Pipeline, rail, barge… any way out!
Prior to the shale and tight oil boom in the U.S. and significant expansion of oilsands in Canada, the U.S. and Canadian system for delivering crude oil to market was stable and relatively predictable. In general, the U.S. and Canadian crude oil pipeline networks were originally designed for taking crude oil into the U.S. Midwest. Then matters started to change as production started to rise, and pricing for WTI at the Cushing, Oklahoma hub, which had always run in close parity with Brent, started to disconnect. Discounts deepened, affecting essentially all inland lower-48 crude grades, as well as Canadian crude oils (since these are also priced off WTI). Since January 2011, these discounts have been steep and have been considered “structural” (see Figure 1). This article will delve into the issues that are behind this soaring discount.
West Coast or the Gulf Coast
The North American crude pipeline system was caught off-guard by expanding production in Western Canada, as well as the Bakken and other shale plays in the U.S. As a result, pipelines are operating at near full capacity and delivering crudes to hubs where lack of capacity leads to congestion, as seen at Cushing, which continues to persist today. It has become a race between expanding supply and attempts to put adequate capacity in place in order to move crude oil to markets beyond the U.S. interior and inland Western Canada.
Two such pipeline projects are currently being reviewed. However, no one would have anticipated that these projects would become the focus of “political heat” at the highest levels. The TransCanada Keystone XL project, originally intended as a 700,000 to 900,000 barrels per day (bpd) line to mainly carry oilsands streams from Hardisty, Alberta, to the Gulf Coast via Cushing, has become a focal point of the political and environmental pro- and antioilsands debate in the U.S. Likewise, the Enbridge Northern Gateway project that would initially take 525,000 bpd of heavy oilsands streams west to British Columbia’s port of Kitimat—and then to markets mainly in Asia—has become the centre of heated support and intense resistance in Canada.
Since then, the Keystone XL project has been split into two parts: a southern leg project from Cushing to the Gulf that has received all the permissions necessary to proceed, and which is expected to start operations by late 2013; and a northern segment from Hardisty to Steele City, Nebraska (where there is an existing line onward to Cushing), which hasn’t yet received the U.S. presidential permit. Start-up would likely be no earlier than 2015. For the Northern Gateway project, Enbridge has filed an application with the Canadian National Energy Board (NEB), but a review will take at least until the end of 2013. The expected start-up for this is around 2017, but some delays are likely.
The response to the delays on these two headline projects from the midstream industry has been an almost ever-changing array of new developments and proposals. There are already several project proposals related to modifying existing pipelines and/or taking advantage of existing rights-of way to construct new parallel pipelines. A leading example is the 300,000 bpd Trans Mountain pipeline from Edmonton to Vancouver, which has recently been heavily over-subscribed. Currently a spur pipeline carries the bulk of the crude to U.S. refineries in Washington State and another 50,000 bpd has consistently gone to a refinery at Burnaby near Vancouver.
As a result, historically, less than 50,000 bpd of crude has been exported over the one and only export dock that currently exists for Western Canadian crudes.
Operator Kinder Morgan has obtained sufficient shipper commitments to support expanding the Trans Mountain capacity by 400,000 bpd. Much of the increased throughput would be moved over the Vancouver (Westridge) dock, with destinations mainly in Asia. The expansion has a start-up date of 2016, although this could slip because of concerns over the resulting increase in tanker movements in the already-busy Port Metro Vancouver harbour.
Glitches in the system
The Northern Gateway and the Trans Mountain expansion represent the only pipeline projects that would take Western Canadian crude west to the Pacific. All other pipeline capacity moves Western Canadian crudes south into the U.S. Rocky Mountain and Midwest regions, from which there is an onward pipeline to Sarnia in Ontario. After recent expansions, which include the base Keystone system and Alberta Clipper, there is more than 3.5 MMbpd of cross-border capacity from Alberta into the U.S. interior. There are, however, bottlenecks in moving Canadian crudes through and out of the Midwest. We have identified such pinch-points along these crucial oilsands export routes.
- Enbridge Lakehead System south of Clearbrook/Superior – Enbridge’s Mainline moves a variety of crude types from Edmonton and Hardistry to Superior, Wisconsin. From Superior, crudge is delivered to the Pine Bend refinery and Line 5 moves product father east to Sarnia, but the bulk of the throughput is moved south to refineries in the Chicago area via the Lakehead system. These southern legs are capable of transporting 1.3 MMbpd, but also have accommodated increasing amounts of Bakken production at the Clearbrook connection point, northwest of Superior. Enbridge’s Southern Access project added 400,000 bpd of capacity south of Superior when it came online in 2009, and the addition of further pumping stations is planned that will raise the line’s throughput. Nevertheless, growing oilsands and Bakken volumes will keep this route near its stated capacity. Beyond the Lakehead System, Enbridge’s Spearhead System transports crude from Flanagan, Illinois to Cushing, Oklahoma. Spearhead’s capacity is 193,300 bpd. Enbridge is moving ahead with the Flanagan South project, which would add an initial 585,000 bpd of capacity (expandable to 800,000 bpd) along the route of Enbridge’s existing Spearhead Pipeline between the Flanagan, Illinois Terminal, southwest of Chicago, to Enbridge’s Cushing, Oklahoma Terminal.1 Other company proposals, such as the reversal of the existing 1.2 MMbpd Capline pipeline to Louisiana, would also help transport oversupplied Midwest markets south.
- TransCanada Keystone XL – TransCanada operates the Keystone base pipeline, which began operations in June 2010 after converting a gas pipeline from Hardisty, Alberta west to Manitoba and constructing a new pipeline from Manitoba south to Steele City, Nebraska and then east to Patoka, Illinois. In 2011, the line’s capacity was expanded to 591,000 bpd, and the pipeline’s southern path from Steele City to Cushing was extended. The proposed Keystone XL project would establish a direct route from Hardisty to its existing Steele City connection point, but U.S. government approval is needed. Keystone XL also includes plans to construct a new pipeline from Cushing to the Nederland, Texas terminal, allowing access to the Gulf Coast market. Approval for the Gulf Coast portion has been granted.
- Kinder Morgan Express/Platte – Kinder Morgan’s Express Pipeline, with a stated capacity of 280,000 bpd, feeds into Casper, Wyoming where the Platte system moves the crude south and east to Wood River, Illinois. Platte capacity shrinks to 145,000 bpd past Guernsey, Wyoming while also adding volumes from the Rocky Mountain region, including Bakken oil. The constrained capacity on this leg of the pipeline limits the amount of volumes capable of moving south on the Express line. Kinder Morgan’s Pony Express, which will be converted from the existing gas to an oil line, has secured sufficient project support in earlier open seasons to transport up to 230,000 bpd from Guernsey, Wyoming to Ponca City and Cushing, Oklahoma starting in the third quarter of 2014.
These potential pinch-points are in addition to the current and wellpublicized shortfall in capacity to move
crude out of Cushing. Increasing supplies
from Western Canada, the Bakken, the
Permian Basin region (West Texas), as
well as from Oklahoma and Kansas, are all
creating pressure to move crudes mainly
from the north into Cushing and out in
multiple directions, but especially south
to the large refining centres on the Gulf Coast. As well, the lack of capacity to move oil supply directly to the U.S. Gulf Coast is leading to increased volumes arriving at Cushing and adding to the oversupply problem there.
Continental inland expansions, extensions, and conversions
Until recently, there was only one pipeline that flowed south to the Gulf Coast—the Pegasus line (up to 93,000 bpd)—which runs from the Chicago area to the Gulf and carries mostly Western Canadian crude. This pipeline was reversed in 2006 and expanded to its current capacity in 2009. The Seaway line used to flow north to Cushing, but this has recently been reversed with current capacity of 150,000 bpd. It will be expanded to a capacity of 400,000 bpd from Cushing to the Gulf Coast by early 2013 and to 600,000 bpd by mid-2014.
Associated with these expansions is a new Flanagan South line that will use the right-of-way of the existing Spearhead line to add nearly 585,000 bpd of initial capacity from Chicago to Cushing with an in-service date of mid-2014. This will help relieve the bottleneck in the Chicago area and will enable Canadian and Bakken crudes to flow via Seaway to the Gulf Coast. A Seaway reversal and expansion, with the Keystone XL southern leg, will add over 1.65 MMbpd of capacity out of Cushing to the Gulf by 2014. This will substantially alleviate the “Cushing congestion” and should, consequently, narrow the WTI-Brent spread, as well as Western Canadian-WTI differentials, at least in the short term.
Growing Western Canadian and Bakken supplies have also led Enbridge to propose modifying its existing pipeline through Eastern Canada. The system already carries western crudes east as far as the refining complex at Sarnia. Another line (Line 9) originally used to run east from Sarnia to Montreal, but was reversed; is now bringing imported crudes west via Montreal and a connecting Portland, Maine to Montreal Pipeline (PMPL) into Sarnia. Enbridge has now proposed to re-reverse Line 9 with a stated capacity of 240,000 bpd so that it runs east to Montreal, where there is access to two refineries in Montreal and Quebec City. This could also tie-in with a possible reversal of the PMPL to take Western Canadian and Bakken crudes out to the Atlantic, from where they could reach refineries in the Canadian Maritimes, the U.S. East Coast and potentially beyond. Enbridge has already reversed a short section of the line and has applied for permits that would allow full reversal. This project, like Northern Gateway and Keystone XL, is meeting some resistance on the environmental basis, and so its timing is uncertain. TransCanada is also considering switching one or more existing gas pipelines that run from Alberta to Quebec into crude service. The concept is attracting interest and a possible capacity range of 400,000 to 900,000 bpd is being discussed. The main objective would be to carry Western Canadian crudes, including oilsands, synthetic crude and/or diluted bitumen, through to Quebec and then to the 300,000 bpd Irving refinery in New Brunswick. At the moment, this is only an idea and has not been taken to the formal “open season” stage to test the level of commercial commitment. Part of the impetus behind this possible gas line conversion and the Enbridge Line 9 project is uncertainty over the major projects that would move Western Canadian crudes to the west, and south to the Gulf Coast. To the extent that either the TransCanada or Line 9 projects go ahead, they will enable light-sweet and medium-sour crude oils to move to Eastern Canada and possibly also to refineries on the U.S. East Coast and free up some room on the pipelines heading south to the U.S.
Rail, Rail and More Rail
Uncertainties over key pipeline projects, and steep discounts in U.S. lower-48 and Western Canadian crude prices, have spurred the above-mentioned proposals to modify and expand existing pipeline Infrastructure, but they have also led to a growing role for rail. This is especially visible in the Bakken. There has been marked growth in Bakken “takeaway” capacity via rail. Faced with a lack of existing infrastructure in North Dakota, mainly smaller producers and transport companies began a rapid expansion of rail terminals in 2009. These use “unit train” technology (load dedicated 60,000 to 75,000 barrel trains, often one or more per day) that then move crude to corresponding receiving terminals with no stops along the route. Williston Basin rail takeaway capacity went from 30,000 bpd in 2008 to 275,000 bpd by 2011 and was anticipated to reach nearly 730,000 b/d by the end of 2012 (see Figure 2).
Pipeline takeaway capacity is also expanding rapidly, but what is new here is that rail is becoming established as an important mode for moving crude oil, at scale, to multiple destinations. Most delivery terminals for Bakken crude are in the Gulf Coast, but movements are expanding to both the West Coast and, especially, the East Coast. These movements are taking Bakken production—which recently passed the 640,000 bpd mark and is expected to go much higher—into coastal U.S. markets.
The year 2012 may also be the point in time when crude movement via rail starts to catch on as a means to move Western Canadian crudes. Small volumes of Western Canadian crudes have recently moved to the Western U.S., the Gulf Coast and the East Coast, as well as Ontario via rail. What is new is that longer-term commitments and unit train developments are starting to surface—for instance, for the movement of Bakken crudes at scale to the Irving refinery in New Brunswick.
Rail movement via “manifest” train can be three times the cost of pipeline. However, unit trains narrow the gap and shorten the delivery time. Moving oilsands bitumen by rail can come even closer to pipeline costs as less diluent is needed; even bitumen with no diluent can be carried if the rail cars are heated. Given the severe price discounts on heavy Canadian crudes, rail looks to be an attractive option. Both pipeline and rail are also tying in with barge movements, notably from the Midwest to the Gulf Coast, using rail or pipeline for part of the way and then barges down the Mississippi River for the last leg. Within the Gulf Coast, midstream companies are also expanding their options to move crudes along the coast (for example, Eagle Ford crude east along the Gulf and also via tanker up to the East Coast) and to move crude west to rail terminals in St. James, Louisiana.
The end result is that the combination of pipeline expansion, and rail and barge transportation options will enable U.S. lower-48 and Western Canadian crudes to flow in an increasingly less restricted way to coastal markets. Data showed that as of the third quarter of 2012, U.S. and Canadian oil movements by rail had already increased by 650,000 bpd, compared to their historical level. This is consistent with the surge in rail loading and offloading capacity that, by the end of 2012, was anticipated to see over 700,000 bpd of receiving capacity in operation: with over 200,000 bpd on the Eastern coast (U.S. and Canada), 450,000 bpd on the Gulf Coast and 50,000 bpd on the West Coast.
By the end of 2013 and into 2014, this rail capacity will have essentially doubled to over 1.4 MMbpd, with nearly 600,000 bpd of receiving terminals on the Eastern Seaboard, close to 750,000 bpd on the Gulf Coast and around 110,000 bpd on the West Coast. By 2015-2016, this new capacity may well have grown further and will have been joined by 1.65 MMbpd of new pipeline capacity to the Gulf Coast from Cushing. Hence, a total of over three MMbpd of capacity will exist to take U.S. and Western Canadian crudes to coastal markets in the U.S. and Canada. This capacity growth is well-underway; it is developing rapidly and is substantial.
The Bottom Line
The net effect of all these developments is that the U.S. and Canadian crude oil logistics system is changing rapidly as it seeks to adapt to a new reality of steadily growing oil production, both north and south of the border. There is appreciable uncertainty, however, over how the system will evolve in the longer term. It will depend in part on whether (and when) a few major pipeline projects are brought online, as well as on how much Western Canadian crude ends up moving west and to Asia versus south into the U.S. and east into Eastern Canada. By 2014, WTI discounts could be partially alleviated, but we are witnessing a race between production growth and infrastructure restructuring. Crude oil discounts could persist to 2020—and even beyond—if U.S. shale production rises at optimistic rates. The emergence of rail is an important new factor. Although rail car availability is a constraint in the shortterm, terminals are low-cost compared to pipelines, can be put online within 12 to 18 months, and offer shorter payback times.
1 http://www.enbridge.com/~/media/ www/Site%20Documents/Investor%20 Relations/2012/2012_ENB_Investment_ Community_Booklet.ashx
2 Spectra Energy Corp. bought the Express-Platte Pipeline System for $1.25 billion in cash; the deal is set to close in the first half of 2013.
3 Enbridge. Investment Community presentation. December 2012.
4 Expandable to 800,000 bpd.
5 OPEC. “World Oil Outlook”, 2012
6 North Dakota Pipeline Authority. “Great Plains and EmPower ND Energy Expo”. November 2012.
7 EIA. The estimate might include some volumes of NGLs.
Originally published in the November 2012 issue of CERI Commodity Report – Crude Oil. Reprinted with permission.by