Is WTI Weakness Purely Physical?*

Prices for crude oil benchmarks WTI and Brent have historically been related. In general WTI light sweet crude oil sold at a 5% premium to Brent crude oil between 1994 and 2010. However, this relationship between WTI and Brent crude oil totally collapsed in 2011. Brent crude oil sold at an average premium of $13/bbl in 1H11, or 13%, reaching $23/bbl in mid-June.

That Brent might sell at a premium over WTI is not a new phenomenon. However, the magnitude as well as the duration of the current episode raises questions about the causes of this new ‘reality’ and whether the weakening of WTI and strengthening of Brent prices are temporary or permanent phenomena.

This report has long noted the combination of rising Canadian and onshore US crude supplies into Cushing, sluggish US demand recovery after the recession and bottlenecks facing the shipment of crude from Cushing to the Gulf Coast refining hub. This has led to a record increase in stock levels in Cushing, thereby depressing WTI crude prices, especially in the front end of the futures curve. At the same time, lower supplies from the North Sea, the loss of Libyan crude and rising demand from large emerging market economies, especially in China, have exerted upward pressure on the price of Brent crude oil. In addition,  given its ready access to seaborne markets, Brent crude  has gained more acceptance as a global oil benchmark.

Some argue that physical supply and demand cannot explain the magnitude and the duration of divergence between Brent and WTI crude oil prices due to a simple economic principle: arbitrage. The arbitrage principle suggests that if goods are close subsitutes and easily transported, then they should sell for a similar price. Since the transportation costs from Cushing to the Gulf of Mexico is not more than $10/bbl, and carry from the Gulf of Mexico to Europe costs an additional $3-4/bbl, a spread above $15/bbl should not be sustainable. The suggested conclusion is that, aside from physical demand and supply considerations, financial factors might have played some role in widening the spread between Brent and WTI crude oil prices.

There are basically two arguments as to how financial markets could play a role in widening the spread. In the first place, it is argued that, amid anticipation of rules on hard position limits by the CFTC, some speculators and commodity index dealers may have migrated from New York (Nymex) to London (ICE) in order to avoid such position limits. However, open interest data published by CFTC and ICE Europe do not support this hypothesis. Open interest in CME NYMEX WTI contracts has risen by 3.5% since January. Open interest in both ICE WTI and Brent contracts declined by 20% and 7% in the same period, respectively. Although data on speculators’ positions on Brent crude oil became available too recently (June 2011) to allow for comparisions, evidence from the ICE WTI contract suggests that the overall positions of speculators and swap dealers declined between January and July.

The second hypothesis is related to changes in the structure of the forward curve for Brent and WTI crudes. Up until December 2010, both WTI and Brent were in deep contango. However, starting last December, the Brent forward curve displayed a steep backwardation, while the WTI curve has stayed in contango. One of the components of the returns in total‐return commodity index investments is the roll yield, which is generated by the rolling of expiring nearby futures into the first-deferred contracts. Depending on whether the forward curve is in contango (when longer‐dated futures prices are higher than nearby contracts) or, in backwardation (when nearby prices are higher than longer‐dated futures prices), the roll yield is either negative (in contango) or positive (in backwardation). All other things equal, and assuming flexibility to switch positions across Brent and WTI, investment by commodity index traders in Brent contracts relative to WTI futures contracts should be expected to have increased since December due to a more positive roll yield in Brent. This shift should also be reinforced by the increase in the share of Brent and decline in the share of WTI crude oil in commodity indices. However, empirical evidence on the role of commodity index traders on prices is mixed; therefore further study is needed to assess the role of index funds on widening the spread.

At the same time,  many market observers disagree  with the notion that massive WTI-Brent spread necessarily imply a violation of the arbitrage principle. In the first place, those commentators argue that Brent and WTI, although close substitutes, cannot be directly interchangeable due to different gravity and sulphur content, not to mention restrictions on US crude exports. Second, there is not enough capacity to move oil from Cushing to the Gulf Coast. This last fact suggests that, until we see improvements in the transportation infrastructure from the Midwest to the Gulf, which might be available from 2013 onwards, WTI may continue to trade at well below its historical relative value. Pronounced WTI discounts may primarily be a physical phenomenon after all.   

*IEA Oil Market Report-July 2011