A hard nut to crack

This year has proved profitable for US oil refiners, but it could have been even better, had they not posted losses from forward product sales. Are refining companies learning from their trading mistakes? Joe Marsh reports

Oil refiners in the US have something to learn about using derivatives, if losses from their forward sales this year are anything to go by. But recent energy price volatility has certainly made it trickier to pick the right hedge. Connecticut-based Premcor, Houston-based Marathon Oil Corp and Valero Energy Corp in San Antonio, Texas are three of the refining companies that saw substantial trading losses after forward-selling crack spreads in 2004.

Crack spreads, or ‘cracks’, are calculated based on the difference between the price of a barrel of crude oil and the price of the refined products – typically gasoline and heating oil (or distillate) – made from it. Cracks are perhaps the clearest measure of how much revenue oil refiners are likely to make, and probably the contracts used most by such companies, says Bryan Caviness, senior energy analyst at rating agency Fitch in Chicago.

Refiners are beset by energy price exposure on three sides: from the price of crude oil that they buy to make their products; from the price of the newly refined products they sell; and from the natural gas they need to power their facilities. Given the high and volatile price of natural gas and crude in recent months, they might have been struggling badly to achieve returns – as they have done for most of the past three decades.

But high refined product prices – thanks to strong demand from a buoyant economy – have brought record refinery profits. In the nine months to September 30, Premcor posted net profits of $324.5 million, approaching treble the $127 million it made in the same period in 2003, while Valero reported a $1.315 billion profit to September 30, up from $490 million the year before.

Yet they could have made even more. In the third and fourth quarters of 2004, Marathon Oil sold crack spreads forward using derivative contracts at values higher than it thought sustainable in the months they were due to expire. “Forward prices ended up being much better than we expected,” says Kenneth Matheny, vice-president of investor relations and public affairs at the firm. The company lost $81 million specifically on the forward crack sales – the only trading activity Marathon viewed as speculative, says Matheny. This loss contributed to a total net derivative loss of $256 million in the refining, marketing and transportation (RM&T) segment in the first three quarters of 2004.

The company’s other open derivatives contracts in the RM&T segment are more conservative, however. Marathon’s only outstanding hedges – for the last quarter of 2004 and first quarter of 2005 – are mainly collars, a purely risk-management strategy to limit upside and downside gains, he says. A collar encompasses a call (an option to buy when the price drops to a certain level) and a put (an option to sell when the price rises to a certain level). “With these hedges, we were more concerned with protecting a value than speculating on price,” says Matheny.

Marathon also hedges the price of crude oil, and has been doing so longer than it has traded crack spreads – something it only started doing last year. The last time the company hedged any of its crude oil supply was in September 2003, says Matheny, and these contracts expire before the end of 2004. It has no such hedges in place for 2005, he adds. Is this down to the fact that Marathon posted derivative losses of $338 million in the exploration and production (E&P) segment in the first nine months of 2004, compared with $120 million for the same period in 2003? No, says Matheny – the company had some capital expenditure to hedge last year; its balance sheet is stronger now, so it sees no need to put hedges on for 2005 at present.

Meanwhile, Valero reported “after-tax losses of $150.8 million on certain cashflow hedges, primarily related to forward sales of distillates” in the three quarters to September 30. And Premcor locked in a price for cracks in the first quarter of 2004, thinking margins would not get any better, says Fitch’s Caviness. But the company lost out when it came to the second quarter – a ‘blowout’ period when refiners posted particularly big profits due to very strong gasoline cracks (see graph). Premcor had not returned calls for comment as Energy Risk went to press.

Rethink
Some refiners need to work on their risk-management strategies, says Leigh Parkinson,principal consultant at Calgary-based consultancy RiskAdvisory. “Those refiners who locked-in margins with the exclusive aim of profiting from their hedgeswill need to rethink their hedging programmes,” he says.

This is not to say such companies are being reckless in their use of derivatives. Producers who use hedging in a disciplined fashion to manage undesirable cashflow volatility should stay the course with their programmes, says Parkinson. If there was a sound risk management objective behind the hedges, and the programme was applied consistently, refiners will be in a strong position to defend themselves to shareholders and analysts, he adds, despite the magnitude of opportunity losses associated with the risk-management activity.

But refiners should not allow existing hedge losses to cause them to abandon hedge programmes altogether, says Parkinson. That would leave them without any protection if the market were to collapse as it has in the past, he says.

On the other hand, hedging can limit a refiner’s flexibility to react to price moves, says Fitch’s Caviness. As a result, refining companies are likely to use derivatives less than upstream exploration & production companies, he adds

Still, refiners are hardly showing signs of abandoning hedging. While some integrated integrated energy companies do not hedge – ExxonMobil, for one, only does physical trades – others, such as Shell and BP, actively trade futures, says Mark Routt, senior consultant at Energy Security Analysis Inc (Esai) in Boston. And pure refiners have shown increasing sophistication in their hedging techniques – particularly in the past few years, says Routt, who was formerly a crude oil trader.

Trading the margin
One thing they do is trade the refining margin, says Routt. If a companyhas a refinery with a capacity of 100,000 barrels, of which it can really onlycommit to filling 85–90% of capacity, it might hedge the excess 10–15% capacity using a ‘synthetic hedge’. That is, a margin specifically tailored to reproduce the exact mix of products made, and crude oil used, at the refinery. The company may even hedge the price exposure for the specific location whereits refinery is located, adds Routt.

Not all refiners run New York Mercantile Exchange West Texas Intermediate (Nymex WTI) crude or produce the kind of gasoline traded on Nymex, he says. One way for a refiner to match a unique product mix is with a specific counterparty through the over-the-counter (OTC) market, using swaps or swaptions. However, many refiners trade gasoline-crude cracks on Nymex as a proxy for their product mix and WTI as a proxy for the exact grade of crude they use.

For example, Premcor’s Port Arthur refinery in Texas has historically produced gasoline and distillate in roughly equal parts, says the company in its November 10Q SEC filing – an unaudited financial report submitted quarterly to the Securities and Exchange Commission. Hence, Premcor feels the Gulf Coast 2/1/1 crack spread – a calculation assuming a ratio in the oil barrel of half gasoline and half distillate – reflects its product mix. Port Arthur’s normal crude oil throughput is 80% heavy sour crude oil and 20% medium sour crude oil. Accordingly, Premcor uses the WTI/Maya oil price differential as an adjustment to the benchmark crack spread (Maya is a heavy Mexican type of crude).

Another common hedge for refiners is to trade the locational differences between US Gulf Coast distillate and New York Harbor distillate, says Esai’s Routt – also know as basis risk. Companies have a similar problem to tackle with regard to their natural gas fuel supply. Natural gas transaction prices are often based on industry reference prices – such as Nymex Henry Hub – that may vary from local market prices in, say, the western US, says Marathon in its November 10Q. Marathon therefore uses OTC transactions to manage exposure to a portion of basis risk.

Looking ahead
In the coming months, refiners will be looking to build gasoline and heatingoil stocks, without killing the golden goose of high prices, says Thomas Bentz,senior energy analyst at French bank BNP Paribas in New York. For example, heating oil-crudecracks are very high at the moment – around $11 a barrel (see graph), given that heating oil inventories are relatively low.

In the past year, the main problem has not been obtaining crude, as some believe, but turning crude into products, says Bentz. US refiners simply do not have sufficient capacity to meet demand. And any increase in inventories is unlikely to be enough to avoid a big price spike in the coming cold months.

Esai’s Routt takes a similar view. When warmer weather arrives next year, demand for gasoline will reappear and sorely stretch refiners’ ability to produce it, he adds. Once again there will be an issue over the supply of products – but not crude – that is directly related to the lack of refining capacity in the US. While winter in the northern hemisphere is typically one of large energy demand, there’s likely to be another run-up in product prices until early next year, he says.

So the present ‘golden age of refining’, as some have dubbed it, seems set to continue. Most refiners continue to recognise the inherent volatility of the industry by taking advantage of strong refining margins to strengthen their balance sheets through debt reduction or refinancing at lower rates, says Fitch’s Caviness. Perhaps their hedging decisions will reflect their growing market experience in the months to come.