Bakken: If Railing Crude is Wrong I Don’t Wanna Be Right

Overbuild!!! The word strikes at the heart of any midstreamer.  Call it the curse of capacity.  There is a shortage of capacity to do something, like transport crude oil, or store natural gas, or process ethane.  You see the opportunity and jump on it, developing infrastructure where it is needed.  But you are not the only smart person out there.  Others jump on the opportunity as well.  And before you know it…. Overbuild!!! Too much capacity has been built and the economics that supported the new infrastructure have been crushed. 

And where do you think the foul word has been used most often lately?  You guessed it.  That island of capacity shortage, the Bakken.   Yes, capacity is still short today.  But there are a lot of projects in the works.  So many pipeline and rail projects out of the Bakken that an overbuild is being predicted by some of the most knowledgeable folks in that market.  In an overcapacity situation it is the high cost alternative that loses out.   And that is rail. 

On Thursday of last week at Day #3 of Bentek’s Benposium conference in Houston the seemingly unlikely bearer of this news was Tad True, one of the most knowledgeable players in all things Bakken, VP of the True Companies and scion of the True family businesses which includes pipelines, production, trucking – just about anything related to the crude oil business.  The True family was into Bakken crude way before it was cool.

Tad laid out a model that his team developed along with the North Dakota Industrial Commission to explain and forecast crude oil take away capacity from the Bakken – and then he explained why the model is ‘wrong’.  As you might expect from our title, the model indicates that railing crude is wrong in the sense that much of the rail capacity being built is in excess of what is needed to move crude production out of the region for much of the next decade.  But that’s not right.  Confused?  Let’s look at the model assumptions.

The model presented by Tad assumes that crude oil flows out of the basin will seek the most preferred route, and when that is full go to the next most preferred, when that is full the next, and so on.   (a) In this model the most preferred route is via pipeline to traditional Bakken local markets – Tesoro’s Mandan refinery, the Guernsey hub, and the Clearbrook hub  (for more about Guernsey and Clearbrook see Bakken and a Rollin).  (b) The next most preferred market is via pipeline to non-traditional markets (Cushing, Casper & Salt Lake City).  (c) Next is unit train rail volumes (since unit trains of 80-100 cars are the lowest cost rail alternative.  And finally we have manifest trains – less-cars-than-unit trains.  Manifest rail transportation is both slow and expensive.  In summary this model says that barrels will move on the least cost transportation alternative available.

Tad provided a lot of detail on individual projects that we don’t have time to review here.  But this graphic from his presentation makes the point.  Each project is a “layer” in the stacked area graph, with pipeline projects in green and rail projects in blue.  The black line is the projection of crude oil production.  We’ll talk more about that line later. 

From 2007 through mid-2010 there was no rail, and take away was constrained by pipeline capacity.  Rail kicked in and provided much needed relief.  More new rail terminals were built and remain a critical part of the equation today.  By late 2012, several new pipeline projects (True/Butte, Enbridge, etc.) will get pipeline capacity to the point that it will be able to handle 100% of all production.  It stays like that off-and-on for another three years when production again needs the rail capacity.  But not all of it.  Only 20-50% of rail capacity is required to move the production in the post-2015 years, based on this model’s view of the world.

Basically this scenario says that rail is a critical piece of the equation until the end of this year.  Then a lot of the rail capacity will be surplus for three years until production catches up.  For brand new facilities needing to pay back their investors, this sounds like railing crude is wrong.  But not so say the rail terminal operators, and Tad agrees.  The model is wrong for at least five reasons.  Four are from Tad, the fifth I’ve added.

1. It ignores volume commitments to projects (e.g., Take or Pay commitments to pipelines and rail facilities).  Terminal operators don’t build these things on spec. (most of the time, anyway).  They have commitments from producers for fee guarantees that they can take to the bank.  Otherwise a lot of these terminals would never get built.  So the volumes will flow through the terminals anyway.  Or at least the dollars will flow to the terminal operators.
2. It ignores Intra-Regional Constraints.  Producers have to get their crude to one of these facilities, frequently by truck.  Even if the pipeline has a better deal, the pipeline receipt point may be too far away to matter.  It is particularly important to minimize the distance that trucks must travel so the routes don’t tie up too many trucks traveling long distances, resulting in a shortage of trucks.
3. It ignores different rail constraints.  The surplus capacity situation might not be nearly as large as it looks because a lot of these rail facilities might not be able to operate at anywhere near their stated capacity.  That could be due to a shortfall in the number of available railcars, limitations on delivery facilities at the other end of the line, etc.
4. It ignores geographic price differentials.  Pipelines take crude from A to B.  Rail can take advantage of moving crude from A to wherever the market is best.  If B gets oversupplied from pipeline deliveries, railcars can simply go somewhere else.
5. And there is one more big factor.  This model is based on a peak Bakken production number of 1.1 MMb/d.  I figure that the number could be 10-20% higher.  Perhaps more.  The higher the number, the better for rail.

So rail is not wrong after all.  There are lots of reasons why the rail terminals will see significant throughput even when the pipelines are not full   And I’ll bet as markets in along the East Coast, West Coast and Gulf Coast build out more receiving terminals, crude oil producers will find the optionality that comes along with rail shipments will offset higher transportation costs in a lot of situations.  But don’t cry for the pipeliners either.  Most of the pipes will see high capacity utilization, most of the time.   

We’ll take a closer look at some of True’s pipeline projects in another posting later this week.

[If loving you is wrong I don't want to be right was a hit by Luther Ingram in 1972 and has been covered by scores of other artists for the past 40 years.]