In the BusinessWeek article, “The Giant Sucking Sound: Why is Oil leaving Cushing’s Tanks So Fast?” , Matthew Philips takes a look at why a “historic glut of crude that built up in Cushing over the past three years” has dropped 40% (to 34 million barrels from 50 million barrels) at the close of August 2013. Let’s take a look at why the increase in crude oil leaving Cushing, Oklahoma, (where WTI, the North American oil benchmark, is priced) is important. The graphs I created below will illustrate Philips’ contentions but will also show the impact that the changes in the WTI market have had on the gap between Brent and WTI.
Philips points to a number of key reasons why this rapid decrease of stored oil is happening, including increased refinery capacity (particularly at BP’s refinery in Whiting, Ind.) and an augmented pipeline capacity of 300,000 bpd through the Seaway pipeline.
The storage levels can be seen in Figure 1 which shows crude stock levels at Cushing reported by EIA dropping from 49,652 TBbl on June 25 to 42,119 TBbl on July 23.
Philips also points out that change in the futures markets for WTI, which has had previously been in contango, has recently moved into backwardation. This means that the price for the nearest WTI contract had previously been cheaper than futures months incentivizing storage of the crude for sale at a later date. This change to backwardation can be see in the Figure 2 that shows the difference between the NYMEX prompt contract for WTI vs. the prompt +1 contract.
As Figure 2 below shows, this change in contract price correlates very strongly with the decrease in stock level at Cushing.
This draining of oil from Cushing is important because over the last couple of years WTI has sold at a discount to Brent crude. This decline in stock and transition to backwardation signal a significant change in the market for WTI that could impact the long term pricing of crude and refined crude products.
All of these graphs were easily produced in the ZEMA Suite and help to illustrate Phillips points. For details on how these graphs were built please reach out to us for a demonstration.