting aspect has been the superior performance of the non-energy futures (industrial metals, grains, softs, livestock and precious metals) and their slightly less marked correlation with equities (on both a daily basis and over long horizons).

While energy appears highly correlated with equities and the economy more generally, some other parts of the complex, especially agricultural futures, appear to offer far more diversification. Investors and fund managers seeking more unique, less correlated returns are likely to up their allocations to agricultural commodities significantly in future.

There may be heightened interest in non-energy indices and discretionary funds to try to get better diversification and strip out some of the macro correlations with equities. Most banks and hedge funds have limited expertise in this area at present, compared with oil, so they will need to expand their teams and competition for skilled professionals will be intense.

But recent volatility has undermined the old case for passive commodity indices and treating commodities as a separate asset class and likely marks the demise of the old approach to marketing these products. In future, commodity investments will be marketed on the basis of the skill and expertise of fund managers.

Ends --

By John Kemp, Reuters market analyst – for Commodities Now.

The views expressed here are his own.