The record differential between the front-month and more liquid second-month contracts at expiry last week once again raised pointed questions about whether the NYMEX light sweet contract is serving as a good benchmark for the global oil market, or sending misleading signals about the state of supply and demand.

The expiring January 2009 contract ended down $2.35 on Friday at $33.87, while the more liquid February contract actually rose 69 cents to settle at $42.36 - an unprecedented contango from one month to the next of $8.49.

Criticism of the contract is not new, and past calls for reform have been successfully sidelined. But with policymakers taking a keener interest as a result of wild gyrations in oil prices this year, and a continued focus on regulatory changes to improve market functioning in future, there is at least a chance changes will be adopted as part of a wider package of futures market adjustments.

AN UNREPRESENTATIVE PRICE

During the surge to $147 per barrel earlier this year, OPEC repeatedly criticized the NYMEX reference price for overstating the real degree of tightness in the physical market and causing prices to overshoot on the upside.

While rallying NYMEX prices seemed to point to an acute physical shortage and need for more oil, Saudi Arabia could not find buyers for the 200,000 barrels per day (bpd) of extra oil promised to U.N. Secretary-General Ban Ki-moon or the 300,000 bpd promised to U.S. President George Bush in June.

Bizarrely, rather than acknowledge there was something wrong with the reference price, some market participants suggested Saudi Arabia should increase the already large discounts for its physical crude to achieve sales in a market that clearly did not need the oil, and was not paying enough contango to make storing it economic (contango is where the futures price is above the spot market).

The NYMEX WTI price may have achieved unprecedented media fame as a result of the “super-spike”, but a futures price to which producers and consumers were paying ever larger discounts for actual barrels was clearly not a good indication of where the market as a whole was trading.

Now the market risks overshooting in the other direction. Intense pressure on the front month in recent weeks has more to do with the contract’s peculiarities (in particular storage restrictions at the delivery point) than a further deterioration in oil demand or a market vote of no-confidence in the 2.2 million barrels per day further cut in oil production announced by OPEC at the end of last week.

The collapse in NYMEX prices nearby risks exaggerating the real degree of oversupply  and demand destruction, sending the wrong signal to producers and consumers about the wider availability of crude in the petroleum economy.

DOMESTIC PRICE, GLOBAL BENCHMARK

The NYMEX contract is for a very special type of crude oil (light sweet) delivered at a very special location (Cushing, Oklahoma) in the interior of the United States. It is not representative of the majority of crude oil traded internationally (most of which is heavier and sourer) and delivered by ocean-going tankers.

These specifications made sense when the contract was introduced as a benchmark for the U.S. domestic market.

U.S. refiners have a strong preference for light oils, for which they were prepared to pay a premium, because of their much higher yield to gasoline. The inland delivery location, centrally located and near the main Texas oilfields, rather than one on the coast, made sense for a contract that tried to capture the “typical” base price for crude oil paid by refiners across the continental United States.

But these specifications make much less sense now the NYMEX price is increasingly used a benchmark for the global petroleum economy, in which light sweet crudes are only a small fraction of total output. Just as NYMEX prices sent the wrong signals about physical oil availability on the way up, distorting the market and triggering more demand destruction than was really necessary, they now risk sending the wrong ones on the way down.

Earlier this year, the problem was a relative shortage of light sweet crude oils at Cushing, while all the extra barrels being offered to the market by Saudi Arabia were heavier, sourer crudes that could not be delivered against the contract. Moreover, extra Saudi crudes would have arrived by ship, and the pipeline and storage configurations around Cushing would have made it difficult to deliver them quickly against the contract.

Financial speculators were able to push NYMEX higher safe in the knowledge Saudi Arabia could not take the other side and overwhelm them by delivering physical barrels to bring prices down. The resulting spike exhibited all the characteristics of a technical squeeze:  tight contract specifications ensured there could be shortage of NYMEX light sweet inland oils even while the global market was oversupplied by heavier, sourer seaborne ones.

Now the opposite problem is occurring. Crude stocks at Cushing have doubled from 14.3 million barrels to 27.5 million since mid-October. Stocks around the delivery point are at a near-record levels and approaching the maximum capacity of local tank and pipeline facilities (https://customers.reuters.com/d/graphics/CUSHING.pdf).

As a result, the market has been forced into a huge contango as storage becomes increasingly expensive and difficult to obtain, ensuring the expiring futures trade at a substantial discount.

But Cushing inventories are not typical of the rest of the U.S. Midwest (https://customers.reuters.com/d/graphics/PADD2_EX_CUSHING.pdf) or along the U.S. Gulf Coast (https://customers.reuters.com/d/graphics/PADD3.pdf), where stock levels are high relative to demand but nowhere near as overfull as in Oklahoma.

Once again the problem is geography. Coastal refiners have responded to the downturn by cutting imports of seaborne crude, limiting the stock build. But the inland market is the destination for some Canadian crudes that have nowhere else to go, and the pipeline configuration means they cannot be trans-shipped to other locations readily.

Light sweet crude has been piling up in the region, with refiners choosing to deliver the unwanted excess to the market by delivering it into Cushing.

NEW GRADES, NEW DELIVERY POINTS

The easiest way to make NYMEX more representative would be to widen the number of crude grades that can be delivered, and open a new delivery point along the U.S. Gulf Coast. Both reforms would link the contract more tightly into the global petroleum economy.

NYMEX already permits some flexibility in delivery grades. Sellers can deliver UK Brent and Norwegian Oseberg at small fixed discounts to the settlement price, and Nigerian Bonny Light and Qua Iboe, as well as Colombia’s Cusiana at small premiums.

In principle, there is no reason the contract cannot be modified further to allow a wider range of foreign oils to be delivered at larger discounts to the settlement price.

More importantly, NYMEX could open a second delivery location along the Gulf Coast, increasing the amount of storage capacity available, and linking it more closely into the tanker market.

If prices spiked again, a coastal delivery location would make it much easier for Saudi Arabia to short the market and deliver its own barrels into the rally. By widening the physical basis, it would also make it easier to support the market by cutting international production and avert a glut trapped around the delivery location.

So far, the market has continued to resist change. But there are signs policymakers might enforce one.

Earlier in the year, Saudi Arabia strongly hinted western governments should look at reforming their own futures markets rather than call for production of even more barrels of oil that could not be sold at the prevailing (unrealistic) price.

Naturally, some of the reform impetus has ebbed along with prices and demand. But policymakers continue to show interest in structural reforms, as was evident at last week’s London Energy Meeting, and there is an increased willingness to challenge unfettered market dynamics.

It is still possible the incoming Obama administration might force contract changes as part of a wider package of reforms designed to improve the functioning of commodity markets, reduce volatility and send clearer, more consistent price signals to the industry and consumers.

For previous columns by John Kemp, click here.

Even by the standards of a deep-cyclical industry, the “golden age” of oil refining has proved remarkably brief, lasting no more than three years, before giving way to a new dark age.

Particularly in the United States, refiners have returned to the state of chronic unprofitability that plagued the industry before 2005.

U.S. refiners now have too much capacity and produce the wrong products (gasoline) in a fuel economy increasingly dominated by ethanol and diesel. Capacity cuts of as much as 0.5-1.0 million bpd (equivalent to 4-8 average refineries) and expensive investment to reconfigure the system to increase the diesel yield seem inevitable.

EVAPORATING PROFIT MARGINS

In May 2007, U.S. refiners paid an average of about $64 a barrel to acquire high quality West Texas Intermediate (WTI) crude (less for other grades) and sold gasoline for $97 per barrel - a margin of $33 per barrel or 52 percent.

By November 2008, U.S. refiners were paying $62 to acquire WTI but selling gasoline at a loss for just $52 - a negative margin of $10 or 16 percent.

Other outputs are still profitable (notably diesel and heating oil) and many refineries will have acquired lower-quality crudes for less than the WTI price. The overall gross margin was still (just) positive.

But the NYMEX benchmark 3-2-1 crude oil-gasoline-heating oil has shrunk from $30 per barrel to just $3. Once operating costs (including natural gas, electricity, water and catalysts) as well as capital expenditures (building, maintaining and upgrading refineries) are taken into account, the industry is making little or no profit.

DEMAND DESTRUCTION

Demand for gasoline and other refined products has been falling for more than a year, initially in response to high prices and now as a result of a weakening economy, leaving refiners with a huge overhang of unused capacity.

The total volume of refined products supplied to the domestic market averaged just 19.2 million barrels per day (bpd) in the four weeks ending Nov. 21, down 1.7 million bpd (8 percent) from 20.9 million bpd in the same period last year. The volume of motor gasoline supplied (9.0 million bpd) was down 300,000 bpd (3.3 percent) compared with last year (9.3 million bpd).

Refiners have responded with run cuts and record exports of both gasoline and distillates to avoid flooding the domestic market and collapsing prices further.

Operating rates have been below year-ago levels since the start of 2008 (https://customers.reuters.com/d/graphics/US_RFRT1208.gif).

Refineries processed 15.2 million bpd of crude and other inputs in the week ending Nov. 21 - using just 86.2 percent of their 17.6 million bpd maximum capacity, and leaving more than 2 million bpd of crude distillation capacity idle.

Refiners also sent increasing volumes of refined products abroad to avoid flooding the domestic market. Refiners and merchants ramped up gasoline exports from 38 million barrels in Jan-Sep 2007 to 50 million in Jan-Sep 2008 (+32 percent) and distillate exports from 52 million barrels to 146 million (a massive increase of +182 percent).

It has not been enough. By Nov. 21, reported gasoline inventories stood at 200 million barrels (22.3 days of supply) up from 197 million barrels (21.2 days cover) in 2007.

ETHANOL DISPLACEMENT

Refinery gasoline is increasingly squeezed out by ethanol. U.S. ethanol production has tripled from 260,000 bpd in Sep 2005 to 640,000 bpd in Sep 2008, with another 80,000 bpd of ethanol imported. As a result, ethanol is cutting almost 750,000 bpd of demand for fossil-fuel refinery-derived gasoline (https://customers.reuters.com/d/graphics/US_GSETH1208.gif).

In Sep 2005, some 8.9 million bpd of gasoline was supplied to the domestic market, of which 8.7 million bpd came from refineries and just 0.3 million bpd was sourced from ethanol distilleries.

Three years later, in Sep 2008, the volume of gasoline supplied had fallen 400,000 bpd to 8.5 million bpd. But while the volume of ethanol sourced from distilleries had risen by 0.5 million bpd to 0.7 million bpd, the volume of gasoline sourced from refineries was down by a massive 1 million bpd to 7.7 million bpd.

Roughly half the refinery demand lost over the last three years is due to increased ethanol (500,000 bpd), while the remainder is due to cyclical factors (400,000 bpd).

The displacement of refinery gasoline is an explicit objective of federal policy to reduce U.S. oil imports. It has been accelerated by the surge in crude oil prices during 2007-2008, encouraging widespread voluntary blending of cheaper ethanol into the domestic fuel supply.

But increased blending volumes threaten to strand many U.S. oil refineries as white elephants with no long-term future. Refinery utilisation rates have been trending down since the start of the decade, but the loss of demand has accelerated notably since widespread ethanol blending commenced in 2005 (https://customers.reuters.com/d/graphics/US_RFRTA1208.gif).

As a result, there is an increasingly wide gap between system capacity and actual throughput. More than 2.0 million bpd of crude distillation capacity is sitting idle. The last time the refining system had more than 1 million bpd of spare capacity was in the early 1990s, when refiners responded by mothballing facilities and closing plants, cutting capacity by more than 500,000 bpd between 1992 and 1994 (https://customers.reuters.com/d/graphics/US_RFRTB1208.gif).

Even with refinery shutdowns, the long-term outlook is bleak. The Energy Information Administration (EIA) projects gasoline consumption will increase from around 142 billion gallons in 2006 to 151 billion gallons in 2030 (based on an increasing population and rising car use, partly offset by improved fuel efficiency).

But the fossil-fuel content of that gasoline is scheduled to drop from 136 billion gallons to just 125 billion gallons as the ethanol content rises from 5.5 billion gallons to 25.8 billion gallons to comply with Renewable Fuel Standard (RFS) targets.

GASOLINE-DIESEL MIX

As if falling demand and the increasing challenge for ethanol were not enough, U.S. refiners face a deeper structural problem.

Most of the world relies on diesel rather than gasoline for transportation fuel and heating demand. According to the International Energy Agency (IEA) the world consumed just 0.75 gallons of gasoline for every gallon of diesel in 2005, and the refinery system was configured to produce the two fuels in roughly the same proportion (https://customers.reuters.com/d/graphics/FL_CNSP1208.gif).

The U.S. petroleum economy is highly unusual in that it is tilted towards consumption and production of gasoline. The United States consumes almost two gallons of gasoline (1.97) for every gallon of diesel; the European Union consumes only 0.40 gallons and China consumes 0.48 gallons.

Until recently, that led to a mutually beneficial trade, with the United States exporting surplus diesel, while Europe and China exported surplus gasoline (https://customers.reuters.com/d/graphics/REFINEPRDS1208.htm).

But U.S. refiners now face the problem that in the fastest-growing parts of the petroleum economy (China, Asia, the Middle East and Africa) the marginal demand is for diesel, while their marginal supply is gasoline, for which demand is stagnating.

The global economy now faces a structural surplus of gasoline and a structural shortfall of diesel. By implication, the world has too much capacity for producing gasoline (much of it concentrated in the United States) and not enough capacity for producing diesel (especially in Asia).

As a result, U.S. refiners face increased competition in their domestic market from imported gasoline, while they struggle to produce enough diesel to sell abroad. This mismatch explains why U.S. diesel exports have risen much faster in the past year than gasoline, even though it is the domestic gasoline market which is most oversupplied.

The United States now has too many refineries for its increasingly ethanol-based economy, and they produce the wrong product mix for a dieselised global economy.

U.S. refiners have begun to reduce gasoline production (https://customers.reuters.com/d/graphics/EIA_REFGS1208.gif) and prioritise distillates (https://customers.reuters.com/d/graphics/EIA_REF1208.gif). But yield changes have been marginal (1-2 percentage points), reflecting the technical limitations of the existing refinery units.

In the short to medium term (12-24 months), it seems virtually certain U.S. refiners will have to cut total capacity sharply, perhaps as much as 0.5-1.0 million bpd, 4-8 average refineries. In the longer term, they have no choice but to undertake substantial capital expenditures to shift the system towards more diesel.