Sao Paulo, 4 May 2010

For almost three years, returns from owning a diversified basket of commodity futures have lagged behind returns on common equity indices such as the S&P 500, as persistent contangos and missed dividends have cost investors dearly.

The impact on comparative returns has been substantial, as the attached charts show. Comparative Performance, Commodities versus Equities since 2001: (Download the charts below)

Via the contango, in which commodities for future delivery are more expensive than those for near-term delivery, long-only commodity investors are paying a significant cost to roll positions forward each month (negative running yield). In contrast, owners of common stocks are benefiting from a stream of dividends that has spiced up the performance of share prices (positive running yield).

Chart 1 shows the performance of the most widely-tracked commodity index, the Standard & Poor's Goldman Sachs Commodity Index, and the S&P 500 equity index, since 2001. The basket of commodity futures has clearly outperformed equities.

Despite losing 61% of its value between June 2008 and February 2009, the GSCI basket has subsequently regained half its previous loss, and is up by a massive 142% compared with January 2001. In contrast, the S&P 500 index is still down 12% from January 2001, in spite of recent rallies.

Total Returns

But measures of spot prices give no clue about actual returns to investors, since they take no account of profits or losses rolling the futures, or dividends paid on the stocks. Offering documents carefully point out that the spot GSCI is not a return actually available to investors.

Chart 2 shows the total returns to investors from a position in the GSCI and the S&P 500 index. While GSCI returns are reduced by the contango, especially after 2005, S&P returns get a small uplift from dividends. The GSCI has still outperformed equities, but the margin is far smaller. Total returns on the GSCI since January 2001 have been almost 18%; returns on equities have been 3%. Commodity returns still outperform, but by a smaller amount, which is significant given the much higher volatility the asset class has exhibited.

Much of the roll cost that bedevils the GSCI stems for its large weighting towards WTI crude oil, which has shifted into a large and persistent contango structure since 2005. Chart 3 strips out the impact of oil and other energy products by showing the spot-price performance and total returns on the non-energy contracts in the GSCI and. Non-energy components benefit from a smaller contango, but gain less from upside exposure to crude oil prices. The net result is a small improvement in total returns. By end-April 2010, the non-energy components of the GSCI had made a total return to investors of 22% since 2001, comfortably outperforming equities (3%) and slightly outperforming the main GSCI (18%).

Recent Performance

The problem is that most of the outperformance in commodity prices over equities stems from the earlier and middle parts of the decade. Chart 4 concentrates on the performance of the GSCI, GSCI non-energy and S&P 500 since the beginning of 2008. It embraces the big run up in commodity prices in H1 2008, the global recession which hit all asset classes in H2 2008 and H1 2009, as well as the subsequent recovery.

Equity and commodity prices have both rebounded strongly from early 2009. On a spot basis, the S&P 500, GSCI and GSCI non-energy have all recovered most of their previous losses and are now about 10-15% below the level at the beginning of 2008.

But total return performance is much more varied. While equity investors are down 9%, investors in the GSCI non-energy are down 28% and investors in the GSCI itself are still down 38%. It is total returns (as well as their own volatility and contribution to overall portfolio volatility) that matter to investors. On a total return basis, commodity futures have significantly underperformed in the last three years as a wide range of markets have traded in persistent and large contango.

Some of this underperformance can be blamed on the stunning collapse in oil prices, from which the market has only partially recovered. If oil moves back to its pre-crisis highs, as some analysts believe likely, the gap between equity and commodity futures returns will narrow.

But it is not all oil, as the underperformance of even the non-energy GSCI shows. The bigger difference lies in the impact of the contango (and to a lesser extent dividend payments). Unlike in equities, where investors receive dividends to compensate them for the use of their capital, investors are actually paying a hefty price to remain long in commodity futures.

Need For a Re-think?

For the long-only index strategy to start working again, the overall contango in many commodities markets (from oil to natural gas and grains) must narrow substantially, or overall prices will have to rise explosively. Both are possible. Many analysts and traders have predicted the contango in crude will disappear by the end of the third quarter this year, perhaps even replaced by a backwardation as physical demand improves.

These forecasts did not account for the sudden increase in crude inventories around the NYMEX delivery point at Cushing and the flaring out of spreads, especially for the second half of 2010.

If contango reflects cyclical weakness, it could still vanish, or at least narrow, as demand picks up. But if it is more structural, reflecting the increased impact of financial investors and the shift to carrying higher inventory levels, even a cyclical recovery may not do much more than narrow it slightly, leaving investors nursing a long-term problem.

The portfolio diversification justification for holding commodities has also weakened significantly in the last three years, as correlations between many commodity prices and equities have strengthened significantly. But this could change once commodity-specific fundamentals rather than cross-asset correlations reassert themselves. And of course there is still the potential for commodity prices to rise explosively in the next few years, either as the result of a breakout in inflation stemming from central bank liquidity injections, or a more narrow surge in commodity prices linked to continuing supply-side constraints.

Three Key Questions

Even so, long-only commodity investors face three tricky questions due to the persistent contango and its associated drag on commodity portfolios:

(1) Is the current period of pronounced contango across so many markets the result of (temporary) cyclical weakness, or does it stem from more fundamental, structural changes, such as the influx of investment money, that are more likely to be permanent?

(2) If the contango is here to stay, are the expected increases in commodity prices over the next two to five years large enough to justify the cost of remaining long in the meantime?

(3) Will commodity futures resume their role as inflation hedges and portfolio diversifiers as the recent period of high correlation breaks down, or has their movement into the mainstream permanently changed the way these instruments are traded, making them less useful in this regard?

MORE: Commodities Show Positive Hedge Pressure

Long-only commodity investors are becoming ever-more reliant on big increases in spot prices, especially crude oil, for their returns, as roll yields and "risk premium" have evaporated.

In their famous 2004 paper on "Facts and Fantasies about Commodity Futures", which helped spur the index investing boom, Gary Gorton and Geert Rouwenhorst estimated that owning a fully collateralised, equal-weighted basket of 25 futures would have generated an average annual return of around 10.7 percent between 1959 and 2004.

Returns were not based on a simple directional bet on prices (which actually fell in real terms over most of the period). Much came from interest payments on the Treasury securities held as collateral ("collateral yield") and the ability to capture the embedded risk premium in futures contracts (closely related to the "roll yield").

But as index investors and other longs seeking to hedge against inflation, diversify their portfolios or benefit from industrialization in the emerging world, flood the markets, market returns have shifted significantly to benefit the shorts.

Capturing Risk Premium

The concept of embedded risk premium is subtly different from market structure and roll yield. A risk premium is present whenever an investor can buy a futures contract for less than the market thinks the spot price will be when the contract matures (current futures prices . Charts 2 and 3 show the same de-composition for the energy and nonenergy components of the index respectively.

For the index as a whole, investor returns were generally positive between 2002 and 2007. Most of the gains camefrom rising spot commodity prices, with collateral and roll yields making only a very small contribution.

Changing Market Structure

From 2005 onwards, however, roll yields turned clearly and consistently negative. Investors have become dependent upon rising spot prices for their returns. In fact, investors need spot prices to rise significantly just to cover losses on rolling their futures positions forward each month.

In 2009, a huge spot price increase of 50% was significantly offset by negative rolls costing investors 37%. The pattern is being repeated at the start of 2010. While spot prices rose just over 5% in the first four months, roll costs consumed 3%.

Negative roll returns are a novelty in energy contracts (Chart 2). But investors have suffered losses rolling their contracts forward every year since 2004. The problem is not confined to oil. Negative roll returns are just as much a problem for non-energy futures contracts, especially grains, livestock and gold. Negative rolls in non-energy contracts have proved endemic, negative every year since 2001 (Chart 3).

Long-only index investors have become locked in a race against the contango, which gradually eats into their returns every month.

Some operators have sought to reduce this cost by promoting variants which roll positions less frequently, move them further into the future, where contangos tend to be less severe, or shift to a dynamic or discretionary roll procedure. None seems likely to do anything more than delay the contango impact. There are already signs that steep contangos are spreading further along the curve as investors move their positions further forward.

Negative Hedging Pressure

In fact, there are good theoretical reasons explaining why investors are no longer making money from being long commodity futures. The existence of a positive risk premium captured by investors running a long position was based on two assumptions, neither of which now holds:

(1) Futures markets were assumed to exhibit "negative hedging pressure". Producers seeking to put on short hedges against a fall in prices were assumed to outnumber consumers wanting to put on long hedges against a rise. The basic intuition: producers had more to lose from a sharp drop in the price of any one commodity, while consumers, who purchase a wide range of raw materials as well as labour, would see much less impact.

For speculators to cover the imbalance, they would need to be net long. That would only happen if they expected actual prices to turn out higher than those indicated by the forward curve. The risk premium captured by net long investors stems directly from the hedging imbalance.

(2) Implied by the first condition, speculators were assumed to form only a small fraction of the market, picking up and benefiting from the residual imbalance in hedging demand.

Both assumptions are now open to challenge. It is not clear that most markets are characterized by "negative hedging pressure" any longer. In fact, hedging pressure may actually have turned positive.

Commodity producers have shown less appetite for strategic price hedging in a high price environment. Much of the new interest is coming from the consumer side following fluctuations in prices over the past decade. To capture the risk premium implied by this hedging pressure, investors need to be short, not long.

Returns to Being Short

Moreover, in recent years, much of the growth in open interest on the futures exchanges has come from commodity indices, hedge funds and other "non-commercial" participants. Increasingly, investors are not taking the other side of a trade from producers or consumers, but from other investors.

Since many investors have entered the market on the long side, to diversify their portfolios, hedge against inflation or benefit from the rapid industrialization of emerging markets, they are creating positive hedging pressure themselves. If the buyers of commodity indices want to "hedge" against inflation or other risks in their portfolios, they must pay someone to take on that risk from them. The price they pay for futures contracts must be higher than the actual spot price the market expects when those contracts mature.

Because financial investment in commodities has outstripped hedging demand from both producers and consumers, index investors and other "longs" have helped move the market against themselves, creating a source of positive hedging pressure, and ensuring the embedded risk premium is now received by others who are running net short of futures contracts.

Who benefits? The opposite side of the trade is mostly being taken by the major commodity banks, physical commodity merchants, and hedge funds. In many cases, these institutions are running short (spread) positions to cover the cost of financing and storage, as well as hedge, large and rising volumes of inventories.

The only way for this to remain a positive sum game, with both the indices and those taking the short spread against them benefiting, is if commodity prices continue to rise sharply.

Ends --

By John Kemp, Reuters market analyst - for Commodities Now. The views expressed are his own.

Charts here:

Attachments:

GSCI-EQUITIES.xls

150 Kb