Middle East Economic Survey
VOL. LI
No 4
OIL PRICES
Asymmetry In Dubai Oil Spreads
By Shawkat Hammoudeh and Mark A Thompson
The following article was written for MEES. Shawkat Hammoudeh is professor of economics at Drexel University and Mark A Thompson is the Cree-Walker chair of business administration at Augusta State University. They can be reached at hammousm@drexel.edu and mthompson@aug.edu, respectively.
With virtually hundreds of different grades, crude oil is classified into three major groups based on specific gravity and sulfur content. The API gravity property classifies crude grades into three main groups: Light, Medium and Heavy. The sulfur content property grades oil into Sweet and Sour. In terms of pricing, each of these groups has a benchmark or marker that is traded on various commodity centers. In addition, the spread between any two benchmarks is often traded and also vital in the price discovery process of crude oil and its derivatives.
The globally recognized benchmarks for the light groups are the West Texas Intermediate (WTI) in North America, and Brent in Europe and Africa. Brent crude is typically graded with an API gravity averaging 38.3° and a sulfur content of 0.4%. Therefore, it is considered a light, sweet crude oil. WTI has an average API gravity of 39.6° and a sulfur content of 0.2%. Thus, the oil market recognizes WTI to be a lighter and sweeter crude than Brent, and thus (usually) prices it higher.
The medium crude group is benchmarked by Dubai-Oman crude which has an API Gravity 31.0° and sulfur content of 2.0%. Most of the crude oil exported from the Middle East to Asian markets, currently over 10mn b/d, has been directly or indirectly benchmarked to the Dubai oil price. In contrast, the heavy crude group is benchmarked by Mexican Maya, which has an API gravity of 22° and a sulfur content of 3.3%. It is a heavy, sour crude and sells at a significant discount to WTI and Brent. Roughly half of Mexico's oil exports are heavy Maya crude which is exported to the US and Europe.
Emphasis On Sour Crudes
As the world becomes more critically reliant on heavier and higher-sulfur streams, the emphasis is placed more on sour crudes. As such, the price discovery process in relation to the Dubai benchmark is even more important, and not just because of trades on its own contracts but also due to its possible role in pricing other oil crude grades. The Dubai benchmark representing the medium, sour crude is priced in balance to WTI and Brent. This benchmark crude (which is now supplemented by Oman crude) is traded on the Dubai Mercantile Exchange (DME), among other centers in Asia. The establishment of DME in June 2007 has provided more transparency for the trading and pricing of Middle East sour crude oil as DME sets records after records in volume, open-interest and physical delivery.
Since changes in the spreads could bring arbitrage opportunities, it is important to understand the differential between these oil markers in order to be able to maintain a balanced pricing relationship among the different grades of crude in their categories. Due to active arbitrage activity in the oil market, these benchmark prices form a long-run relationship leading to an equilibrium (spread) position. However, there are several reasons to expect that the nature of the adjustment process in these oil markets to be asymmetric. In other words, the speed of spread adjustment to equilibrium spread may be different depending on whether the shock that hits the spread is positive or negative resulting in the spread narrowing or widening. The presence of risk-averse (financial) market participants with heterogeneous expectations in North American, European, Middle Eastern, Asian and Latin American markets can lead to price asymmetry in these different grades. Noisy trading activities may also result in asymmetric adjustment process where a price is pushed up (or down) causing the spread to widen (or narrow) until informed traders react to the temporary deviation and push prices back to the equilibrium position. Other researchers have suggested that transaction costs may also explain the asymmetric adjustment back to the equilibrium position. Financial market frictions, as well as institutional and regulatory constraints, are also major factors in oil market price mechanisms and may affect the spread convergence to equilibrium.
Dynamic Behavior
Thus, understanding the dynamic behavior of these benchmarks is important. In our paper with Bradley Ewing, we examine and model the spreads between different benchmarks.1 We briefly describe some of our results related to the Dubai price spreads.
The spread between the light, sweet WTI and the medium, sour Dubai price benchmarks, WTI price responds to both widening and narrowing of the spread in the long run and to its own past changes in the short run, again reflecting WTI’s prominent leadership role. The WTI price speed of adjustment to the long-run (mean) spread is faster when the spread is widening after a negative shocks “breaks” WTI as the leading benchmark. Dubai only responds to past changes in WTI and Dubai. This means that the long-run adjustments reflect more changes in the fundamentals of WTI than in the fundamentals of Dubai. The Dubai benchmark adjusts to short-run changes in WTI and its own past changes, perhaps because it does not have the leadership prominence that WTI enjoys.
The Brent-Dubai spread may be more popular in the commodity centers than the WTI-Dubai spread. Brent responds to both positive and negative shocks in the long run, as is with WTI in the WTI-Dubai spread, and also to past price changes in both Brent and Dubai in the short run. However, in contrast to WTI which adjusts faster from widenings in the spread in the previous case, Brent adjusts much faster to narrowings in the spread, perhaps because traders are more sensitive when Brent becomes more expensive relative to Dubai than when WTI becomes more expensive. In addition, Dubai responds only to past changes in Brent and Dubai in the short run, which is similar to the WTI and Dubai spread in the previous case, again reflecting lack of world-wide leadership.
These results indicate that when the WTI-Dubai and Brent-Dubai price spread gets out of alignment from the long-run position, there may be arbitrage opportunities enhanced by understanding the asymmetric adjustment process for both spreads. While WTI and Brent are actively traded and adjust to the long-run equilibrium spread, Dubai is less responsive to spread changes over time because it was not actively traded. These results reflect less leadership stature on part of Dubai and more on WTI and Brent. The establishment of DME in Dubai and the supplement Dubai crude with Oman crude may result in the Dubai benchmark being one to watch for in the near future. Therefore, Dubai is more likely to undergo adjustment changes in the future as their institutional and fundamental factors change. For Dubai, the change may come soon after the successful trading at DME.
1. See S Hammoudeh, B Ewing, and M Thompson, “Threshold Cointegration Analysis of Crude Oil Benchmarks,” Working Paper. A copy of the paper is available upon request.