Home View from Underground Stringer: MTA failing to meet demand for off-peak subway service

Stringer: MTA failing to meet demand for off-peak subway service

by Benjamin Kabak

Once upon a time, before the 2010 cuts decimated transit service in NYC, the MTA worked to ensure that off-peak subway service met increasing demands. From 2000 until 2010, as off-peak demand, especially between 5 a.m.-7 a.m., increased, the MTA responded in turn, adding more service so that the percentage of off-peak trains generally aligned with the percentage of riders using the system during these early hours. But since 2010, as New York City’s economy has rebounded from the depths of the recession and off-peak commuting has increased, the MTA hasn’t added service at a commensurate level, and this service gap is hurting New York’s lower-income workers, according to a new report by NYC Comptroller Scott Stringer.

“During rush hour, we’re packed into subway cars like sardines. But outside traditional ‘9-to-5’ travel, off-peak service is fundamentally failing to meet demand. It’s a crisis within a crisis, because over the past decade, the nature of our economy has changed,” Stringer said in a statement, “and ridership late-nights and early mornings has risen while actual service to match it has not. That means that those who need service to get to work during non-traditional hours are stuck with a crisis not of overcrowding, but of infrequency.”

Subway service to and from the Manhattan hub has not kept pace with growing ridership in the early morning (5 a.m. to 7 a.m.)

The crux of Stringer’s argument is the chart above. Based on the nature of New York City’s economic recovery, as more jobs have shifted outside of the 9-to-5 realm, the MTA has not added service to meet demand. Thus, as Stringer put it, more riders are waiting longer for more crowded trains in the morning even as the MTA has engaged in an on-again, off-again PR campaign to ask riders to shift travel outside of the peak hours. Furthermore, riders who need to commute during these off-hours are generally earning around 20% less than those who travel for traditional peak-hour jobs. They can ill-afford sub-par transit service, but that’s what they’re getting.

The numbers are stark: Since 2010, ridership to and from Manhattan has jumped by 14% in the two hours before 7 a.m. and by 13 percent in the four hours after 7 p.m. Meanwhile, over the same period of time, train service declined by 3 percent between 5 a.m. and 7 a.m. and increased by only 3 percent between 7 p.m. and 11 p.m. Thus, anecdotes of 10:30 p.m. trains packed to crush load are borne out by numbers that show service failing to keep pace with ridership trends. As Stringer notes, these unreliable commutes can lead to reprimands at work, docked pay, termination and a general slow-down in productivity. His office had previously estimated that subway delays cost around $400 million annually in lost productivity and wages.

Subway service in and out of the Manhattan hub did not keep pace with growing ridership in the early morning and evening from 2010 to 2016

Stringer’s report has some interesting trend information concerning commutes. As ridership has exploded over the past 30 years, commuting patterns have undergone a dramatic shift. In 1985, a whopping 49 percent of daily ridership into the Manhattan central business district occurred between 7 a.m. and 9 a.m. Now that number is only around 28 percent. In essence, all of the people riding in 1985 between 7 a.m. and 9 a.m. are still doing so (and then some), but off-hour commuting which was non-existent to a combination of safety concerns and lack of jobs has exploded.

Meanwhile, people are waiting much longer. Stringer’s report notes as follows:

While the MTA may be restricted from increasing frequencies in peak hours due to aging infrastructure and underinvestment, these capacity constraints are less relevant in the early morning and late evening. In fact, the MTA runs 60 percent fewer trains from 5 a.m. to 6 a.m. than it does from 8 a.m. to 9 a.m. and 38 percent fewer from 9 p.m. to 10 p.m. Clearly, there is ample train and track capacity at these hours.

This dramatic decline in off-peak subway service, of course, leads to significantly longer wait times for riders. During the morning rush hour, for instance, 87 percent of subway lines run more than six trains per hour and 36 percent run more than 12. That means that passengers will rarely wait more than five or ten minutes for a train—provided all goes well.

By contrast, in the evening (8:30 p.m. to 10:30 p.m.) and early morning (5:30 a.m. to 6:30 a.m.)—when a growing number of workers depend on public transit—subway service is significantly diminished. Just 47 percent of subway lines maintain headways of less than 10 minutes in the evening and only 10 percent maintain frequencies of less than 5 minutes. In the early morning, meanwhile, only 43 percent have headways of less than 10 minutes and zero have headways of less than five minutes. Looking at individual lines, the biggest drop-offs in the early morning are on the 1, 6, 7, B, and W trains, where hourly throughput declines by more than 50 percent in the peak direction, as compared to the A.M. rush. In the evening hours, the 4, 5, A, B, and F lines experience a similar fall in service.

This is, of course, not a surprise to those who have tried to ride outside of peak hours. Waits are noticeably longer, and trains tend to be nearly just as crowded off-peak on some lines as during the peak due to these increased headways. As I mentioned, it’s particularly jarring in the context of numerous MTA officials urging those with flexibility to commute at off-hours to do so. Percentagewise, at least, a significant number of commuters have moved to these so-called off-peak timeslots, but the MTA hasn’t provided a carrot in the form of increased service to go with the stick.

Where I find Stringer’s report lacking however is in its messaging. The recommendations are on point; the MTA, he says, should perform a comprehensive review of scheduling and strongly consider how latent or induced demand can further drive up off-peak commuting. But in releasing the report to the public, he also tried to use it as a crudgel to beat on the drum of city support for the MTA’s subway action plan. He says city support of the subway action plan will ensure the trains run on time, but he could use this report to call for the subway action plan to directly add more off-peak service. It would be a better use of the money anyway than the SAP band aid that many think won’t actually fix anyone’s commute.

Ultimately, though, the overall message is one worth remembering: Yes, the MTA needs a new signal system and a comprehensive overhaul of its procurement and construction processes. By as with the signal timers, commutes are worse than they should be because of choices made at the MTA. The agency is not a bystander in the accelerated slow-motion collapse of the city’s transit system.

You may also like


Stephen Bauman March 26, 2018 - 10:11 am

In 1985, a whopping 49 percent of daily ridership into the Manhattan central business district occurred between 7 a.m. and 9 a.m. Now that number is only around 28 percent. In essence, all of the people riding in 1985 between 7 a.m. and 9 a.m. are still doing so (and then some), but off-hour commuting which was non-existent to a combination of safety concerns and lack of jobs has exploded.

That’s not true. There’s been a shift away from the peak hour and peak period ridership in both realand relative terms. If one takes the trouble to look at NYMTC’s Hub Reports, one would find that the number of peak hour (8-9am) inbound CBD riders fell from 461K in 1985 to 373K in 2016. Similarly the number of peak period (7-10am) inbound CBD riders fell from 912K to 851K during the same period. Finally, the 24 hour inbound CBD totals rose from 1.541M to 2.171M. The peak hour and peak period numbers pale in comparison to the 1963 totals: 583K (8-9am); 1.118M (7-10am). However, the 1963 24 hour total was only 1.926M (less than 2016).

the MTA needs a new signal system

The MTA and its apologists claim the MTA signal system is what’s preventing the MTA from running more trains during peak periods.

Were that true, how did the Transit Authority manage to transport so many more people during the peak periods? The answer is that they ran many more trains during the peak hour than the do today. They managed to do this with the same signal system that they want to replace.

Here’s what the Transit Authority managed to run in 1954. The Board of Transportation reported similar results in 1949, and here.

One should note that the Canarsie Line operated a peak of 24 tph, whereas the CBTC total is only 20 tph (to be improved to 22 tph electrical power permitting). The Flushing Line operated 36 tph in 1954 vs. 28 tph today. The Third Avenue El operated 42 tph back in 1949 before its operation was cut back.

The point is that the existing block paradigm is capable of operating at far greater service levels. What’s needed is an upgraded implementation that does not rely on “vital relays”. The second point is that the high service levels of yore are not required due to today’s reduced peak hour demand. That’s why peak hour travel remains tolerable, with reduced service levels. The MTA has been responsive to reducing service levels, when it is not required. It has been unresponsive to increasing service levels, when it is required. They are mired in the early 20th Century mindset that only peak hour performance matters.

Benjamin Kabak March 26, 2018 - 10:31 am

I’m seeing different numbers than you in the hub-bound travel reports. The 1985 report showed 908K between 7-10 a.m. on the subway vs. 921K in 2016. The overall point about off-peak service remains, but there seems to be some discrepancy regarding peak volume.

Stephen Bauman March 26, 2018 - 3:14 pm

I’m reading from Table 14 on page 22 for the 1985 Report. The 7-10 am numbers are:
Manhattan: 298580
Brooklyn: 375190
Queens: 238620
Tot 1985: 912390

I’m also reading from Table 14B on page II-2 (44) from the 2016 Report. The 7-10 am numbers are:
All Express: 218570
All Locals: 702163
NJ PATH: -70043
Tot 2016: 850690

AndrewC999 April 6, 2018 - 5:10 pm

” the existing block paradigm is capable of operating at far greater service levels”

I am far from an expert, but could it be there are other factors, such as longer trains now, stricter safety regulations due to accidents in the ensuing decades, other equipment differences etc. at work here? Do the individual trains run at higher speeds now, requiring more stopping distance?

In my experience, when it seems so obvious things like this should be easily fixable and they haven’t been, things are not usually as simple as they appear. It could be a lack of willpower/corruption etc., or there could be legitimate reasons.


AndrewC999 April 6, 2018 - 5:13 pm

Interesting: I just read this in yesterday’s post, which seems to answer my question at least in part:

But in the aftermath of the crash, the MTA determined that many of the then-septuagenarian signals were too close together to adequate stop modern trains faster than their late-1910s counterparts, and thus, the MTA in conjunction with Parsons Brinckerhoff began to implement speed limits throughout the subway system


Larry Littlefield March 28, 2018 - 7:41 pm

So what is the MTA doing with all the money from cutting service and packing in riders? Burning $100 bill in a furnace for heat?

Why don’t Stringer — and DiNapoli and Lhota — ask that question. Could it have something to do with their political supporters? Could it have something to do with deals that were cut one and two decades ago to provide a score for some people in the present (now the past) at the expense of everybody still around in the future (now the present)?

Where is the money going?


Leave a Comment