London, November 2010

"He that sells what isn't his must buy it back or go to prison," is an old stock-market axiom regularly cited by derivatives dealers and investors with long positions when shorts and hedgers accuse them of squeezing the market.

Dealers and investors with long positions often express an almost moral outrage that shorts have sold something they do not own -- and then seek to avoid having to pay whatever the longs demand by appealing to the exchange authorities or regulators to intervene.

The reality is far more complicated. At the heart of the futures markets is an ambiguity about the nature of the delivery obligation. This ambiguity about delivery is what makes futures markets work. It also explains why position limits are often needed and why futures prices can sometimes drive the physical market, rather than the other way around.

DELIVER -- OR PAY UP

In a narrow, legal sense the requirement to deliver or pay must be true. The foundation of every commodity futures contract is an obligation to make or take physical delivery of the underlying product or pay the difference between the price at inception and the price when the contract expires or is closed out.

The potential for every futures contract to result in physical delivery, and the resulting arbitrage with the cash market, is what ensures prices converge with the physical market on expiry. It is what makes futures an effective hedge for physical price exposures.

But in a broader sense, contractual rights associated with futures can never be absolute -- or the market would cease to exist. Every futures contract can in theory be taken to physical delivery, but hedgers rely on the fact that most will not in order to make the futures markets useful.

The closest parallel is with reserve banking. In theory, all depositors have the right to demand repayment at short notice. In practice, if everyone tried to exercise the right at the same time, a bank would be illiquid and possibly insolvent.

Reserve banking is possible only because depositors do not all exercise their rights simultaneously all the time. The same ambiguity is central to futures markets. If the longs were to demand delivery too often -- or a substantial premium to close out positions or roll them forward -- the market would lose its usefulness for hedgers.

CARGILL VERSUS HARDIN

In the landmark case of "Cargill versus Hardin", decided in 1971, the U.S. Court of Appeals for the Eighth Circuit observed: "While the obligation to make or take delivery is a bona fide feature of the futures contract, in reality the futures market is not an alternative spot market for the commodity itself, and indeed the functions performed by the futures market would be severely hampered if it were turned into an alternative spot market.

"Most parties who engage in futures transactions are in no position to either make or take delivery, and if they were required always to make preparations to fulfil their obligations to make or take delivery, the number of persons who could effectively participate in the futures market would be substantially restricted, thus reducing the liquidity and volume of the market.

"The main economic functions performed by the futures market are the stabilization of commodity prices, the provision of reliable pricing information, and the insurance against loss from price fluctuation. The functions can be fulfilled only if both longs and shorts can be assured that they can offset their contracts at non-manipulated prices".

HEDGERS ALWAYS VULNERABLE

By definition, hedgers will never (or rarely) want or even be able to make or take delivery for operational reasons. Take the example of a motor manufacturer hedging its expected future purchases of 1 million tonnes of aluminium by buying 1 million tonnes of futures contracts. The automaker will never actually want to take delivery on the futures; otherwise it would end up owning 2 million tonnes, double what it actually needs.

The same is true for a processor using short positions to hedge its inventory (raw materials, work in progress and unsold output). In theory, the processor can deliver stock against short positions if squeezed. In practice delivering against futures contracts would leave the firm without the inventories needed for operational requirements, and its regular business would be forced to shut down. Much of the hedged inventory may not even be in the right form, quality or location to satisfy delivery obligations.

Physical hedgers' unwillingness or inability to make or take delivery is what makes them so vulnerable to squeezes, particularly inventory hedgers with short positions. If shorts always had to make preparations to fulfil their obligations, many would abandon hedging as impractical.

Commodity derivatives markets would fail to achieve one of their central functions and might cease to exist altogether, or become casinos for side-betting on cash commodity prices. Futures markets are a delicate ecosystem of speculators, dealers and hedgers. Occasional squeezes are part of the environment and keep dealers and speculators interested. But if they occur too often or aggressively, the supply of shorts, many of them hedgers, will dry up and the ecosystem break down.

It is the vulnerability of the shorts, and need to protect them to make the market work, that justifies a system of position limits to prevent abusive squeezes.

FUTURES CAN DRIVE PHYSICAL

Many analysts insist that inflows of speculative money into the futures markets cannot affect commodity prices and that the physical market drives futures prices rather than the other way around.

But this is not how markets work. By now it should be clear that futures and physical positions are not perfectly interchangeable since not every contract will (or can) go to delivery. Futures and physical markets are to some extent separate; opportunities for arbitrage are more limited than theory predicts, at least in the short term.

Futures prices are determined by the supply and demand of futures positions. Cash market prices are determined by the supply and demand for the physical commodity. The two are not the same and the arbitrage is incomplete.

While the possibility of delivery should force some convergence between futures prices and the cash market at expiry, physical prices may adapt to futures prices more often than the other way around.

It is true that every long position must be matched by a short, but the supply of short positions at a given price is not infinite. If enough investment money flows into the futures market, prices must rise until sufficient short interest can be found or induced. So large-scale financial inflows can and do have an impact on futures prices.

The physical market may not always be able to anchor futures prices, at least in the short term. In theory, if futures prices rise too much, physical supplies will be delivered against short positions. In practice, there is never enough physical material readily available to deliver against all the shorts, so rising futures prices can only be offset in the futures market by buying back contracts rather than making physical delivery.

It takes time for the market to divert and accumulate sufficient physical supplies from normal business channels to meet a rise in futures prices driven by speculative rather than fundamental factors. Physical fundamentals will always prevail eventually.

But in the meantime (which can last from days to months or even years) futures prices are more likely to drive the physical market than the other way around. The futures tail really can wag the physical dog.

Ends --

John Kemp, Reuters market analyst - for Commodities Now.

The views expressed are his own.