It was never all in your head. The New York City subway really is moving slower.
The trains aren’t just stuck in the tunnels due to a broken rail or signal problem or “sick passenger,” although that happens a lot, too. Even when there is supposedly “good service,” even when the trains are moving relatively smoothly, they are going slower than they have 15, 10, even five years ago. And it is mostly on purpose.
Like many well-intentioned but ultimately detrimental programs, the great subway slowdown, which has mostly occurred in the last five years, was originally implemented in the name of safety. And doesn’t everyone want the subway to be as safe as possible? Indeed, two decades ago the Metropolitan Transit Authority (MTA) had good reason to be concerned about safety after two high-profile train crashes in the first half of the 1990s. In 1991, a drunk motorman plowed into Union Square and killed five people. Four years later, a train rear-ended another on the Williamsburg Bridge — likely because the motorman fell asleep, although investigators couldn’t say for sure because he was the only person to die in the crash.
Like many well-intentioned but ultimately detrimental programs, the great subway slowdown, which has mostly occurred in the last five years, was originally implemented in the name of safety.
After the Williamsburg Bridge crash, the MTA knew it had to do something to eliminate the potentially catastrophic effects of human error. Its solution was to slow trains down. The following year, the MTA eliminated the “express” mode on every train, reducing maximum speed from 55 or 60 miles per hour to about 40 miles per hour on a flat grade. At the same time, it began to roll out something called signal timers, a series of devices that automatically trigger the emergency brake if the train is going over the posted speed limit.
Again, this all sounds good, right? After all, nobody wants trains to crash. And for a while, it was good. The subway was safer. But the changes also led to long-term consequences no one had anticipated and which took years to take effect.
Even after the subway became safer, the MTA kept slowing trains down. Since 1996, the MTA installed some 2,000 signal timers across the more than 665 miles of mainline track on the subway system, not just on parts of the track with sharp curves or steep grades, but also on relatively safe portions that didn’t obviously need strict speed restrictions.
Critically, as the MTA installed more and more of these devices, the authority had no program in place to maintain them properly. This is a huge problem because a malfunctioning signal timer is a huge problem. For safety reasons, they are designed to err on the side of caution when not working properly and assume the train is going faster than it actually is. But this means malfunctioning timers will trip the emergency brakes even when a train is going under the posted speed. And train operators get penalized for tripping the emergency brakes due to supposedly violating the speed limit.
Tripping the emergency brake is bad for a number of reasons, most obviously because it takes as much as ten minutes to restart the train after the brakes are deployed, which in turn causes a delay in service. But broken signal timers also pose a much more fundamental ill to the subway system. There is only one way to find out if a signal timer is broken: to travel across it at the posted speed and see if it works. In other words, there was no good way for the MTA to know which timers worked and which ones didn’t.
Now, imagine you’re a train operator. You have perhaps heard from other operators that there are some broken timers out there, or perhaps you’ve tripped a broken one yourself before. But, there’s no way of knowing which of the 2,000 sensors throughout the system aren’t working properly. What are you going to do? You’re almost certainly going to go much slower than the posted speed limit through many of them just to make sure you don’t trip the brakes and get in trouble with your supervisor.
This, in a nutshell, is why the subway got so much slower, particularly in the last five years. After decades of not maintaining the timers, coupled with staff turnover that deprives the MTA system of its institutional knowledge, operators started to treat most every timer like it could be a broken one. This means trains may be going five, ten, 15 miles per hour slower than they ought to. Slower trains means fewer trains coming through the station to pick up and drop off passengers, more crowding on the platforms and more time trains have to spend in the station.
Slower trains means fewer trains coming through the station to pick up and drop off passengers, more crowding on the platforms and more time trains have to spend in the station.
All of this results in delays, which is why the subway delays due to unknown causes increased by a whopping 1,190 percent from January 2012 through December 2017. All the while, MTA management has attributed these delays to “overcrowding,” which is akin to saying your carpet was ruined because your apartment flooded without questioning why there was water on your floor in the first place.
Fortunately, there’s potentially good news. A year ago, the MTA hired a new president for New York City Transit, Andy Byford, who promptly acknowledged the problem was not overcrowding after all. Byford, who previously served as the chief executive officer of the Toronto Transit Commission, created a special unit called the S.P.E.E.D. Unit — short for Subway Performance Evaluation, Education and Development — that rides around the subway in an empty train to make sure the speed limits are as high as they can safely be and to test the signal timers. With almost the entire subway tested, this team has found 267 malfunctioning timers, or about 13 percent of all the ones in the system, which are in the process of being fixed.
As Byford readily acknowledges, the best way to make sure the problem doesn’t repeat in the future is to do away with signal timers and the human element entirely by installing a modern signaling system that uses computerized signals and sophisticated software to control the trains. This technology is already in use on the L train, where train operators merely push a button and the train goes until it stops. This system will soon be operational on the 7, too. Byford has proposed installing these new signals to most of the system by 2030, although it will cost billions of dollars to do so.
But an advanced signaling system won’t address a more fundamental issue, which is the bureaucratic rot that allowed this problem to fester for years as MTA management looked for excuses why subway performance was tanking right in front of their very eyes. Whether the root causes of that institutional failure can be fixed is still a matter of debate. But at the very least you aren’t imagining things on your morning commute.
To learn more about Aaron Gordon's work documenting the problems with the MTA, listen to his recent conversation with Chris Hayes on this episode of "Why Is This Happening?"