A newsletter about what the hell is going on with the subway.

What to make of fare evasion?

Thank you so much to everyone who became a paid subscriber! I’m excited to put together some fantastic special editions for you.

If you missed yesterday’s announcement, you can read it here. And my apologies to everyone who had trouble signing up yesterday. Substack, the newsletter platform, had some ill-timed technical difficulties. But everything is back to normal.

Subscribe now

On Monday, the MTA Finance Committee discussed a long-awaited report on fare evasion. The “report” was really just an eight-slide PowerPoint woefully short on details and methodology breakdowns. When I asked the MTA if there is a more detailed, full report and when it will be released, a spokesman replied, “This is the report.”

According to the MTA, there are around 550,000 fare evaders on the subway and buses every single day. The MTA further estimates $215 million in “lost” revenue due to fare evasion this year, almost double the amount “lost” due to fare evasion in 2015. Here is the graph they used to illustrate the problem (chart nerds, avert your gaze):

I’ve put the term “lost” in quotes because I have absolutely no idea how NYCT came up with this estimate. The only clue they give in the presentation is this bullet point:

Evasion rates are calculated against paid ridership figures.

I don’t know what assumptions went into calculating those revenue figures or the daily evader projections. And those assumptions are incredibly important when extrapolating from observation-based samples. We know they counted people who didn’t swipe at various subway stations and bus routes, and that ultimately their computers churned out these estimates and figures. But, we don’t know, for example, if they’ve changed the method, location, or sample size of their observations over the five-year period, as they have with many other statistics. Everything in between is a total black box.

So, we’re left to evaluate these numbers as best we can. The MTA is saying that one out of every 25 subway riders are fare evaders, and about one out of every five bus riders doesn’t swipe. As Chase Tralka on Twitter pointed out, this averages out to one fare evader at every subway station every three minutes.

Without knowing more about the methodology, its possible there is some serious over-counting as a result of the MTA’s complicated fare structure, poorly-designed control areas, and service change policies. At its core, I worry this analysis conflates two distinct problems: people not paying for their ride versus the MTA not accurately counting riders.

For example, the presentation made the point that fare evasion through the subway service gate has increased four-fold since 2011, which accounts for the “vast majority” of the increase in fare evasion on subways. But it is quite the leap to assume everyone who enters the subway through the service gate is not paying their fair share. According to the most recent MetroCard report to the board (page 191), more than half of all trips are taken on unlimited MetroCards. Surely, many people entering through the service gate are doing so with unlimited cards in their possession simply because it’s easier or less crowded. Or perhaps they have bulky items and don’t feel like going through the cumbersome process of alerting the station attendant, swiping the card, backing up, and going through the service gate. Maybe there’s no station attendant at all.

And this is to say nothing of local buses, where NYCT estimates as much as one out of every five riders does not scan their card (evasion is lower on SBS, but still higher than the subway). Fare evasion is a much bigger problem on buses for sure—there is almost no enforcement and drivers are instructed not to intervene—but I do wonder, again, how many people simply don’t bother to swipe because they’re on a free transfer, or are waved through by a driver trying to speed things up.

In any event, this potential for over-counting and total lack of transparency on the report’s methodology would be bad enough if the presentation didn’t try to make the exact opposite point. In fact, NYCT argues, they were likely under-counting the problem because of such factors as “the inherent limits of human observers in a dynamic environment” and “the mere presence of field staff may limit evasion.”

These last notes concern me a great deal. If NYCT was truly interested in studying fare evasion, they would take basic measures to minimize these factors, such as dispatching plain-clothed employees. This, to me, raises a big red flag. Either they’re working backwards from a desired conclusion or not thinking through their methodology very carefully.

That’s not to say I think fare evasion is a non-problem. But this eight-page PowerPoint (OK yes the PDF is nine pages but no way am I counting the title slide) does not constitute transparency on the issue. Instead, I would characterize it more as an argument. And it’s not a particularly useful one, particularly given all the other challenges the MTA faces.

Even if everything the presentation says is accurate, it’s not even clear what should be done about it. As Byford has repeatedly stated, fare evasion is a fact of life on any transit system. At least as the subways are concerned, even these higher figures are not that far out of line with what the NYCT presentation dubs “peer metros” (which metros are they and what methodology do they use? Good question! The report doesn’t say).

But let’s grant the premise that fare evasion is a huge problem. Then what? Post cops at every turnstile? 89 percent of those arrested this year for fare evasion are black or Hispanic. In any formulation about how to move forward with fare evasion policing, this is a fact that simply cannot be brushed aside. What does equitable fare enforcement look like? How resource-intensive is it? Until there is a clear plan on this front, the MTA runs the risk of further encouraging discriminatory enforcement practices.

But the main reason this issue rings hollow to me has to do with why the report was even generated in the first place. The MTA board asked for it.

Indeed, the board’s disproportionate focus on fare evasion these last few months speaks volumes about their priorities. To me, it’s indicative of the MTA board not knowing its purpose. Does the MTA exist to be run as a self-sustaining entity? Or is it a public good funded primarily by tax revenue? And what is the board’s actual role in figuring that out?

Answering these questions would go a long way towards making the fare evasion question clearer. If the MTA ought to be self-sustaining, then it would have a rigorous cost-benefit calculus as to the optimal amount of fare evasion—which, by the way, is not zero, as that would require far too much enforcement—and how to best achieve it. On the other hand, if the MTA is delivering a public good, then very little energy should go to enforcing fares through draconian policing. Insofar as it would be interested in curbing fare evasion, the MTA would in this case be focused on design changes to the fare control areas. For example, turnstiles could be re-designed to prevent hopping—it almost seems to me like our turnstiles were specifically designed to facilitate hopping, unlike other fare control areas— and control doors closed again.

Further, the astronomical rate of fare evasion on buses would suggest it’s time to have a discussion about whether buses should have a fare at all. (There are a lot of other elements in favor of free buses, including social justice, income distribution, and traffic reduction/environmental arguments. That’s for another time.) Or, perhaps the MTA would simply need to adjust baseline assumptions for how much revenue is “lost” to fare evasion every year and work within those confines.

A note to Signal Problems readers

Almost one year ago, I created Signal Problems to help keep New Yorkers informed about what the hell is going on with the subway. I’m thrilled by the daily notes from readers telling me how much they appreciate the newsletter. I want to keep Signal Problems going. But I need your help.

Starting today, you can support Signal Problems by signing up for a paid subscription. For $5/month or $50/year, paid subscribers will:

  • Receive at least 25 special editions of Signal Problems per year. That’s 50 percent more Signal Problems (the newsletter, not the subway delay), drilling deeper into what ails the subway and how to fix it. Some special editions will cover current events, similar to the Amazon HQ2 Special Edition. But others might be Q&As with interesting people, investigative projects, etc.

  • Help keep Signal Problems ad-free. I’m not a salesman. Paid members make sure I spend my time doing what I do best: reporting and writing.

  • Be able to comment on all posts.

The Friday editions of Signal Problems will remain free. My hope is if you enjoy those Friday editions, for about the cost of a fancy coffee per month, you’ll consider supporting the newsletter.

Writing Signal Problems, interacting with readers, and sharing your Dog In A Bag photos has been the most fulfilling undertaking of my career. I don’t merely want to keep it going. I want to make it better. But I need your help to do that.

Subscribe now

Thanks for reading as always,

Aaron Gordon, transportation reporter and founder of Signal Problems

The 7 Train, CBTC, and the trouble with expectations

The 7 train finally cut over to CBTC service! Hooray!

NYCT Subway@NYCTSubway34 St-bound 7 trains are running with delays because of network connectivity issues. We are doing everything we can to keep trains moving and get you to your destination quickly.


NYCT Subway@NYCTSubway34 St-bound 7 trains are running express from Willets Point to 74 St as our signal vendor troubleshoots a connectivity issue with a train at 111 St. We are moving a train out of service now and other trains will be moving again shortly.


NYCT Subway@NYCTSubway7 trains are running with delays in both directions because of a network communication problem at Queensboro Plaza. There is no 7 express service.

OK, so the first week of the CBTC on all of the 7 line—which has been more than a decade in the making, five years behind schedule, and at least $157 million over budget—hasn’t gone well. The 7 has been hit with delays or service changes each day during rush hour. Via SubwayStats.com, here is how the 7 train looked during the first two days after the cut over:

Not all of those service disruptions are due to the cutover—someone was struck by a train Tuesday night for example—but most have been a number of “network” or “connectivity” issues affecting the signals.

So what happened here?

First, if you’re new to Signal Problems or need a refresher course on CBTC, here’s a primer I wrote for the Village Voice.

Now, what’s this about the “cutover?” Over the last year or so, CBTC has been gradually rolling out in segments across the 7 line, moving from east to west. When CBTC is ready for a new segment, they call it cutting over. Before last week, CBTC was the signaling system until just before Grand Central. The final cutover this week completed the entire line, which seemed like something to celebrate, right?

Well, not so fast. For those paying attention, previous cutovers had their issues, too. For example:

NYCT Subway@NYCTSubway@chichaaang Hi Joyce. Supervision advises that 7 trains have experienced delays recently between Mets-Willets Point & Main St. related to the introduction of our Communications Based Train Control (CBTC) system. We're working to address this issue. ^BD
NYCT Subway@NYCTSubway@dubious_d Hello -- supervision advises that 7 trains have experienced delays recently between Mets-Willets Point & Main St. related to the introduction of our Communications Based Train Control (CBTC) system. We are working to address this issue. ^BD

Furthermore, it might sound like the CBTC project ends with the final cutover, but it doesn’t. As Robert Gomez, a program officer in Capital Program Management told the MTA board earlier this month, there are still weeks of fine-tuning that take place afterwards. Right now, CBTC is the 7’s signaling system, but trains are not running in full automatic mode, called Automatic Train Operation (ATO), in which the motorman presses a button and the train goes until it stops. Instead, they’re operating in what’s called “restricted manual” mode, which is a kind of hybrid between human-operated and ATO, in that it tells train operators when they can proceed and at what speed.

So, even though the CBTC cutover has occurred, the most important benefits of CBTC are yet to kick in. A lot of focus on CBTC is how many more trains the MTA can run along the line, which is indeed important, but that’s only a piece of the CBTC puzzle. The reliability improvements are just as important. Those improvements come via two main features: first, more stable hardware so the signals rarely malfunction, and second, ATO, so the trains always strike the right balance between speed, safety, and spacing between trains.

Right now, the 7 is not getting those benefits because the project isn’t finished yet. Contractors are still tinkering with the software to prep for ATO, and until that is done, CBTC on the 7 has not been, to use the MTA’s technical terms, “substantially completed.”

My initial inclination was to blame NYCT—naturally—for misleading customers about the nature of the cutover and saying the project is done when it’s not. After all, we all know NYCT has issues communicating with customers. But I actually don’t think this is one of those cases. In fact, I think they’ve been pretty clear that there’s more work to be done after the cutover, both in board meetings and in communicating with the public. See, for example, their tweet from Monday announcing the cutover:

NYCT Subway@NYCTSubwayToday is the first day that modern signaling tech is live on the entire 7 line. After we finish optimizing and automating the system and performing other signal work over the coming weeks, we’ll be able to run more trains, more reliably. Thank you for your patience.

I then wondered if it maybe would have been better for them not to say anything about the cutover at all since it wouldn’t substantially change the customer experience. But it’s my opinion the MTA should always err on the side of transparency. It is, after all, a public authority. In fact, I can’t really think of a good argument against the MTA announcing the cutover as they did. So, while we’d all prefer to live in a world where the network problems didn’t happen this week, I have a hard time faulting NYCT or the MTA for anything other than the signal problems themselves.

Instead, I think 7 riders, who have put up with almost five years of service disruptions, were understandably anxious to finally start experiencing the benefits of that sacrifice. And they will, very soon. Unfortunately, that angst led to misunderstandings, and now there’s even more resentment and frustration, which tends to happen among all parties towards the tail end of very delayed and troubled projects regardless of what is said or done.

This causes me to worry. Not that the benefits won’t come, because they will. But the project has been so delayed for so long and caused so much inconvenience while the rhetoric around upgrading the signals have been so heightened to the point where it’s all too often presented as some magical panacea for all that ails the subway, no signal upgrade in the world could possibly deliver the outsized benefits riders now anticipate. It will be better. But will it be good enough?

News You Probably Can't Use, But About Which You Can Certainly Brood


In recognition of Thanksgiving—and the short holiday week that generated very little in the way of subway news)—I’ve decided to do a special holiday edition of Signal Problems featuring something I am very, very thankful for: dogs in bags. So let’s celebrate and give thanks to the one thing that makes any subway ride a good ride.

MTA Rules of Conduct Section 1050.9 Subsection (h) Paragraph 2: no person may bring any animal on or into any conveyance or facility unless enclosed in a container and carried in a manner which would not annoy other passengers.

Have a dog in a bag photo? Reading this on the subway and see a dog in a bag? Take a picture and send it to signalproblems@substack.com.

Photo credit: Jordan Bleckner

Photo credit: Mike Zappitello

Photo credit: Adam Reynolds

Photo credit: Anonymous

Photo credit: Emily

Photo credit: Ezra Kurtz

Photo credit: Rosemary McGuinness

Photo credit: Vivek Bhagwat

Photo credit: Matthew Bambach

Photo credit: Fritz Brantley

Photo credit: Michael Pollack

Photo credit: Seth Rosenthal

Photo credit: Alana Casanova-Burgess

And last, but certainly, certainly, not least:

Photo credit: Leonor M.

This has been another edition of Signal Problems, a weekly newsletter helping you figure out what is going on with the subway, made every week by Aaron Gordon, freelance transportation reporter. Read on the web or view the archives at signalproblems.nyc.

If you’re enjoying this newsletter, please share it with others. It’s the best way you can say thanks.

As always, send any feedback, subway questions, or Dog in a Bag photos to signalproblems@substack.com. I’d love to hear from you. As someone on a stalled Q train once told me, we’re all in this together.  

Stories we tell

Every budget tells a story. Sometimes, the stories are difficult to read and have many competing interpretations. Other times, the authors come right out and tell you what the moral of the story is. That doesn’t make it the only moral, but at least their intentions are clear.

During a presentation to the MTA board on Thursday, CFO Bob Foran took the latter approach. He made it clear that the agency’s cost-cutting goals for recurring expenses have pretty much run their course. They’re having a hard time finding additional savings. “The low hanging fruit,” he lamented, “is gone.”

Coupled with lower-than-projected revenue over the coming years, the MTA is in dire financial straits. They need more money.

The headline item, of course, was the proposed fare increases, with the flotation of service cuts on the horizon getting second billing. To be clear, the board did not vote on anything except to begin the process to raise fares. It’s not at all certain what any new fare structure will look like—although the MTA did put forward two options—or whether the board would even vote for it.

I haven’t had a chance to comb through the 600+ page financial report released on Thursday. In fact, some board members were so irked that they would not have another chance to discuss this material before voting on the proposed budget in December that they agreed to hold another meeting in the next few weeks. So there’s plenty more information and debate to come.

That being said, this budget’s story is not about what the MTA is telling riders. It’s about what the MTA is telling its funders, most importantly the lawmakers in Albany.

How do I know this? The titles of each slide in the presentation tell a pretty clear story. I will paraphrase that story:

We need more money. We already knew we’d have very little money, even if we make riders give us more money. We don’t want to make riders give us more money, but we have to, or else we’ll have even less money. This isn’t our fault. We’ve tried everything, but we’re running out of ways to save money. We could provide a worse product, but even if we wanted to, it would barely save us any money. In sum, please give us more money.

It’s a good story! But like many stories, the authors are writing a clean narrative for the sake of clarity and brevity. In this way, the full 600+ page report is akin to the reporter’s notes, which provide all those messy details and reveal how the story could have been told differently, without even getting into the issue that this is absent capital program expenditures, a whole other budget with its own spending problems.

Since July, the MTA lowered projected revenue by $485 million over the next four years—the biggest adjustment in the new plan—due to lower ridership. As the MTA’s presentation from July made clear, there are many reasons for lower ridership. The board preferred to focus on fare evasion, even though at this point the MTA has presented no evidence whatsoever to suggest fare evasion is a bigger problem now than it was two or four or ten years ago (the MTA promises a robust presentation on fare evasion next month, which will hopefully clarify this issue).

That being said, there is plenty of evidence to suggest poor service, particularly on nights and weekends, is the biggest driver for declining ridership. Weekday ridership impacted by construction nearly doubled from 2017 to 2018. Similarly, weekend ridership is down 7.3 percent year over year. Why? Because weekend work has doubled over that time. If a large portion of the subway system is not running, a large portion of potential riders will not use it. Many things about the subway system are complicated, but this is not one of them.

Further, the Subway Action Plan has been a key driver of those service changes. Notwithstanding the questionable benefits it had for the system, Byford admitted that the agency hasn’t fully analyzed the revenue impact of the Subway Action Plan in terms of lost ridership, conceding “I think it’s probably not as advanced as it needs to be.” But he rightly added any such analysis would be tricky and imprecise.

Still, this general issue of how planned work affects ridership—and therefore revenue—isn’t going anywhere soon. The L shutdown, for example, begins next year. The MTA predicts the vast majority of trips will still take place within its ecosystem, but it’s easy to imagine ridership falling due to discretionary trips not being taken or a higher-than-projected rate of folks opting for rideshare or bicycling instead. Indeed, the MTA now predicts a 1.1 percent decrease in ridership in 2019, following a 2.8 percent decline this year. This is a major revision from the July plan, where they predicted ridership *increases* in 2019 and 2020 despite acknowledging the L shutdown. Their logic: the economy is good.

These explanations are more confusing than insightful. Pegging ridership trends to future employment projections may be accepted practice but it’s been demonstrably unreliable in recent years due to fundamental changes in how we work, shop, and travel. I, for one, anticipate the July 2019 plan to further revise ridership downward due to the impact of the L shutdown as people work from home or a local coffee shop more often, don’t travel to Williamsburg on weekends, and ditch a monthly Metrocard as a result.

But there’s an even bigger red flag in their ridership projections. If the MTA does get funding to move ahead with the Byford Plan, entire trunk lines in Manhattan as well as major branches in Queens and Brooklyn will be shut down on nights/weekends for months if not years on end. In other words, the most extreme planned work shutdowns in the city’s history will occur in the next decade if Andy Byford gets his money. Ridership will almost certainly suffer.

That’s not an argument against doing the work, but merely a consideration therein, especially when projecting budgets. But, as of now, the MTA is predicting flat ridership for 2020-2022. Of course, the MTA cannot budget for a plan that has yet to be funded, but they don’t even flag this as a potential risk. This is emblematic of the agency’s tendency to get caught flat-footed by predictable ridership trends.

Still Foran cautioned against reading too much into the 2022-2024 projections, otherwise known as the “out years,” because predictions that far out are less reliable. This could also serve as yet another suggestion that perhaps the service cut warnings, which wouldn’t be applied until 2020, are more of a threat to politicians in Albany than an actual, serious proposal. (For what it’s worth, my early prediction is the board will vote for fare increases but won’t entertain service cuts, at least not directly.)

Acting MTA chairman Freddy Ferrer took a more existential approach to even some of the near-term predictions. “Bear in mind,” Ferrer cautioned, “we haven’t shut the L line down yet. So a lot of this is an estimate. What we may lose in subway ridership we will undoubtedly pick up in bus ridership.” (This is very much in doubt.) “So will it be person for person? I don’t know, we don’t know, it is not knowable right now.”

What I do know is there’s one moment from the press gaggle that’s sticking with me, because it doesn’t mesh with the story we were told. Among all this talk about dedicated revenue streams, congestion pricing, and all that jazz, AM New York transit reporter Vin Barone asked Ferrer if the agency has truly exhausted all of its options when it comes to cost-cutting. For example, Barone cited, health care takes up 13 percent of the MTA’s budget.

“I’m not even going to get into a debate about health care,” Ferrer replied, because it’s “not my job,” but also because “that’s very important to the men and women who work for the MTA and all of our agencies, for their families, for their children.”

Loading more posts…