Back in January, when collapsing the Brent-WTI trade was all the rage after the spread had hit all time wides, we cited a JPM report which contrary to Goldman (which 6 months ago had seen WTI higher than Brent) warned that the spread was likely to persist and even widen, and for once, agreed with Jamie Dimon's firm, cautioning: "those who believe that a compression trade between the spot curves is a slam dunk: be very careful." Sure enough, some decided to be brave and sell Brent while buying WTI. Considering today the spread just hit a new all time record of over $23, those brave souls have now been wiped out. Yet that does not explain why the spread continues to diverge, and has recently taken a sharp $7 jump in just the past few days. Below we present David Greely's latest thoughts on what the reason for this unprecedented divergence is, on why he has been dead wrong, and why he believes, eventually, he may be proven right, even as Goldman's prop desk has almost certainly milked this move for its entire duration.
After trading in a range between -$10/bbl and -$15/bbl for the past 3 months, the WTI-Brent spread collapsed at the beginning of June, to a record low close of -$21.80/bbl on Monday (see Exhibit 1). The recent collapse of the WTI-Brent spread raises something of a puzzle in that the usual suspects, the logistical issues surrounding the WTI delivery point in Cushing, Oklahoma, are not to blame.
It is true that the logistical issues surrounding Cushing are responsible to a large extent for the wide discount of WTI to Brent this year. In particular, the WTI-Brent spread set its prior record low in February of this year when the opening of the Keystone pipeline into Cushing and a string of unplanned refinery outages in the US midcontinent led to the dislocation of the US midcontinent crude oil market from the rest of the word crude oil market. Further, the logistical issues surrounding Cushing also seem to bear responsibility for the WTI-Brent spread remaining wider than we anticipated from March through May. However, the recent June decline of $7.77/bbl in the WTI-Brent spread has been primarily driven by the weakening of the US Gulf Coast light-sweet crude oil prices relative to Brent crude oil prices, not a Cushing bottleneck.
More specifically, we can break the WTI-Brent spread into two legs: the WTI-LLS spread and the LLS-Brent spread. The first represents the light-sweet crude oil price differential between the US midcontinent and the US Gulf Coast markets. The second represents the light-sweet crude oil price differential between the US Gulf Coast and Northwest Europe. Typically, we would expect a Cushing bottleneck to widen the WTI-LLS price spread. This is what happened earlier this year in February. However, the WTI-LLS spread has traded in a range between -$12/bbl and -$18/bbl since March, and has not experienced a sharp decline recently (see Exhibit 2).
On the other hand, the LLS-Brent price spread has collapsed recently, with LLS declining to a $5.45/bbl discount to Brent (see Exhibit 3). Because both LLS and Brent can be moved by tanker, the LLS-Brent spread reflects shipping costs. Consequently, the recent decline in the LLS-Brent spread suggests that the arb between the US Gulf Coast and Europe has flipped, and the oil market is now directing light-sweet crude oil away from the US Gulf Coast and toward Europe.
More details on why the recent collapse in the spread can not be blamed on Cushing this time:
While crude oil inventories at Cushing remain high, supporting the wide WTI-LLS spread, they have been drawing fairly consistently since peaking in early April at 41.9 mmb. In the past two weeks, they have drawn by 1.2 mmb, falling back below the 40 mmb mark for the first time since February (see Exhibit 4). Further, storage capacity at Cushing continues to grow. The US Department of Energy recently reported that shell crude oil storage capacity in Cushing reached 57.9 mmb by the end of March, with working storage capacity of 48.0 mmb. This implies that Cushing inventories have room to build 9.1 mmb, or that Cushing inventories are currently occupying 81 percent of working storage capacity (see Exhibit 5). Consequently, while inventories at the WTI delivery point remain high, storage capacity is
by no means full.
Reflecting the spare storage capacity available in Cushing, near-dated WTI timespreads have strengthened considerably since their collapse in February, when a string of refinery outages in the US midcontinent in the wake of severe winter storms undercut demand for crude oil in the region (see Exhibit 6). Further, WTI and LLS prices remain highly correlated. This suggests that the while WTI is trading at a steep discount to LLS, the US midcontinent crude oil market has not dislocated from the US Gulf Coast oil market like it did in February. Instead, the two markets remain integrated (see Exhibit 7).
Consequently, neither the behavior of prices or fundamentals point to the logistical issues surrounding the WTI delivery point in Cushing, Oklahoma as the main driver of the recent June collapse in WTI-Brent spreads. Rather, the evidence suggests that the recent collapse in the WTI-Brent spread has been driven primarily by the flipping of the light-sweet crude oil arb between the US Gulf Coast and Europe, which is now directing flows away from the US Gulf Coast and toward Europe and Asia.
Unlike last time around, Greely, who still is confident that the spread will eventually have to close, is far less confident (to be expected after loosing clients their capital on any 50% leveraged position).
At this point it remains far from clear if the recent flipping of the transatlantic light sweet crude oil arb is sustainable. We believe that it is most likely that it is not sustainable for more than a few months, and we continue to expect the WTI-Brent spread to narrow. However, we continue to refrain from making a trading recommendation on the WTI-Brent spread as we believe that the risk associated with the fundamental uncertainty is too high. In order to highlight these risks, we can envision several scenarios that may play out, some of which would suggest the spread remains wide, others that it collapses quickly. In each scenario, we find that the size of the LLS-Brent spread should be limited by tanker rates which have been running between $3-$4/bbl on the Northern Europe-US Gulf Coast route, which with the LLS-Brent spread now at -$5.45/bbl suggests that downside risk in the LLSBrent spread is limited from current levels, and that the arb has likely overshot recently. However, while we expect the WTI-LLS spread to narrow as Cushing continues to draw, we continue to believe that WTI and the US midcontinent oil market will remain prone to dislocations, which will continue to pose downside risks to WTI-Brent spreads until the logistical issues at Cushing are resolved.
So instead of piling insult on insolvency, Greely this time provides readers a sampling menu of three scenarios as to what may happen to the spread.
Scenario 1: Recent supply disruptions in the light sweet crude oil market are driving the flipping of the arb, arb likely to revert back in coming months
In our mainline scenario, we believe that the flipping of the transatlantic light-sweet crude oil arb is likely a short-term response to the supply disruptions that continue to plague the light sweet crude oil market. Of course, the largest supply disruption has been the loss of Libyan production, which removed over 1.5 million b/d of light sweet crude oil production from the market. More recently, however, other supply disruptions have occurred, including:
- Nexen brought forward planned maintenance on its Buzzard field in the North Sea to June and July from September. The field produced 80 thousand b/d in May, and it is not expected to return to full production until the end of July.
- Royal Dutch Shell’s Nigerian unit declared force majeure on planned loadings of 200-250 thousand b/d of Bonny Light crude oil in June and July. According to a company spokesperson, the pipeline was repaired and production restarted on June 12.
Consequently, it is likely the case that the flipping of the transatlantic light sweet crude oil arb is being driven by a near-term shortage of light sweet crude oil due to the recent and ongoing supply disruptions. As these supplies reenter the market, we could then expect to see the arb revert to its former direction, with LLS rising back above Brent.
Such a reversal could be accelerated by a resumption of Libyan crude oil production, or more likely in the near term, by increases in Saudi production. Reports have stated that Saudi will increase its crude oil production to 10 millon b/d in July. To the extent that the new Saudi super-light blend can offset the shortfall in light sweet crude oil, this increased supply could allow the LLS-Brent spread to revert to more normal levels.
Scenario 2: The process of resource reallocation has accelerated, and will sustain the flipping of the arb as oil is redirected to the emerging markets
The downside risk to the spreads scenario is that the flipping of the arb is sustained over a longer-term period as the market’s means of redirecting oil on a structural basis to meet the growing demand of the emerging market countries. We have long argued that in response to tightening supply constraints, oil prices are rising in order to bid oil away from developed market consumers in order to supply the growth in emerging market demand. We have called this process “resource realignment.” It could be the case that the flipping of the transatlantic light sweet crude oil arb is another step in that process.
As we have been observing for some time now, the drawdown on the overhang of US petroleum inventories built up during the economic recession has been driven not by renewed strength in US oil demand, but by a reduction in net US oil imports. Up until now, this has been largely a feature of the US petroleum product markets, where lower imports and higher exports to South America and Europe, have driven the draw on US petroleum product inventories. Meanwhile, US crude oil inventories remain high, even along the US Gulf Coast (see Exhibit 10), supported by low US refinery runs. Consequently, it could be that the oil market is now directing light-sweet crudes away from the US Gulf Coast as an extension of the rebalancing process that has been under way since the beginning of the US recovery.
Further the impact on the LLS-Brent spread of this process of resource realignment could be augmented by the growth in US lower-48 crude oil and condensate production. With the US lower-48 becoming one of the leading growth regions for Non-OPEC supply, we may be beginning to see a trend of light-sweet crudes and condensates making their way past and around the Cushing bottlenecks to the rest of the United States. With crude oil production from the Bakken shale expected to grow by 180 thousand b/d this year and the Eagle Ford increasing production by 75 thousand b/d over the past 12 months, the United States will likely require less light-sweet crudes from the seaborne markets, allowing these crudes to flow to the rest of the world (see Exhibit 11).
Under this scenario, the reversal of the transatlantic light sweet crude oil arb would be sustainable. It would also suggest that the tightening that we expect in the oil market in 2H2011, which we expect to push the oil market to critically tight levels in 2012, has likely accelerated significantly. This would suggest this scenario presents upside risk to our Brent crude oil forecast.
And last, Scenario 3: Brent prices have risen too quickly, with a market correction to be driven by increased crude oil flows to Europe.
In the upside risk to spreads scenario, it could be the case that Brent have risen too quickly in the anticipation that the prior scenario is unfolding. That is, the market is pricing as if the tightening has accelerated, even if it hasn’t. Under this scenario, we would expect a market correction in the near term as the flipping of the transatlantic light sweet crude oil arb would motivate participants in the physical market to redirect crude oil flows to Europe, forcing Brent prices back in line with fundamentals. The ability to sell Brent forward to lock in the profit on shipping an LLS cargo would suggest that the correction could begin even before the physical volumes would arrive in Europe. While there are no clear signs this is the case, Brent prices have risen faster than we have anticipated based on our fundamental outlook and have been quite strong in the face of disappointing economic data. This scenario would, of course, present downside risk to current Brent crude oil prices.
All that said, Goldman would still like clients to sell the spread.. to its prop, pardon, flow traders.
Under any of these scenarios, however, we believe that the downside risk to spreads from current levels is limited.
The LLS-Brent spread closed Monday at -$5.45/bbl. Tanker rates between Northern Europe and the US Gulf Coast have ranged between $3.50-$4.00/bbl in recent months according to Drewry. This suggests that the current spread would cover the cost of pulling oil from the US Gulf Coast to Europe, and is more than large enough to pull West African cargoes to Northern Europe and away from the US Gulf Coast. This suggests in turn that the declines in the LLS-Brent price spread will likely be limited from here, and that the arb may have overshot recently.
This implies that any further deterioration in the WTI-Brent spread would need to be driven by Cushing-related issues, which would weaken the WTI-LLS spread. However, should the WTI-LLS spread hold in its recent range of -$14/bbl to -$18/bbl, the WTI-Brent spread would likely not fall significantly below -$22/bbl. As we expect that the inventories in Cushing will continue to draw and the WTI-LLS spread will compress in the coming months, we continue to expect WTI-Brent spreads to strengthen substantially from current levels.
Bottom line: at this point, courtesy of bizarro centrally-planned markets, expect the spread to do the opposite of the logical, and of what Goldman "expects" it to do.