Checking the numbers

STB hopes rail performance data will provide clearer view of operating conditions.    In October, the U.S. Surface Transportation Board began requiring all Class I railroads to file weekly performance data to provide a more transparent picture of the rail issues that have recently plagued the nation. 
   The order is an extension of the board’s previous requirement that Canadian Pacific and BNSF file weekly status reports regarding grain shipments. A joint weekly submission by railroads operating out of Chicago, which has been sent to the board each week by the Association of American Railroads, gives a summary of Chicago operating conditions.   
   Beginning Oct. 22, seven carriers filed weekly reports to the board covering everything from terminal dwell times and the volume of goods moved through their system to average train speed. The board has also been taking a closer look at grain levels, requiring the railroads to submit data on grain carload totals by state.  
   Looking at the numbers so far regarding dwell times and total system volumes for the first six weeks of the program, it’s easier to see how an increase in volume adds to the time it takes railcars to continue moving through terminals.
   Throughout November, Canadian Pacific saw its dwell time increase, while total cars on the network in the United States stayed relatively similar. During the week of Oct. 19-25, CP ran 31,787 cars through its network; each car spent an average of 18.3 hours in a terminal before it could move on. The next week, that car measurement climbed to 32,682, but the dwell time increased to more than an hour per car, at an average of 19.4 hours. A trend like this is expected because of the added volumes, and dwell time mirrors changes in volumes for the next two weeks, even falling back down to an 18-hour average with an increase in total cars during the middle of November. But things started to change in the last half of the month.
   During the week of Nov. 16-22, dwell time increased to 20.3 hours with a total volume of 32,142 cars, a drop of more than 750 cars from the previous week. During the last week in November, car volumes rose back to 32,514, but dwell time increased to an average of 22.9 hours.   
   To explain the large rise in dwell time, Robert Johnson, senior vice president of operations at CP, wrote to the board that Thanksgiving volumes backed up the system. 
   “The biggest driver of our metrics this past week was the Thanksgiving holiday,” he explained. “Volumes into the holiday were strong at Chicago, in particular traffic bound for eastern Canada. The understandable slow-down throughout the supply chain resulted in elevated dwell.”
   Johnson noted that he expected improvement during the week after. 
   He added, “we continue to work on the problems in the logic used to generate the data points for cars on line that have not moved in revenue service for more than 120 hours, or have not moved in more than 48 hours but less than 120 hours.”
   On the whole, Canadian National had faster dwell times for a larger volume, but it also saw dwell times increase over the period, and that increase is not necessarily tied to a ramp-up in cars on the system. For the first week measured, terminal dwell time averaged 15.4 hours for 47,934 cars. The dwell time and car count stayed similar the following week, but as carloads moved past 50,000 in the middle week of November, dwell times suffered, moving up to 17.53 hours, even with a nearly 1,000-car average drop from the previous week. By the last week of November, dwell time had climbed to an average of 18.95 hours with a volume of 50,250 cars, a 2,000-car decline from CN’s busiest week in November. 
   Judging from CP’s and CN’s results, the overriding trend is an increase in dwell times as the month progresses, no matter the car count each week — though there are exceptions. While performance concerns started out in the agricultural field, the STB has required statistics across a wider range of commodities. Taking a look at the last week in November, using reported data regarding performance expectations versus actual performance, CN had planned to load an average of 325 cars of coal each day, and the railroad exceeded that number by 51 cars. (CP shipped no coal.)
   CP and CN also have to submit to a weekly volume-minimum from the Canadian government to make sure enough grain is being moved throughout the system. It goes without saying that their volumes are more extensive north of the border. Starting in March 2014 and running through June, the government required railroads to carry 500,000 metric tons of grain per week. According to the U.S. Department of Agriculture, a railcar can carry between 100 and 112 tons of grain. Assuming each car carries 100 metric tons of grain, the Canadian government had required Canadian railroads to carry 50,000 carloads per week, a steep increase in grain volumes compared to the total volumes in the United States. (For the week ending Nov. 15, CP loaded 888 cars of grain in the United States.) The requirement inched up to 535,230 tons per week from Aug. 1 to Nov. 29.  
   The Canadian edict came with fines of up to $100,000 per day for non-compliance, a little more incentive than what the STB could muster. When the new policy was first announced, Rep. Kevin Cramer, R-N.D, argued in a letter to the Canadian government that the fix for rail problems up North would only make issues seen in the United States that much worse. He noted the United States will have to “reciprocate” to counteract the negative effects of Canada’s policy. 
    “The potential discrimination this effort fosters against CP’s and CN’s U.S. customers is counter to Beyond the Border and overall general cooperation between friendly nations such as ours, as well as the combined economic power our countries have established,” Cramer wrote. “Unilateral parochial endeavors such as this potentially undermine our entire integrated system.”
   Regardless of the backlash, Canada saw benefits from its program, as it subsequently extended grain-volume minimums, eventually requiring railroads to keep up a defined pace through March 28. The government isn’t asking for 500,000 tons of grain per week, but has varied the volume requirements to mirror expected agriculture flows. 
   At the time of its initial passage into law, Canadian Agriculture Minister Gerry Ritz said bumper crops have put “significant pressure” on the nation’s rail system and these minimums are needed to improve service. 
   “This legislation creates the necessary tools to help ensure Canadian shippers have access to a world-class logistics system that gets their commodities to market in a predictable and timely way,” Ritz said in a statement. “Farmers and our economy need a system that works today and tomorrow, with the capacity to move what is grown.”
   Through Jan. 3, Canadian railroads must move 200,000 tons from Dec. 21 to Jan. 3. The requirements ramp back up starting Jan. 4, when carriers must move 325,000 tons of grain each week through Feb. 21. From the end of February to March 22, the government will require railroads to transport a weekly grain load of 345,000 tons. Between March 22 and March 28, carriers must move 465,000 tons of grain.  
   The largest Class I carrier by car count, Union Pacific, reported 293,975 cars moved during the week of Oct. 19-25. These cars spent an average of 28.7 hours of dwell time. That dwell time is on the high end of the scale, but UP’s car count must be considered. Of the U.S.-based carriers Kansas City Southern had the fastest dwell time during that week at 21.9 hours, but it also moved, by a wide margin, the fewest cars. KCS’ volume numbers, however, are nearly comparable to the results turned in by CP.  CSX, with a car count of 207,239, had the second best dwell time at 26 hours. Norfolk Southern, which moved 191,221 cars during the period, had the worst dwell time at 30.5 hours. BNSF’s total for the week came to 259,763 cars spending an average of 29.2 hours at terminals.  
   BNSF was the other carrier being monitored since last spring as laid out by the STB’s earliest status order. It actually saw its dwell times decrease during the last week of November, but carloads fell each week during the month. Throughout the month, dwell times had averaged between 29 hours and 29.3 hours; BNSF saw 257,588 carloads that first week, but was down to 256,296 cars during the third week of November. During the last week, car count was down to 255,966, as dwell times bottomed out at 28.5 hours, the lowest measurement since the last week of October, when the railroad measured an average time of 28.4 hours at 258,902 cars.
   Making the argument that one can’t draw conclusions about service levels simply by looking at a few numbers, BNSF’s weekly submissions to the board come with the caveat.
   “We repeat our earlier caution against drawing firm conclusions based on the absolute values reported in BNSF’s report or across the various railroads that are also submitting data,” the railroad stated. “BNSF will also continue to engage frequently and substantively with our customers through direct conversations, and through broader communications and letters, customer forums, meetings and broadcasts to provide real-time information around our service challenges, our short-term and long-term plans to increase network velocity, and our progress against those plans, and to ensure we hear their perspectives and feedback.” 
   For the rest of the Class I carriers, the trends seen by CN, CP and BNSF are pretty much the same as they experienced. By the first full week in November, and the third reporting week in the study, CSX’s car count was up to 296,687, and dwell time had increased to 30.1 hours. Half of the remaining Class I carriers actually saw a decrease in dwell time on the last week of October, with CSX’s time decreasing to 25.6 hours, and Norfolk Southern’s time getting better by half an hour; in Norfolk Southern’s case, this improvement actually came with an increase in cars. 
   STB has drawn no sweeping conclusions from data submitted but shippers are able to see that the railroads are paying close attention to their results. While the STB can’t penalize poor rail performance as the Canadian government can, this is one step in improving the rail issues that have plagued the U.S. for 2014 and threaten to continue far into 2015, the agency noted. 
   In its decision announcing the status reports, the board laid out its argument that these reports will give everyone a more complete picture of the nation’s rail system.
   “The United States rail system is an interconnected network, and one carrier’s service problems can affect the performance of other carriers.  Although the severity differs, shippers have reported problems on multiple carriers.  Thus, the board views the network as a whole, and seeks to better understand performance across the entire network,” the STB said. “The new reporting requirements will give the agency and stakeholders access to data needed for real-time understanding of regional and national service issues.”

This article was published in the January 2015 issue of American Shipper.