BNSF Coal Stockpile Recovery Curve, May 2014-January 2015

BNSF’s tendencies to overgeneralize plans for improving coal deliveries have annoyed coal users and grain shippers alike. The railroad has publicized their $6 billion capex planning for 2015, while not really clarifying their specific coal service improvement plans. Their slowness in rebuilding coal stockpiles has exasperated and frustrated a very large group of customers and public officials, causing a growing crowd to demand answers of the Surface Transportation Board (STB). Feeling somewhat cornered and irritated, the board answered on the eve of STB Chairman Dan Elliott’s departure.

In March 2014, a frustrated group of electric utilities forming the Western Coal Traffic League (WCTL) sent a letter to Elliott complaining about a rail service crisis caused by BNSF. They had complained a month before, but seeing no turnaround in BNSF’s service, decided to make it formal. They listed problems specifically caused by BNSF’s slow service: low stockpiles, forced price increases, increased electric costs, no train sets for shippers using railroad-owned cars, poor stockpile rebuilding efforts, etc. These were utilities that decades before had committed to purchasing low-sulfur coal from the Powder River Basin (PRB) of Wyoming and Montana to meet increasingly stringent environmental laws, companies whose investments had made it possible for BNSF to grow into a financial behemoth.

Many of these utilities had purchased hundreds of their own coal hopper cars to guarantee continual rail service to their coal-fired power plants. The railroads provided the power; they provided the hopper cars. Over the past several years, many coal-fired plants in the east and some in the west have closed. Some utilities might be stuck with unused hopper cars; it is not beyond reason that BNSF and other railroads would be stuck with unused power.

On October 22, WCTL petitioned the board to require BNSF to submit a coal-specific recovery plan. WCTL member Minnesota Power had idled four of its coal-fired generating units after delayed deliveries of PRB coal forced them to conserve stockpiles in anticipation of the usually harsh winter in northeastern Minnesota. Two days later, the STB ordered BNSF to file a reply to WCTL’s petition by November 3.

BNSF submitted a letter to the board on October 28 describing recently completed and intended infrastructure projects, which, according to them, should benefit its transportation of coal. The letter went on to say that preparing and filing a specific coal recovery plan would not contribute materially to its customers’ perspective on its operations. Their approach was to tell the board about their $6 billion capex plan for 2015 and let the board assume the delivery problems would be solved by overall system improvement, the “rising tide lifts all boats” management philosophy. Unfortunately, this approach did not appear to relieve short-term coal inventory problems that deeply concerned the WCTL members.

BNSF’s reply to WCTL’s petition came on November 3. To say they were opposed to the WCTL request would be an understatement. In fact, their belligerent response was uncalled for, particularly when it practically insulted outgoing Chairman Elliott, who was an excellent railroad lawyer and would be retiring in two months:

1) “The board lacks authority to mandate service requirements;”

2) “The board cannot compel BNSF to take specific actions related to service;”

3) “The board cannot grant the relief requested by WCTL because the ‘vast majority’ of BNSF’s coal traffic moves under private contracts;” and

4) “The relief requested by WCTL would detract from its overall recovery efforts.”

It may very well be true that the STB’s authority is limited in the matter of contract business, but one could easily make the argument that BNSF’s coal service deficiencies potentially affect a much broader spectrum of the American economy than that bounded by individual rail service contracts. It was not likely BNSF could force the board to back down, and they did not push the issue any further after they made their four boasts.

The backing that WCTL received over the October 31-November 19 period was incredible. Starting with comments filed by the American Public Power Association, there were comments by Dairyland Power, ALLETE (Minnesota Power), Otter Tail Power, Ameren Missouri and Consumers for Rail Equity. Support for the WCTL petition was also expressed by Minnesota Gov. Mark Dayton, U.S. Sens. Amy Klobuchar and Al Franken of Minnesota, and from Wisconsin, Sen. Tammy Baldwin and U.S. Reps. Ron Kind, Sean Duffy, Thomas Petri, Reid Ribble and Paul Ryan. Ryan also wrote that if delivery problems persist, the board should require all Class 1 carriers to submit publicly available coal service recovery plans and monitor carriers’ progress through weekly reporting.

During the board’s two rail service hearings and in its recent filings, BNSF generally acknowledged that it had not met customer expectations with regard to its movement of coal. It also noted that it had been working aggressively toward remedying ongoing service issues. To address these issues and improve its coal service, BNSF stated that it had undertaken the following initiatives: increasing hiring, locomotive and car acquisitions, and capital investment in maintenance and capacity expansion; investing in northern corridor infrastructure, including network capacity expansion; adding two double-track projects to its infrastructure investment plan to support its coal route; making network-wide investments, including terminal and capacity expansion projects; improving fluidity on the coal network through maintenance projects across the network; preparing enhanced winter action plans, including new resources for the 2014 winter season to better handle extreme weather; and decongesting the network by strategically removing a small number of coal sets. Apparently, this answer was much too general.

Finally, bone-weary of BNSF’s obfuscation, the STB ended 2014 by directing the railroad to submit coal transport contingency plans by January 29. This was an invitation and a directive to be specific. The only thing missing from the decision was “or else,” possibly because Elliott did not want to start a word war upon his departure, and perhaps he figured Ryan’s threat would get the job done. It was a courageous act that he called BNSF’s bluff, even at the risk of making a powerful enemy when he needed friends. (STB Docket No. EP 724.)

BNSF Executive Vice President Steve Bobb answered the board’s directive honorably, on time and with a fair amount of detail on January 29. There was no impertinent reference to the board’s “lack of authority.” However, whether he responded in sufficient detail to satisfy WCTL, the board and all other complainants remains to be seen.

Bobb said, “the most effective way to address coal stockpile fluctuations is to improve velocity across our network to a level where we are no longer managing resources to respond to critical situations.” He went on to describe BNSF’s “coal customer escalation process, an existing process that BNSF has utilized to address potential concerns arising from recent service issues in coal transportation.”

He provided a chart showing how BNSF has rebuilt stockpiles by 440 days between May 5, 2014, and January 15, 2015, a time of relative quiet weather wise. Total rebuilt inventory gain or loss is plotted over critical coal stockpile customers’ net inventory days gain or loss for central, south and north customers. The vertical bars represent data for entire regions, obscuring much different results for individual plants within those regions. The rising curve should give encouragement to even the most critical observer were it not for the fact that regional statistics can mask what might have been happening at an individual power plant.

BNSF has a very deliberate process that they use on a daily basis for determining service needs and deficiencies, and for responding to specific critical stockpile situations. As Bobb pointed out, they “must have the flexibility to adjust service recovery efforts.”

A key to their recovery effort lies in identifying coal plants with “critical stockpile challenges,” using “days of burn” as the critical metric. If a customer’s stockpile is at 20 days or less, that customer is designated internally as “critical.” When a customer does notify BNSF that it has less than 10 days of burn, there is a heightened alert initiated by service design personnel that identifies the physical location of each train set moving to that utility as well as the current disposition of each train. Also, the board’s staff is provided with updates on the customers who have plants on this critical list; customers are dropped off the list when they have 20 days of burn or more in their stockpiles. The intensification of communications is laudable, but what will the railroad actually do? Bobb described a number of “operational measures available.”

Bobb described several tools for recovery of dwindling stockpiles: adjusting the number of coal train sets in service, reallocation of locomotives (perhaps from one business unit to another), route and gateway adjustments, maintenance planning, and crew management. While stockpile recovery is clearly a short-term problem, all but the first two appear to lack short-term flexibility. Also, adjusting the number of train sets within a given utility’s fleet certainly seems plausible, but only if that utility’s other stockpiles would not be harmed by their temporary absence from service.

The BNSF Coal Desk is in constant communication with customers to get stockpile information, saying they have no direct visibility into the customers’ stockpile levels. Some customers provide stockpile information more regularly than others, and some seem to use stockpile data to secure preferential treatment. The uncertainty problem might be tackled somewhat by requiring each coal plant to submit a three-month rolling forecast of coal needs each week.

Bobb does a good job of explaining the BNSF system for monitoring and dealing with coal stockpile levels with their customers, and it is a good macro system. Missing from his discussion are answers to some questions that deal with specific micro problems such as, “Where are most of your problems now? What are you doing to overcome them? Do you have railroad construction projects that can be shut down on a moment’s notice? Do you expect to obtain some new or leased locomotives in the near future? Do you plan to cut back on a specific business unit(s) if a coal inventory problem arises? In short, how do I know you plan to deliver?”

One may ask why the board’s decision did not include Union Pacific (UP), since both railroads serve the same set of mines in the southern PRB. The simple answer is that the slow service complaints have primarily centered on BNSF. Data filed with the STB for 2014 show that the average train speed for PRB coal for UP was 24.9 mph, while that for BNSF was only 17.7 mph.

It is reasonable to assume the BNSF average speed in the southern PRB was about the same as that of UP, since they use the same sets of tracks.

If their southern PRB velocities are about the same, it follows that BNSF’s average coal traffic velocities on its northern routes were much less than 17.7 mph. The major difference is that UP does not travel the northern routes where the Bakken oil trains originate.

At a Federal Energy Regulation Commission meeting held December 18, Bobb apologized to Minnesota Power’s Dave McMillan for delivery problems, saying those problems were brought on by sharp spikes in shipments of crude oil from the Bakken Fields in North Dakota, bumper crops of grain in the United States and Canada, and increases in the transport of manufactured goods. He did not say whether BNSF was the source of most of those spikes.

In the five years since BNSF began hauling Bakken crude out of the Williston Basin, it has seen a huge increase in the amount it ships. Its network reaches all major coastal and inland markets and directly serves 30% of U.S. refineries in 14 states. Currently, the railroad has 1,000 miles of rail line in the Williston Basin area. It connects to 16 of the top 19 oil-producing counties in central and western North Dakota and five of six oil-producing counties in Montana.

White House and congressional delays in the Keystone XL Pipeline project has enabled both BNSF and Canadian Pacific, which hauls the other half of the Bakken oil, to strengthen their positions as product carriers. Matt Rose, BNSF chairman, said, “We see a path to 1 million barrels per day.” Running 100-car unit trains at 60,000 barrels per train would require 16-17 unit trains per day. In 2013, BNSF hauled about 325,000 carloads of Bakken crude for EOG, delivering the oil to refineries north of New Orleans where it fetches a premium.

BNSF recently announced its 2015 capital projects, revealing plans of investing roughly $1.5 billion across eight states for engineering and expansion projects. Its north region has experienced the most rapid growth in recent years and is a destination point for materials that support the production of crude oil in the Bakken.

The company also recently issued a request for proposal to major railcar manufacturers to submit bids for the construction of 5,000 next-generation tank cars used to move crude oil. Tank cars are normally owned by oil shippers, but many are not constructed to haul the Bakken crude safely. BNSF’s anxiety to order 5,000 cars suggests they do not want wait for private car owners to make up their minds, and that they are anxious to start hauling more crude.

Dave Gambrel is a private transportation consultant. His experience extends to major modes of coal transportation, including rail, barge, ocean vessels and deepwater terminals. He has authored the Transportation Tips column since May 2010 and has also written for Engineering and Mining Journal. He may be reached at