America Could Stop More Than a Third of Train Crashes. Here’s Why It Doesn’t.
From Japan to Mozambique, many of the world’s rail systems have ‘positive train control’—except the United States.
PORTLAND, Oregon—Penn Central Commuter Train N-48 was en route to New Canaan, Connecticut, with a three-man crew and 60 to 80 passengers aboard, when disaster struck. The train’s engineer, Frank J. Bojarski, inexplicably ignored an order that should have prompted him to stop the train at the crossing and yield to another train, N-49, headed in the opposite direction.
“He just kept coming at me,” N-49’s engineer, Edward May, would later tell Hearst Newspapers.
Bojarski and a passenger were killed, along with N-49’s conductor and flagman. Forty others were injured. It took five hours to cut May out of the wreckage.
A year later, the National Transportation Safety Board released its findings about the crash. The engineer had ignored an order to pull onto a side track and let May pass, NTSB wrote. (A copy of that order was found on Bojarski’s body.) Among NTSB’s recommendations aimed at preventing another tragedy were to “study the feasibility” of a form of “automatic train control.”
That was on Oct. 23, 1970.
Nearly a half-century later, what’s now known as Positive Train Control has yet to be activated on most passenger rail lines across America. If an engineer doesn’t slow down at an upcoming crossing, curve, or obstacle, PTC kicks in and stops the train. Some 40 percent of all rail accidents are caused by improperly lined switches or human error; by an engineer distracted or literally asleep at the switch.
PTC is on most rail lines across Europe, Japan, and even in large swaths of relatively poor countries like Mozambique and India. But in America, only patches of the rail network have track control. All those countries subsidize safety for passenger rail, which doesn’t make as much money as its lucrative brother, freight rail. The United States’ subsidies for passenger rail are minuscule by comparison.
The latest avoidable tragedy came a few days before Christmas, in DuPont, Washington, when an engineer somehow missed admonitions that he slow Amtrak 501 train from 80 mph to 30 mph as it reached a curve ahead of a bridge across Interstate 5. The 14-car train flew off the tracks, throwing passengers from busted-out windows, killing three people and injuring dozens of others.
NTSB immediately dispatched a team that will within some months discern the cause of the crash. Was the engineer distracted? Asleep? Did his equipment fail? But it doesn’t take a federal investigation for every expert who has studied the crash to agree on one thing: Positive Track Control would have prevented this derailment, and the unnecessary deaths of Benjamin Gran, Jim Hamre, and Zack Willhoite. Positive Track Control would have prevented 145 train crashes between 1969 and 2015; 298 deaths, and nearly 7,000 injuries, according to NTSB. And yet, the implementation of PTC limps along in the U.S., with lawmakers and government officials and rail companies all blaming one another for the delays.
NTSB’s 1970 recommendation was largely ignored, thanks in part to a defiant response from the Federal Railroad Administration, which called the frequency of accidents due to incapacitated engineers’ “so very small” that the installation of an expensive system wouldn’t be worth the investment. Automatic Train Control had been installed on sections of the Northeast Corridor rail line since the 1930s, but it enforced the speeds at signals not at curves. In Europe and Japan, governments began deploying similar technologies beginning in the 1950s.
Years after NTSB called for it, in 1987, Burlington Northern deployed a system of 17 locomotives on 250 miles of tracks in northern Minnesota. The system worked fine, Steven Ditmeyer, the railroad’s former head of research and development, told The Daily Beast, and NTSB added track control to its “most wanted” list of transportation safety improvements in 1990.
The trade group ordered the study destroyed, Ditmeyer says. Burlington Northern’s train control-enamored chairman retired in 1991, and his successor shut down the project, Ditmeyer said. “PTC went into hiatus for the better part of the decade,” he added.
In 1993, the Association of American Railroads put together a study laying out the business case for PTC: Railroads would save fuel and labor costs, traffic control would work better, and the system could monitor the health of locomotives.
The rail industry has since spent tens of millions of dollars lobbying Congress to keep the government from mandating the installation of PTC, all the while complaining publicly that the government doesn’t provide it enough capital for infrastructure improvements.
It wasn’t until 2007, after an increase in preventable rail crashes and after Democrats took over Congress, that lawmakers finally found the political stomach to mandate the installation of PTC, but only on the highest-risk rail lines: passenger and commuter railroads, and those lines that carried more than 5 million gross tons of hazardous materials. The initial deadline was 2018. A 2008 crash in California inspired lawmakers to push that up to 2015. But the railroads successfully lobbied for a three-year extension, back to 2018, with an option for two more years to get the job done, if railroads met certain stipulations.
Ironically, PTC was installed on the rail line of last month’s Amtrak crash in Washington State. It just hadn’t been activated yet.
“This was preventable, the technology was available, and it was installed,” Ditmeyer said. “That makes me angry.”
While Amtrak owned the train, a local agency called Sound Transit owned the tracks, which it purchased from freight operator Burlington Northern Santa Fe. Amtrak’s locomotive was equipped with PTC, and so was the track, but the system hadn’t been switched on. That’s because the line needed testing, Sound Transit spokesman Geoff Patrick told The Daily Beast.
Sound Transit is much farther along installing PTC than most of its national counterparts. The majority of the commuter line’s routes run on BNSF tracks, and 56 percent of the companies trips all have PTC installed and operating. “We’re ahead of two-thirds of commuter rail systems in the country,” Patrick says.
American railroads are a complicated web of ownership scattered amongst freight companies, and local and regional commuter lines. Of 200,000 miles of railroad in the U.S., some 97 percent carries freight, not passengers, Dr. Allan Zarembski, direct of the Railroad Engineering and Safety Program at the University of Delaware, told The Daily Beast.
The freight industry can afford to make infrastructure improvements to its lines because it makes more money on its cargo, Zarembski says. But even the freight industry has made “foolish” decisions that have delayed the rollout of PTC, according to Ditmeyer. The business case for PTC was clear: It wouldn’t just improve safety, it would help railroads run trains more efficiently, closer together, allowing the companies to increase track capacity and use their assets smartly.
But the railroads’ method for installing PTC was to remove existing systems, install brand new ones, and then add track control on top of that, a move that “doubled the cost of PTC and virtually eliminating the possibility of reducing running time,” Ditmeyer says. “They knocked out the business benefits and doubled the cost,” he said, from $5 billion to $10 billion.
Despite that, most of the freight system has finally had PTC installed on its tracks, but passenger rail waits on government handouts for infrastructure upgrades.
Congress has consistently hacked away at federal budget allocations that could have paid for more PTC. In 2016 alone, House Republicans cut Amtrak’s capital grants by $290 million.
The federal government spends more each year on freeway construction than Amtrak has received for capital upgrades in its entire 47-year history.
“There’s no passenger, commuter, or transit system in the world that makes money,” U.S. Rep. Peter DeFazio (D-OR) told The Daily Beast. “They’re always in fiscal distress. They’ve got to choose between keeping the trains running, maintenance, contracting, or new equipment.”
Despite Congress’ mostly unfunded mandate, some tried to meet the original 2015 deadline for PTC, says Zarembski. That’s when they ran into another reason PTC isn’t installed on more rail lines across America: the Federal Communications Commission.
For PTC to work requires an allocation of radio “spectrum,” which is sort of like bandwidth, by the FCC. Freight railroads started buying up all the spectrum they could get their hands on years ago, but the feds have to approve those purchases, and telecommunications companies have fought any allocation outside of those allotted to cellphone and internet bandwidth.
The Association of American Railroads has complained to Congress that there’s a backlog of submissions to the FCC for a review of wayside antennas that have delayed freight rail lines’ ability to implement PTC by 15 months, according to a report from the House Committee on Transportation and Infrastructure.
“The phone companies want to gobble up all the spectrum in the world, so you can have 10 gigabyte Twitters,” DeFazio, the committee’s ranking member, says. “They’re trying to block us from having spectrum for smart cars, and railroad communications, too.”
NTSB could be weeks or months away from completing its investigation into the Washington State derailment, but the government’s 50-year failure is already clear. This disaster will certainly ramp up pressure on the government and rail companies to make trains an even safer mode of transit than they already are.