(John Kemp is a Reuters market analyst. The views expressed are his own)
By John Kemp
LONDON The most important fundamental in commodity markets is not supply, demand, inventories or spare capacity but the passage of time itself.
Successful traders learn time is the most important influence on their profits. Not just in the simple sense of picking peaks and troughs but on the cost of holding positions and time's impact on all the other variables affecting prices (investment in new capacity, demand destruction and stock building).
Time enters explicitly into formulas used to price options as one of the "Greeks" (theta). It features in all elements of futures pricing (the cost of financing, cost of storage and convenience yield).
But discussions about time's impact are almost entirely absent from commentary on commodity markets and few institutional investors seem to appreciate it fully.
In too many instances, investors are expressing a very long-term view on supply and demand trends (for example peak oil), often using short or medium-dated instruments, ignoring the roll costs of maintaining the position, and the extent to which both supply and demand become highly variable over periods of more than about 2 or 3 years ahead.
Failure to comprehend the full effect may explain why returns for many investors have proved consistently disappointing.
In commodities, at least, it answers Fred Schwed's famous question "Where are the customers' yachts?" They have disappeared because investors forgot to account properly for time's passage.
By now most investors in commodity futures and first-generation buy-and-hold indices are familiar with the contango problem. The cost of carry (finance and storage) embedded in futures prices has eaten up almost all the returns from rising spot commodity prices over the past six years.
Marketing literature accompanying first-generation indices was careful to note broad-based indices of spot prices are not "investable" and investors should not expect to receive a return based on spot price appreciation alone.
The second and third-generation indices now being marketed seek to minimise the problem by shifting investors into contracts with longer maturities, employing more dynamic roll procedures, or rotating exposure among commodities based on those that have the most favourable contango/backwardation characteristics.
These strategies can minimise the drag on returns, but cannot avoid it entirely, and their effectiveness is likely to diminish in future as take up increases and more investors chase the same source of returns.
Because of carrying costs, investment in commodity indices is often based on a fundamental error. In many cases, investors are trying to express a long-term view about the direction of commodity prices (based on emerging market industrialisation and scarcity of raw materials) by owning short-term instruments, which must be constantly rolled forward to maintain the exposure.
It is as if they were trying to express a view on long-term interest rates by owning short-term bills instead of a 30-year bond. It is not very efficient. Costs of rolling are high, and the accumulated returns on short-term instruments do not track returns on a long-term investment closely.
But there is a more basic reason why investors should carefully about time's impact on the value of their commodity investments.
Charts 1 and 2 show monthly averages for spot Brent (actually the front-month Brent futures contract) and three specific futures contracts (Dec 2012, Dec 2013 and Dec 2015) popular with investors (here).
Front-month Brent surged to an average of $135 per barrel in July 2008, plunging to $43 in December 2008, recovering to $123 by April 2011, and recently easing off to $110 this month.
If an investor could capture the spot price on a long-term basis, they would have suffered a rollercoaster ride. But by April 2011 they would have made up most of the previous losses, as prices closed to within just $12 of the last peak. The market seemed almost as tight as it had been three years earlier.
In reality, an investor with a position in front-month futures, consistently rolling them forward, would have done much worse, as steep contangos, especially in late 2008 and 2009, would have bitten deeply into actual returns. In fact, carrying costs would have turned a $135 investment in oil in July 2008 into just $60 now.
What about an investor who expressed a long-term view about growing consumption and continued supply problems by buying contracts for oil delivered in December 2012, 2013 or 2015?
The long-term investor would have avoided roll costs and much of the volatility in the spot market. From a peak of $135 in July 2008, futures for 2012-2015 never fell below $65-70 per barrel, and recovered to around $106-115 per barrel by July 2011.
But performance has still been poor. December 2012 Brent futures have traded around $105 this month, $30 (22 percent) below where they were three years ago. For December 2015, performance has been even worse, with the futures contract still $40 (30 percent) off its previous peak.
Proponents of long-term investment via futures will argue this comparison is unfair; the three year period includes the biggest shock to the global economy since the 1940s; prices for long-dated commodity futures have actually held up in the circumstances.
But that misses two crucial points: (1) prices for long-dated futures contracts will always be affected by a mix of secular trends and cyclical position; and (2) while the advanced economies may have suffered a severe financial and economic shock, emerging markets are in robust health, and global oil consumption has surpassed its pre-crisis peak.
Investors in 3-year or 5-year commodity futures cannot ignore the cycle. Nor can investors ignore the impact of significant shifts in supply and demand trends on long-dated futures contracts. Production and consumption may be relatively fixed in the short-term, up to 2 years, but beyond that commodity markets exhibit more flexibility than analysts and investors give them credit for.
Global oil demand is almost 3 million barrels per day higher in July 2008. Yet spot oil prices have fallen more than $25 (18 percent) and the price of December 2013 futures is down $35 (25 percent). Consumption has been rationed effectively and supply has responded more vigorously than most analysts and investors predicted three years ago.
BRINGING TIME BACK IN
Time is the most powerful fundamental of all. Yet too often it is completely absent from analysis of the market.
There is a tendency to assume if conditions do not fulfil a particular forecast this year, the prediction or investment strategy can simply be pushed forward by 12 or 24 months. But that ignores both the carrying cost and the extent to which as predictions are deferred other factors also become increasingly variable.
Commodity futures do not have the required characteristics to serve as a separate "asset class". They are best thought of as instruments for risk management and speculation over the short-to-medium term (from a few months up to 2-3 years ahead). Not coincidentally, that is where most of the open interest lies in commodity futures markets.
As short-to-medium instruments, investors should be cautious about arguments which seek to justify values by appealing to long-term trends, while minimising short-term "fluctuations". Futures performance is dominated by the short-term.
Investors should be constantly aware of time's impact on both the cost of carrying their positions and potential changes to the fundamental assumptions that caused them to assume the position in the first place.
(Editing by Anthony Barker)