City Blueprint: The Second Avenue Subway (Part 1: Fixing The Money Pit)

86th Street Station, in all its hugeness.

It was like a dream come true to many when, on New Year’s Day, 2017, the first section of subway under Second Avenue finally opened to paying customers. As you can imagine, the enthusiasm wasn’t hidden: Blocks-long lines, filled to the brim with eager Upper East Siders, stretched from subway entrances, trains on the line quickly became packed (both with curious straphangers and VERY excited train buffs), and the station crowds soon resembled a calmer version of rush hour at Times Square. In short, it was a day of tremendous enthusiasm and good cheer that couldn’t be tempered, even when the MTA acted like the MTA (I still laugh at the fact the the first ever uptown train was delayed due to non-existent train traffic).

The enthusiasm from Upper East Siders that day wasn’t undeserved. They were finally seeing the end result of a major transit project that they had been promised for nearly a century. What was lost in the excitement, though, was a very important realization: this was only the first phase. With at least 3 more phases to go, the Second Avenue Subway is far from being fully built out, and the struggle to build it may only get more difficult if the MTA doesn’t solve several key issues that have not only raised the project’s budget to levels not seen before, but has also led many to ask what, exactly, the MTA is doing.

The Most Expensive Subway In The World

At the time of writing, Phase 1 holds a rather dubious record: costing $4.5 Billion and only consisting of 2 miles worth of new subway, it is the most expensive subway project ever built per mile, breaking by $150 Million the previous per mile record held by the 7 Line Extension to Hudson Yards. This, justifiably, invited harsh criticism and scorn from urban planners and engineers from both inside and outside New York, with one noted criticism being how, compared to New York, other cities, specifically in Europe and Asia, have shown the ability to build much more subway milage for amounts equal to the $4.5 billion that the MTA spent on Phase 1, with Paris being singled out for comparison by many for its ambitious Grand Paris Express expansion plan (shown below).

Paris’s $29.5 Billion Grand Paris Express expansion plan would add 125 miles to the Paris Metro network for only $236 million per mile, with the lines represented here in bold.

Unfortunately, however, if these criticisms reached the MTA, they were swiftly ignored, for the current price estimate for Phase 2, which would extend the line a mere 30 blocks to 125th Street, is $6 Billion. It’s a price estimate that has left many in the urban planning and engineering communities shocked and appalled. $6 Billion for 1.7 miles of new subway would easily break Phase 1’s record as the most expensive subway project ever built per mile, a fact that becomes even harder to swallow when one considers that, because of an unsuccessful attempt to build the line in the 1970s, a half mile’s worth of tunnel is already there, waiting to be used.

White House infrastructure document on Phases 2 and 3 of the Second Avenue Subway

While this figure, then, is clearly reason alone to criticize, the worries of urban planners and engineers go beyond this. Ever since Phase 2’s price estimate was announced, the biggest question on their minds, outside of wondering if the stations are planned to be made of diamonds, is this: If constructing 1.2 miles of new tunnels and stations will really cost $6 billion, how much will the rest of the project, with all of its new tunnels and stations, be? At the time of writing, the clearest estimate we have comes from a White House document listing infrastructure priorities, which lists the combined costs of Phases 2 and 3 as being $14.2 Billion, a staggering amount of money that, as Phase 4’s costs are omitted, may grow even larger.

With this figure, and all the previous elements of Phase 1, in mind, the question posed in this piece is this: what, exactly, can be done to shrink the Second Avenue Subway’s cost down to a more manageable level?

Making the Second Avenue Subway Affordable

As I’ve discovered from reviewing what the MTA plans on doing with regards to Phases 2, 3, and 4, making the Second Avenue Subway cheaper would not be as simple of an endeavor as many would hope. A large part of this comes from the fact that, because of the way the stations and tunnels have been designed, a thorough redesign would involve changing or downright eliminating most aspects of the project. So, for the sake of clear organization, these changes will be split into two categories: changes focused primarily on stations, and those focused primarily on tunnels between stations.


Second Avenue Subway MTA costs
MTA figures on station and tunneling costs associated with Phase 1 (numbers in millions)

Of everything associated with Phase 1, there is a wide consensus that this is the aspect that inflated the Second Avenue Subway’s cost the most for one simple reason: they’re huge. The platforms are wider than most. They all come replete with full length mezzanines from entrance to entrance, making the stations two stories tall end to end. And, because of this, as shown above, they cost a fortune.

The reason why these elements inflated the stations’ cost is simple: it’s a common understanding in subway engineering that the more you excavate or blast for a station, the more the station’s cost goes up. So, among other things, in order to reduce the construction costs of future stations, their footprints must be made smaller. The approaches to this will vary based on how a station is built, so, for the purposes of this piece, these approaches will be split between stations built using mining techniques and stations built using cut-and-cover techniques.


Making a mined station cheaper is a simple process because, as the costs associated with mined subway construction are only contingent on how much rock is blasted and removed, it simply involves shrinking the station’s size, which can be done primarily by using partial mezzanines.

The partial mezzanine at Roosevelt Island

Phase 1’s full-length mezzanines were criticized by engineers precisely because, outside of looking nice, they serve no real purpose, with the extra blasting required for their construction unnecessarily raising construction costs. Because of this, it’s very important to note that when looking at mezzanines in subway station designs, the only areas of a station where a mezzanine is strictly necessary are where street level entrances and exits are. After this, extra mezzanine space, unless used as storage or ancillary space, does not play a vital role in a station’s functionality. If partial mezzanines, then, were to be used in future stations, they wouldn’t be a detriment to how well a station operates, with the reduction in the basting required for a station directly reducing its cost.


The lessons for mined stations also apply to cut-and-cover. However, cut-and-cover stations are much more complex to slim down because, while a mined station can be made cheaper no matter the depth, a cut-and-cover station’s construction costs are contingent directly on how deep they are. This is because, while mined stations can be built without the need to dig up much of a street, cut-and-cover stations are built by, essentially, digging a big trench, building a station in it, and filling the space on top of the station back up to street-level. Thus, the deeper a cut-and-cover station is built, the more earth is required to be excavated, raising construction costs.

The most convenient solution, than, would be to make cut-and-cover stations shallower. However, while this is certainly a better option than digging a 50-foot deep trench, I believe that there is a cheaper solution than even this that would allow cut-and-cover stations to be deep, but not in a way where they cost prohibitive amounts of money. It comes courtesy of Singapore, where, in order to build a pedestrian underpass below a street, instead of digging the street up, they used a novel solution: a square-shaped, jack powered tunnel-boring machine as shown below (Apologies in advance for the music. I didn’t put it in. I hate it as much as you do):

The application of this in New York would be very similar to this. At each cut-and-cover station site, two trenches would be dug that would act as a launch box and a retrieval point, with the rings shown in the video being supplied from street-level. The TBM itself would be 50 feet wide and 25 feet high to both give the station the airy feel of Phase 1’s stations and to allow a tunnel boring machine to fit through the station tunnel (more on that later). When the station tunnel is completed, the TBM would be reused at the next station site, side platforms would be installed (again, more on that later), and the launch box and retrieval point trenches would be converted into the station’s entrance and exit points, with mezzanines being built in them above platform level.

By using this approach, because the only significant cost of TBMs is the buying price, station construction costs would be significantly reduced as the only aspects of the station that would need to be built with traditional cut-and-cover techniques would be the trenches. Plus, ground level work and disturbance, while still a necessity with this approach, would be reduced as the trenches would be the only areas with significant ground-level disruption.


Example of a Double-O-Tube (DOT) TBM that could be used for the Second Avenue Subway (Photo by SCG Canada)

Admittingly, tunneling was not a major concern from a cost perspective during Phase 1. That being said, I still believe that it could have been done differently. This is because, while the MTA did use a tunnel boring machine for the project, they used one wide enough for only one track, meaning that, after its first run between 96th Street and 63rd Street, they had to disassemble it and then reassemble it at 96th Street for its second run, adding to the total duration of the project. To quicken the tunneling process, then, the MTA could just as easily use a TBM that can build a two-track tunnel in only one run, which, while more expensive up front due to the increased cost of a larger TBM, would quicken the tunneling process while also indirectly making other aspects of the subway cheaper (smaller track crossover caverns, for instance). As hinted at in the last section, a consequence of using this type of TBM would be the use of side platforms at new stations, with exceptions at stations that act as terminals like 125th Street or Hanover Square.

DOT tunnel
Example of a tunnel dug by a Double-O-Tube TBM


This, then, is how the MTA can make the Second Avenue Subway less of a money sink. But, like the subway itself, while this piece took a ridiculous amount of time to write, there is still more to talk about. So, coming up in Part 2, I will look at how these techniques can be used in the Second Avenue Subway’s next phases, and where the line could go after Phase 4, assuming that it’s built out that far before the end of time.


2 thoughts on “City Blueprint: The Second Avenue Subway (Part 1: Fixing The Money Pit)

  1. Yikes, this is way over my head. I just finished the epic “The Path Between The Seas” (a great book about the building of the Panama Canal) and as I began to read this, I was reminded of it. . . without the malaria and the yellow fever of course, but full of the tsunami of money. But I’m glad somone seems to know and understand enough to take a good hard look at this project. BTW, I didn’t see any mention of the fact that the limited mileage only covers the high-income UES, leaving the 125 Phase 2 extension as a lower priority.


    1. I left any socio-economic aspects out of the piece because I wanted it to be focused primarily on its financial aspects, but yes, I completely agree that it’s a bit weird and a little bit suspicious that the project stopped right at the border of Spanish Harlem, especially as they could have easily (and cheaply) extended the line through the existing tunnels to 116th Street. Also, I admit that I nerded out a bit with this one, so apologies if it can be a bit too technical for its own good at times.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s