A record WTI-Brent spread, a new paradigm | FT Alphaville

A record WTI-Brent spread, a new paradigm

Here it is, the all time record spread between CME WTI front-month future crude and ICE Brent front-month future crude:

According to our assessment that’s a record on a like-for-like basis on both Bloomberg and Reuters terms (Reuters data quoting a slightly higher historical record from January 2009).

Nevertheless, a record $12.10 premium of Brent to US crude has now been hit on Thursday.

A fundamental paradigm shift?

As we reported earlier, distortions of this calibre have become increasingly more common since North American production volumes began rising in the mid naughties — especially on account of additional flows from Canada.

Until now, though, the Brent premium effect tended to be a temporary glitch, mostly affecting the front end of the curve — an indication that the market believed WTI’s premium to Brent would eventually return at some point.

But, as FT Alphaville reported earlier this month, this was no longer the case in 2011. The Brent premium had escalated across the entire curve — which is odd given WTI’s superior quality.

It’s a point mulled by Olivier Jakob of Petromatrix on Thursday.  He theorises, for example, that the last time the spread hit these sorts of levels it was due to a very different reason. More down to ETF fund flows than real fundamentals.

As he explains:

If in 2009 the Brent premium to WTI was only visible in the front month it was because it was primarily the result of the weakness in the front spread of WTI where the contango widened to extreme levels. While high storage in Cushing was responsible for a WTI contango in early 2009, our opinion has always been that the supercontango in WTI then was due to the USO ETF rolling (and being front-run) 70 to 90’000 contracts in a single day. The Brent premium of WTI in early 2009 was not really a premium but purely the side-effect of the extreme contango on the front month of WTI when Feb-March WTI went down to a contango of -9 $/bbl and Mar-April down to a contango of -8 $/bbl.

In other words, back in 2009, the Brent premium was more of a reflection of the trading play going on in WTI due to its extreme contango in the first two months of the curve only.

Now the distortion is clear across the entire curve.

Note the graph from 2011, which does not see WTI return to a premium anywhere:

And the chart from 2009, which reveals a crossing point around two years out:

This could, of course, account for the discrepancy between Bloomberg and Reuters in judging the record spread. Should you take the record set by the permanent differential between the WTI front and Brent front, or the record set between the like-for-like contracts (that is January WTI versus January Brent) despite them expiring at different times?

Here’s more from Jakob (FT Alphaville’s emphasis):

The Brent premium to WTI in 2011 is very different as it is not due to extreme weakness in the WTI contango. WTI is in a contango and at -2.00 $/bbl it is a wide contango but nowhere near the -8 or -9 $/bbl of 2009, about in line with the contango seen in August or September 2010 and narrower than the contango seen in April and May of 2010. Overall stocks in the US are on the high side, new pipeline capacity is being filled to deliver more northern crude oil to Cushing, and all is justifying a contango structure in Cushing but what is being priced in the current extra-ordinary premium of Brent to WTI is not an extra-ordinary weakness in WTI spreads per se (that was the case in 2009 through the timespreads).

It seems he feels it’s not necessarily about extraordinary weakness in WTI time spreads this time round.

Meaning: it may have less to do with the intensity of the contango structure and more to do with a fundamental shift:

As far as inherent strength in Brent leading it to an extreme premium to WTI we have to observe that the timespreads in Brent are not tightening as the premium goes wider and wider, if anything they have slightly weakened.

It is because of the lack of significant strength in the Brent timespreads that the Brent premium to WTI is able to move through the curve. What is trying to be priced currently is therefore not really a market specific supply and demand in Europe or in the US but a move of WTI to a structural discount to Brent and at a very significant level.

The greatest impact, in that case, will be on the pricing of products since these still trade as a differential to WTI:

However given the implications that this has on cash differentials, cracks, arbs etc we remain unconvinced for now about the sustainability of the current Brent premium to WTI. The traditional product cracks (HeatOil versus WTI, RBOB versus WTI) have become totally irrelevant in our opinion as they have become only an extension of the Brent-WTI spread.

Which could mean it’s time for some new spread-sheets on financial energy desks, we would say.

Related links:
WTI’s upcoming ‘Keystone’ problem
– FT Alphaville
Correlation trading and the WTI-Brent spread
– FT Alphaville
US mulls ’supplemental’ draft EIS for TransCanada’s Keystone XL: Clinton
– Platts
Brent’s got its problems too
– FT Alphaville
‘WTI about as useful as a chocolate oven-glove’
– FT Alphaville