Skip to main content
Publish date:

Using the Oil Curve

Can it be trusted? Let's see in part 2 of our feature.

In my last column, I discussed the influence of the crude oil price curve on many issues and instruments, not just oil, and the need to be aware of the constantly dynamic nature of the curve.

The characteristic of "contango," where near-month contracts are less expensive than back-month contracts, has been a relative constant of late. As a result, people have asked whether this upward slope of prices represents a prediction of future prices. I want to tackle that question with two stories. As Cramer might say, with a slight variation: "My job is not just to educate, but to entertain." Hopefully, you'll find these stories quite entertaining.

An Offer They Couldn't Refuse

The curve in gasoline prices has always been a little unique. The price curves upward in the spring, peaks to a premium during the summer driving season, then starts to decline again. The months outside any peak period are known as "shoulder" months.

Now rewind the tape to 1993. After the Berlin Wall fell, the new Russian republic thought oil presented a good start to entering the world economy. Russian crude was readily available, but the Russian refining infrastructure lagged far behind. One government-sponsored group was approached by a German company, Metallgesellschaft, with an idea: Swap discounted Russian crude for refined gasoline.

Metallgesellschaft had long experience in these kinds of arrangements, known as forwards swaps, but had previously used them almost exclusively for currencies and base metals. This was their first foray into oil. It would also prove to be their last.

The Russians wanted a long-term arrangement and created some very attractive components to lure the Germans into a long contract. First, the crude was sold according to the Nymex curve, but with a significant discount. Metallgesellschaft needed merely to hedge the price outright on the crude market to assure a profit, which they figured would be simple.

The gasoline leg of the swap is what proved problematic. The gasoline market was nowhere near as liquid as crude, and the 3,000 contracts a month that needed to be hedged could not be handled in the far-back months at that time. The Russians knew this. It was one reason they were for a long-term contract.

The Russians offered a lot in exchange. They contracted to buy the gasoline at a constant premium to the curve as it was currently represented throughout the life of the swap, which at the time amounted to a 5-cent profit per gallon to the Germans in August and quite a bit more in January. This was too good to walk away from, and the Germans entered into a 12-month agreement.

The final step for the Germans to lock in their profits was supposed to be simple as well. As a synthetic short (supplier) of gasoline to the Russians, they needed to buy the front-month gasoline contract and then liquidate that position while creating an equivalent long position in the next-closest month as the front month expired, an equivalent buy-and-sell operation known as a "roll."

In practice, all you need do is "buy the spread" every month, selling spot as you buy the next month out. The 5-cent profit pad in the first month and close to 10 cents at the farthest should have provided an enormous buffer against a price squeeze. All they needed to do was be wrong about the price curve for less than a nickel in each of those 12 months and they would make a heady profit, which must have seemed close to a sure thing.

When It Rains, It Pours

In the first month, the Germans managed to make a tidy 4-cent profit on 3,000 lots (close to $5 million!), as well as locking in similar gains on the crude oil leg, and the best, supposedly, was yet to come. Unfortunately for them, a few traders in the gasoline pits noticed an unfamiliar broker coming in at the end of the month with only one order to fill -- over and over he'd look to buy a lot of spot spreads.

In the second month, a few traders were watching for this broker to appear. He did and found that he had quite a bit more trouble getting his "roll" accomplished, although he still netted the Germans a tidy profit.

TheStreet Recommends

By the third month, the game was up. Pit traders inflated prices in the days before expiration and "blew out" the front month spread in preparation for the broker's one-sided orders, "helping" to see that his orders were filled as expensively as possible. Hey, what can I say? Traders are in it for the money. I remember feeling particularly sorry for this broker. It's just impossible to play poker when everyone at the table knows what cards you're holding. I imagine the reports he had to give his German clients at the end of a day weren't very happy ones.

By the fourth month, the entire floor was in on this monthly gift. Almost every Nymex trader had a few gas spreads waiting for this poor guy to come in and try and roll his positions. Remember, too, that the Germans had little choice but to try to cover their gasoline risk, and everything was going wrong at once. The "upstairs desks" were also aware of the Germans' problems and weren't about to buy them out of their bad swaps.

Their brokers were having similar difficulties in the crude pits where they were executing "reverse rolls," as the traders there had spotted the pattern as well. Other fundamental factors also conspired to skew the curve against them. When it rains, it truly does pour. Needless to say, the German firm did not survive the life of this swap arrangement and was bankrupted by the sixth month.

Is the Curve Trustworthy?

Was the curve acting normally during all of this? No. The spreads across the curve would explode astronomically until this broker was seen and contract wildly after he finished his roll. Was there any fundamental reasoning to the prices we were seeing? No. Was there anything predictive about the skewing of the curve? Of course not. The only thing you could predict was that a few oil traders were about to make a lot of effortless money.

Metallgesellschaft made the mistake of trusting the curve. But there's nothing in the curve worthy of trust. It changes often and for specious and trivial reasons. While companies must use it to plan their future energy costs, there is little reason to believe the price will be anywhere near those levels when those months reach delivery. Its gyrations force complex hedging strategies and fuels an entire risk-management industry. But don't ever believe that it represents anything other than a collective "idea" of what something might be worth in the future.

The Metallgesellschaft story is entertaining, but dated. The markets are more sophisticated and liquid than before, and the curves we see are a lot more predictable now than they were 15 years ago.

Or are they?

Those Who Cannot Remember the Past ...

I leave you to ponder that with a more modern tale of the natural-gas curve. Natural gas has both summer and winter peaks, as gas usage ramps up on really cold weather and really hot weather.

One hedge fund "bet" on the reliability of the natural gas curve, selling shoulder months against summer premium months. When temperatures in the Northeast last year hit 100-plus degrees Fahrenheit for five summer days, the perception of what constituted a "shoulder" month rapidly changed. Suddenly, participants wanted long positions throughout the curve, even in the shoulder months, annihilating the hedge fund's positions.

Additionally, the covering of those positions further skewed the curve, but it hardly mattered to the bankrupted fund's investors. To be sure, the curve reverted to its "traditional" shape after the positions were liquidated, but investors' portfolios were destroyed and liquidity removed from the natural gas markets in general. All of this from just one big bet on the "reliability" and predictive nature of an energy curve.

Have you figured out the fund I'm referring to?

That's right, it's Amaranth.

Dicker has been a floor trader at the New York Mercantile Exchange with more than 20 years experience. He is a licensed commodities trade advisor. Dan designed and supervised the introduction of the new NYMEX PJM Electricity futures contract, launched in April 2003, which cleared over 600,000 contracts last year alone. Its launch has been the basis of NYMEX's resurgence in the clearing of Power market contracts over the last 3 years. Dan has appeared as an energy analyst since 2002 with all the major financial news networks. He has lent his expertise in hundreds of live radio and television broadcasts as an analyst of the oil markets on CNBC, Bloomberg US and UK and CNNfn, for example. Dan is the author of many energy articles published in NYMEX and other trade journals. Dan obtained a Bachelor of Arts from the State University of New York at Stony Brook in 1982. Dan is also on the Board of Trustees at the Temple Israel of Great Neck.