Backwardation

Backwardation is a condition of a futures market in which prices for more distant delivery months sit below prices for nearer ones. The CFTC, the federal regulator of the futures markets, defines it as a "market situation in which futures prices are progressively lower in the distant delivery months", illustrates it with gold quoted at $360.00 an ounce for January and $355.00 for June, and notes that "backwardation is the opposite of contango".

Like contango, the term describes the whole curve of prices for successive delivery dates rather than any single contract, and the comparison is between those delivery months. The CFTC's related entry, inverted market, gives the same condition from the other end: "a futures market in which the nearer months are selling at prices higher than the more distant months; a market displaying 'inverse carrying charges,' characteristic of markets with supply shortages."

The phrase "inverse carrying charges" is the whole explanation compressed into two words. Holding a physical commodity costs money, and the CFTC's entry for carrying charges lists the components: insurance, storage and interest on the funds tied up. Those costs push later delivery months above nearer ones, which is the ordinary shape and the subject of the entry on contango. Backwardation is the market paying somebody to give up the goods now rather than charging them to hold the goods later, and it takes a real shortage to do that. Whoever needs the commodity this month is bidding above what the market will pay for delivery in six.

That is why the condition tends not to last. A shortage is a temporary state in most markets: high prices for immediate delivery draw supply out of storage and pull production forward, and once the shortage eases the curve can revert to charging for carry. The practical consequence is that a curve shape is a description of current conditions, not a durable characteristic of a commodity, and a strategy that depends on the curve staying inverted is exposed to the shape changing rather than only to the price moving.

Backwardation reverses the arithmetic of the roll, which is the part that reaches ordinary investors. Almost nobody wants delivery, so anyone holding continuous futures exposure has to sell the contract about to expire and buy a later one. In contango that swap costs money each time. In backwardation the later contract is cheaper than the one being sold, so the same exposure is re-established at a lower price, and if the market price then holds up as the new contract approaches its own delivery month, the difference is a gain. The CFTC's entry for convergence names the tendency that makes this work: "the tendency for prices of physicals and futures to approach one another, usually during the delivery month."

It is not a free lunch, and the reason is in the same sentence. The roll gain only appears if the spot price does not fall while the position is held. A commodity in short supply is often expensive by historical standards, and the same conditions that invert the curve can reverse quickly. A backwardated market rewards the holder for the roll and still exposes them to the entire price move of the underlying, which is much the larger number.

The condition is not confined to physical commodities either. Any market with a term structure of futures prices can display it. What does not transfer is the storage explanation: where there is nothing to store, an inverted curve reflects who wants the exposure and at what price rather than a warehouse running dry.

Read a futures quote screen for one commodity and you see a price for each delivery month. If each successive month is quoted below the one before it, the market is in backwardation. The gap between two months is what the market will pay to have the commodity sooner.

Start from the CFTC's own illustration: gold quoted at $360.00 an ounce for January and $355.00 for June is a backwardation of $5.00 an ounce for the five months against January.

A hypothetical of what that does to a position. Assume a fund holds futures exposure to 1,000 ounces of gold in the January contract and wants to keep that exposure past January, so it sells January at $360.00 and buys June at $355.00. It still has exposure to 1,000 ounces; it now has that exposure at an entry price $5.00 an ounce lower. If gold is still trading around $360.00 as June approaches delivery, and the futures price converges toward it, the position has gained $5.00 an ounce from the roll alone, or $5,000 on 1,000 ounces (1,000 multiplied by $5.00), with no change in the price of gold over the whole period.

Now run the same trade with the numbers reversed, which is what contango looks like: selling January at $355.00 and buying June at $360.00 leaves the same 1,000 ounces of exposure sitting $5.00 an ounce higher, and an unchanged gold price produces a $5,000 loss on the roll. Same commodity, same position size, opposite result, decided entirely by the shape of the curve.

Pros

• A long futures position gains from the roll rather than paying for it, so the accumulated drag that afflicts futures-based funds in contango works the other way.
• The condition is informative. An inverted curve is one of the clearest market signals that the physical commodity is genuinely tight right now.
• Producers and holders of inventory can sell into unusually strong nearby prices, which is part of how a shortage resolves itself.

Cons

• The roll gain is small next to the price risk. A backwardated market can still fall, and the entire move in the underlying sits on top of whatever the roll contributes.
• It is usually temporary, so a position sized around today's curve is exposed to the curve flattening or inverting back.
• The condition often coincides with disrupted supply, which is exactly when prices are volatile and hedging is expensive.
• Anyone short the futures, or on the other side of the roll, faces the mirror image of the benefit.