Home MTA Economics From the RPA, a call for MTA megaproject reform

From the RPA, a call for MTA megaproject reform

by Benjamin Kabak

A new RPA report, released as part of the group’s fourth Regional Plan, delves into NYC’s high construction costs.

It’s high times for close-ups on the MTA’s out-of-control capital spending, and hot on the heels of The Times’ deep dive into the MTA’s cost problems comes a long awaited Regional Plan Association report on the very same topic. The title is staid — Building Rail Transit Projects Better for Less — and doesn’t exactly roll of the tongue, but the conclusions are forceful. “The entire process of designing, bidding, and building megaprojects needs to be rethought and reformed top-down and bottom-up,” the RPA says, not mincing words.

In a certain sense, the RPA’s report takes a modest approach to cost reform goals. The RPA believes the MTA could save between 25-33% by implementing its recommendations. While those savings would bring costs more into line with the most expensive international transit projects, the MTA would still be a leader in the cost realm if, for instance, East Side Access came in at $7 billion instead of $10 or Phase 1 of the Second Ave. Subway checked in at $3 billion instead of $4.5. But the extra billions in essentially free money would be put to good use.

The full report is available as an 80-page PDF and, while not perfect, is worth the read. It delves into three case studies focusing on the Second Ave. Subway, the 7 line extension (the best of three overly expensive projects) and East Side Access before analyzing New York City in the context of its international peers. When stacked up against London (Crossrail and the Northern Line extension), Los Angeles, Paris and Madrid (along with a few other cities thrown in for good measure), our models fall apart. New York doesn’t adopt best practices when it comes to capital construction and thus cannot keep pace with technological innovations that would improve service reliability and modernize an old system as many international peer systems of a similar age have. It’s an ugly and familiar story all around.

A comprehensive set of recommendations

As with other recent pieces on the topic, the RPA is careful to spread the blame. Much to the frustration of those outside the transit realm, there is no silver bullet. “The extraordinarily high costs associated with building transit projects in New York are due to many factors, at every stage, from decisions made by political leaders at the inception of the projects to the final stages of lengthy planning, design, and construction processes. Long tolerated as an accepted natural consequence of New York’s size and dominance, these costs threaten to strangle the region’s future economic growth. Other global cities have outpaced New York in building modern infrastructure and attracting new business and residents as New York struggles to simply keep up with basic maintenance.”

To fix the problem, the RPA sets forth 22 recommendations. These include a streamlined environmental review process that could cut years off the planning process. Projects in New York can take up to seven years to get off the ground due to the environmental review requirements while similar work in Europe can begin in 18-24 months. The RPA also recommends accurate project budgets and timelines to avoid funding and procurement problems down the line; more of a reliance on off-the-shelf design and construction elements to avoid costs of customization and change orders; industry-standard project management and procurement practices; streamlined post-project review; the ever-popular work-rule reform; a ten-year capital funding pipeline to better align with construction timelines; and a reform of land-use and zoning practices so that the city and state coordinate on value capture. As a break from the familiar laundry list of reforms, the RPA also points to a shortage of skilled and qualified laborers for the work on the MTA’s slate and urges the creation of job-training centers as well.

Inside the Second Ave. Subway case study

While the RPA by and large seems to reinforce what many in the know have urged the MTA to do for years, the case studies, particularly with regards to the Second Ave. Subway, lay bare the the MTA’s problems. You can read Aaron Gordon’s analysis on the impact of cutting the third track at 72nd St., and take, for instance, the cost of phasing out the entire line:

SAS’s first phase offers many lessons for phase two, starting with how best to manage the overlap of a project with multiple phases. To date, the MTA has not completed engineering and design for phase two, which is only just getting started. This has made it impossible for the agency to coordinate the winding down of phase one with the ramping up of construction for phase two, which would have allowed the experienced crews working on phase one to move directly onto the next segment. Instead, the MTA ended major construction in January 2017 and left the neighborhood, with plans to return in roughly three years. The loss of experienced labor, current staging docks, and remote office spaces means the MTA will have to start the process all over again — adding to the time and costs of phase two as well as the disgruntlement of the neighborhood. And without institutionalizing the lessons learned from phase one, the MTA risks the loss of institutional knowledge that agency staff and project managers have developed on the job over the course of building SAS.

The report includes a one-page featurette on the decision to forego development at the six lots along 2nd Ave. that are home to ventilation plants and station entrances. The MTA lost out on approximately $100-$125 million had the city and state coordinated on rezoning and air rights transfers. As costs go, the Second Ave. Subway was subject to thousands of change orders, and the costs add up:

According to the GCA, the SAS had thousands of change orders, with 96th Street issuing 200 orders during the acceleration phase alone (a claim the MTA disputes). Change orders modified more than 30% of the contract plans. An extreme example occurred with the electrical contractor at 96th Street having over 70% of its bid scope of work modified by change orders, which were issued serially as items were discovered. “This required the general contractor to perform other work out of sequence,” noted GCA, “and in many instances, remove and replace (at its expense) work that had already been completed and installed.”

The stories and dollars just keep adding up. Take, for instance, the station costs:

The high costs of the 96th Street station can be partly explained by the inclusion of 65,000 square feet for the MTA workforce in underground facilities and office space, requiring expensive blasting. The station has hundreds of non-public-access employee spaces. This equates to three to four times as many employee spaces as any other station along the line. The MTA’s justification was that 96th is a terminus, which is only temporary because the line will actually terminate at 125th Street when phase two is completed. Instead of spending the extra millions of dollars to build these temporary facilities, the MTA should have explored the cost-effectiveness of providing employee spaces at the surface by renting commercial space.26 There are also redundant employee facilities at 86th Street and 72nd Street.

Of course, another reason SAS’s stations were extraordinarily expensive lies with the materials. The archway entrances, for example, are built of granite that is six to eight inches thick. As one MTA project manager noted, part of the problem is nothing is off the shelf, with all of the granite being custom-produced.27 Granite, by nature, is nonstandard; each piece is unique due to its geologic formation. Under Buy America procurement rules, the MTA was required to buy American materials. Yet the United States has only a few suppliers that could provide granite of this size, custom-cut to curve at around eight inches thick to support the weight of the archways. This was a deliberate material design decision that almost surely should not be repeated in the future if the MTA hopes to contain costs.

The examples go on and on and on, and by the end of the case studies, one may think that the Second Ave. Subway was designed with no regard to cost at all. The construction of Phase 1 was, from start to finish, a textbook example of how not to build a subway line at a reasonable cost and within a reasonable timeline, and the MTA has shown no indication that things will be different for Phase 2.

Criticism of the RPA’s report

For the strength of its examples, though, the RPA report has not been immune from criticism, and Alon Levy, the resident expert on transit construction costs, has weighed in with a thoughtful critique of the report that is also worth the read. Levy highlights mistakes in the details of the RPA’s cost comparisons that are hard to ignore and rightly charges the organization with an Anglo bias in that it “overvalue[s] other English-speaking countries, even when their construction costs are the highest in the world outside the US.”

Levy doesn’t hold back in his conclusion, urging the creators of reports such as these to delve deeper and draw out the necessary comparisons. He writes:

One of the things I learned working with TransitMatters is that some outside stakeholders, I haven’t been told who, react poorly to non-American comparison cases, especially non-English-speaking ones. Ignorant of the world beyond their borders, they make up excuses for why knowledge that they don’t have is less valuable. Even within the group I once had to push back against the cycle of failure when someone suggested a nifty-looking but bad idea borrowed from a low-transit-use American city. The group’s internal structure is such that it’s easy for bad ideas to get rejected, but this isn’t true of outside stakeholders, and from my conversation with Tom Wright about Gateway I believe the RPA feels much more beholden to the same stakeholders.

The cycle of failure that the RPA participates in is not the RPA’s fault, or at least not entirely. The entire United States in general and New York in particular is resistant to outside ideas. The political system in New York as well as the big nonprofits forms an ecosystem of Americans who only talk to other Americans, or to the occasional Canadian or Brit, and let bad ideas germinate while never even hearing of what best industry practices are. In this respect the RPA isn’t any worse than the average monolingual American exceptionalist, but neither is it any better.

The RPA of course knows its audience and knows how far recommendations can go within the MTA. Scott Rechler, the chair of the RPA, is a bright voice on the MTA Board, and he knows that institutional resistance to reform runs deeper. So perhaps in that sense, the report is written for its audience and goes as far as this audience is willing to go. That may be part of the problem. It is another salvo in the MTA’s war on costs, one I would read with Levy’s grain of salt but not dismiss out of hand. The question is: Who’s listening and when and how will change arrive? The city cannot afford to spend more while getting less and less each year.

You may also like


Alon Levy February 11, 2018 - 7:49 pm

The worst part of it is, they do go over two of what I consider the top three problems (procurement, and station costs)… but despite cogent analysis their recommendations say nothing about awarding contracts based on technical score and having less onerous specs. And for some reason they don’t follow through on the analysis of New York station costs by comparing them to other station costs; whereas 75% of the non-overhead costs of SAS phase 1 were stations, only 35% of the costs of the Line 1 extension to Val-de-Fontenay here are stations. The only one of the top three problems they include in the recommendations and not just the analysis is labor, and even there they seem more rooted in local criticism of the unions than in the practices that Rosenthal’s article found.

Christopher February 11, 2018 - 8:12 pm

I’d be concerned about some non-European examples as well. Not that they don’t have things to learn but their relationship with labor and work is not necessarily comparable. That multiple thousands of workers build a station in China thing is neat but couldn’t necessarily be replicated here. We just can’t summon workers in the same way. It’s part of what Ai Wei Wei references in his work and why likes to produce things in huge quantities. There’s even questions to me about how well we can compare ourselves to labor markets with a large social safety net and socialized healthcare too. But I think the difference with the global south and East Asia is stronger.

Alon Levy February 12, 2018 - 8:58 am

Anticipating your criticism, Rosenthal stayed away from comparisons with China and Russia and focused on France. But not all of Asia is China. Japan not only is democratic, but has stronger protections for the individual against the state in eminent domain disputes than the US does.

As for the health care point, Rosenthal’s sources indicated that health care costs only raise US construction costs by 10%; the RPA report doesn’t quantify it but similarly finds a small effect. It doesn’t explain why SAS cost $1.7 billion per km and Line 14 here cost $250 million per km.

michael February 12, 2018 - 6:57 am

While I agree that European and Canadian examples are probably more relevant than Asian, we should not ignore higher cost places such as Hong Kong and Singapore where land, people and environmental standards are pretty rigorous.

As someone who used to work on Crossrail, the real costs in an established urban are works to integrate with existing rail infrastructure, particularly stations but also existing rail lines – SAS Phase 1 has very little of this if which makes the comparisons either more stark.

Alon Levy February 12, 2018 - 9:03 am

Singapore has very high construction costs, without any real reason for them. The Thomson MRT Line, built under a wide road flanked on one side by housing projects and on the other by a forest, looks like $600 million per km.

Larry Littlefield February 12, 2018 - 7:28 am

What everyone is aware of but refuses to discuss is how much of the MTA’s costs are for the retroactively enriched and underfunded pensions of those who worked on public AND private projects in New York 10, 20, 30 years ago, now retired and living in Florida.

But the Empire Center estimates that added cost at 25 percent — over and above what is paid for the future benefits of those employees actually working today.

Moreover, if the MTA has a budget for a project, every contractor knows with the MINIMUM BID for all of them is. When the budget is exceeded, that becomes the new baseline for the next, similar project.

“Costs” are someone else’s revenues.

VLM February 12, 2018 - 8:54 am

Just curious, Larry. Did you read the report or anyone’s coverage of it? For something you claim everyone “refuses to discuss,” it sure is front and center in the discussion these days.

Larry Littlefield February 12, 2018 - 9:57 am

What everyone has started to discuss is the TOTAL costs of labor. Because what everyone wants to do is cut the cost of the workers on the job — to far less that what those who already cashed in took.

It’s the “screw the newbie, flee to Florida” cycle, as in the public sector. And once again the public sector is paying for it.

SEAN February 12, 2018 - 2:57 pm

You know Larry, this may interest you. In May of 2016 I took a trip out to Garden City, but before departing at Minneola I over herd a conversation between another passenger & two conductors.

Female passenger… my daughter just moved to Tampa & she is never coming back. Conductor #1, I have eight years until I retire. Conductor#2, I have ten years. Discussion of moving to Florida ensues. I held back from bursting out laughing while exiting the train, as you were the first thing that popped into my mind.

Larry Littlefield February 12, 2018 - 4:03 pm

People deserve the pensions they were promised when they were hired.

But retroactive increases wrecked this city in the 1970s and will do so again.

And pols here and elsewhere can’t resist failing to fund the pensions workers have been promised to begin with, so they can hand out more goodies.

All and sundry lie to cover their tracks.

Not only do we face soaring costs for public pensions, but also for the private pensions of the politically connected. Meanwhile, the serfs are facing future cuts in Social Security, which is all they have.

Alon Levy February 12, 2018 - 9:07 am

What VLM said. And I’ll add that pensions aren’t why the MTA uses lowest-bid contracting, or why it uses 24 workers to do the job of 9, or why it doesn’t build stations cut-and-cover the way everyone else in the world does when it’s possible (which it was for both SAS and the 7 extension).

The Empire Center has some serious problems when it comes to efficiency. It’s designed to embarrass workers who make a lot of money, not to look at staffing levels. The SeeThroughNY tool has nightmarish UI if you want to figure out how many employees there are in each category (say, train drivers), because it separates categories like “train operator rev veh” and “train operator-rev veh” (the hyphen is the only difference). And Nicole Gelinas repeated union talking points about OPTO being unsafe essentially because OPTO conflicts with her goal of reducing wages and benefits.

Larry Littlefield February 12, 2018 - 10:00 am

Again, this is about the construction companies and their contracts with the construction unions. The Times say the MTA has no control over those contracts, but didn’t mention why their cost has exploded. It’s the cost of the retired.

This article shows how private construction, which got a better deal back in the day by underfunding the pension, is now dropping union construction to avoid the costs exploding as a result — leaving the public sector to pay for it.


Larry Littlefield February 12, 2018 - 10:06 am

Here is the key section of the article.

“The Empire Center report calculated that the government ends up paying 25 percent more for public projects in New York City because of the high prevailing wages.”

“McMahon’s report argues that the union’s pension benefits are so high—in many cases nearly 50 percent of workers’ compensation—because the unions have to fight for very expensive pension funds for workers.”

“The problem is that they have priced themselves out of competition. And it’s not just greed but that they have to back fill their unpaid liability of pension shortfalls.”

I can tell you direct from people in the trades that right after pension increases and pension underfunding in the 1990s stock market bubble, givebacks started for younger and new union members. Early 2000s. This money is for past public AND private projects.

What is fair in this situation is debatable. What is unfair is to sweep it under the rug, so younger generations of subway riders and union members won’t know they’ve been screwed.

Alon Levy February 12, 2018 - 6:52 pm

But pensions for now-retired workers are not a marginal cost of construction. The MTA needs to pay the same if it never engages in system expansion again or if it builds the fantasy maps of every NYC Transit Forums regular.

Larry Littlefield February 13, 2018 - 10:50 am

That may be true in practice, but not in theory.

In theory, if the MTA were to shut down all capital projects for five years and tell the construction industry it was going to hold out until it got a better deal, it wouldn’t be paying for construction union labor. And thus wouldn’t be paying the add-on for unfunded pension benefits.

So it is a marginal cost of all future union construction, even if those doing the work now are paid less.

In practice, even if the MTA shut down the capital plan, the state legislature and Cuomo would probably still order them to borrow a few $billion and send the unionized contractors checks to put in the pension fund.

Adirondacker12800 February 13, 2018 - 12:28 pm

The construction workers don’t work for the city ( or the state ) and don’t participate in the public pension plans that have the problem.

Larry Littlefield February 13, 2018 - 1:54 pm

You need to read the article I linked.

The unionized workers are part of private multi-employer pension funds that have their own problems, as a result of retroactive pension increases and pension underfunding in the 1990s and early 2000s.

These are private plans, but the private sector wants the public sector to pay.

There are similar issues all over the country, but in the rest of the country they don’t have an MTA to siphon money from.


“By the late 1990s, many plans were fully funded. In this environment, unions were concerned that employers would stop contributing to the plans due to limits on the tax deductibility of employer contributions to fully funded pension plans. They were wary of interrupting the flow of contributions, because restarting contributions when markets cooled would require reducing other components of compensation. To ensure that contributions remained tax deductible for employers, plans offset the increased funded levels by repeatedly increasing benefits. The good times ended with the bursting of the dot-com bubble in 2000.”

We just had to increase benefits for older workers cashing in and moving out!!! Now somebody with no pension benefits has to pay!!!!

SEAN February 13, 2018 - 5:09 pm

Curious Larry – how do you know so much about this subject? I really want to know – no sarcasm implied.

Larry Littlefield February 13, 2018 - 6:21 pm

In the late 1980s, when I was a junior city planner and regional economist at the NYC Department of City Planning, a new City Charter passed, mandating a set of reports. I ended up being assigned to write the economics sections of them.

I was told the purpose of the Annual Report on Social Indicators was to provide the City Council, newly vested with budgetary powers previously held by the Board of Estimate, with information. And I discovered all the data collected by the Governments Division of the Census Bureau. So I started compiling and making comparisons for use in the report.

And I started getting pissed off. Then I found other data sources and started compiling them. And got more pissed off. Pretty soon I was pretty much the foremost expert on comparative state and local government around. But I was so pissed off I ran a Don Quixote campaign against the local state legislator in 2004. and exited the public sector.

Before and after working at City Planning I worked at NYC Transit, the second time in capital budget. Which, of course, made me even more pissed off. I left and ran my protest campaign because I didn’t want to see the transit system I cared about go down the tubes right before my eyes.

I’ve worked in the private sector ever since and feel much better.

SEAN February 13, 2018 - 8:57 pm

Explains a lot, thanks.

J Adlai February 12, 2018 - 10:17 pm

You quoted that terrible Aaron Gordon article. In it he says that the southern section of the SAS will only be able to operate at 6 to 7 minute (it’s actually 4) and then he also states that the A can only operate every 6 to 15 minutes during rush hours (it’s actually 3.5 minutes). Since then, outlets like Jalopnik have picked up the article and taken the issue even further. It’s like Gordon didn’t bother to read the FEIS and see that even with 3 tracks at 72nd street, the Q and T were going to share tracks.

Setting that aside, and getting to the crux of this piece, there really is no incentive to cut costs. Larry Littlefield pointed out that rising construction costs have caused the private sector to start hiring non-union labor: The labor unions refusal to give up some work rules on the private side has led directly to loss of work. However, even though there is no mandate that the MTA use union labor (and in fact, sometimes non-union contractors do win jobs), the MTA has no incentive to use non-union labor, or get better deals with unions. Even with all of this recent coverage of the cost of construction out of control, has anyone with any real authority tried to take up the mantle of cost reform?

Subutay Musluoglu February 13, 2018 - 6:58 am

Thanks for pointing this out – you beat me to it. Bad reporting by Mr. Gordon; he is speculating at best. There was never an intention to terminate the Q at 72 Street; as per the EIS and the accompanying analyses, the Q and T services were meant to run simultaneously upon the opening of Phase 3. The third track at 72 Street would have served as a double ended pocket track in specific instances of schedule perturbations which could cause a conflict, such as a northbound T and northbound Q arriving at the same time. It would have enabled to get a train out of the way so as not to slow down the mainline in either direction. It would also have helped during General Orders requiring single tracking north (or south) of 72 Street, so that cross platform transfers can happen to the through service. A pocket track is not an ideal terminal arrangement for every day service, at least in NYC. We can debate the long term impacts of deleting this feature in the interests of saving costs, but frankly if you wanted to run two services of 25+ tph each, then the entire line would have been built with 4 tracks. Under the current scheme, after Phases 3 and 4 are built the T can be complemented by an additional service to and from Queens via the bellmouths at 63 Street, but that would require an additional trunk line in Queens.

J Adlai February 14, 2018 - 11:34 pm

In theory, you don’t need to build another Queens Trunk line. With better terminal operations you could you could split the Queens Boulevard local 3 ways and operate one of those services down the Lower Half of SAS. Or, you could try and get 36 trains per hour out of CBTC (which should be achievable) and operate trains on the E, F and SAS at 5 minute headways.

But connecting the 63rd tunnel to a new trunk line (either operating along the old Horace Harding alignment or the taking over/sharing the Port Washington Line) would be the best use of the line. The new trunk line should be served by the F, while the SAS line should serve the QB express.

Larry Littlefield February 15, 2018 - 9:57 am

Here is another report by the Empire Center on the soaring cost of private construction pensions for public entities.


As is the case for public employees, do you think the fact that their total compensation is so much higher than for the rest of us makes today’s workers more motivated? No.

“The crowding-out impact of rising benefit costs was especially large for workers in 60 of the job titles surveyed, whose hourly pay actually decreased in real terms over the past 10 years. Titles with hourly pay hikes below the 17 percent inflation rate since 2007 included painters, masons, ironworkers and laborers in Buffalo; electricians, ironworkers and plumbers in Rochester; sheet metal workers in New York City; and teamsters in Rochester and Syracuse.”

That’s in the past 10 years, and perhaps it’s even worse if take home pay is considered, because the first thing they did was jack up existing employee contributions to pay for the retired. The guy who told me that, before the data above, WAS a NYC sheet metal worker.

In the 1980s New York’s prevailing wages laws did NOT include fringe benefits. That’s how all that affordable housing got built in the former vacant lots and abandoned buildings. They probably added fringe benefits to make the MTA pay for pension increases and underfunding in the 1990s — with both the construction unions and the construction contractors in on the deal.

Larry Littlefield February 15, 2018 - 10:54 am

This is just one part of the big picture.

There is no federal money because it is all going to Generation Greed’s tax cuts, low gas taxes, and enriched senior benefits those coming after will never see.

There is no state and local money because it’s all going to Generation Greed’s public pensions and debts.

There is no MTA money for both reasons, plus this one. Even at a nation-leading tax burden.

And those who made the decisions that have left us here are still in charge, and expecting to take more.

smartone February 15, 2018 - 2:35 pm

I have a question – the 7 Train to Javis extension was funded by City so it didn’t have many of the required hoops to jump thur that are attached to Federal and State funding and yet the cost of the project construction cost were the same as other mega projects. And yet it implies that cutting State and Fed requirements around Megaprojects would improve cost.

which is it ?

Larry Littlefield February 15, 2018 - 6:37 pm

If you are going to ask that question, you might as well ask this one.

If NYC’s high capital costs result from the need to do the work in a 100-year old system operating 24 hours a day, how come that line and the SAS were also extremely overpriced?

Billy G February 16, 2018 - 12:56 pm


Leave a Comment