Home MTA Economics Known Unknowns, or why does everything cost so much?

Known Unknowns, or why does everything cost so much?

by Benjamin Kabak
At $1.4 billion, the Fulton St. Transit Center cost far more than it should have. Can the MTA find a way to control its costs? (Photo: Benjamin Kabak)

At $1.4 billion, the Fulton St. Transit Center cost far more than it should have. Can the MTA find a way to control its costs? (Photo: Benjamin Kabak)

In addition to a lack of political support from Albany, the highest barrier to MTA expansion efforts concerns costs. The one-stop 7 line extension clocked in at $2.3 billion, and the only subway expansion effort in the world that’s more costly is the first phase of the Second Ave. Subway. The MTA is spending nearly as much to rebuild the South Ferry station as it did to construct it, and the East Side Access price tag is comically high and ever increasing.

In a vacuum, the probably isn’t just the costs alone. We know everything costs so much, but we do not know why. Over the years, observers and experts have blamed everything from stringent federal regulations regarding emergency access, a costly and litigious environmental review process, corruption in the construction industry and the uncertainties of digging up old New York City streets. To me, this reeks again of New York City exceptionalism as these are issues facing most developed nations. Somehow, some way, other countries aren’t spending $2.7 billion per new subway mile.

In the latest issue of Capital New York’s monthly magazine, Dana Rubinstein went in depth on the cost issue. For long-time readers of my site (or infrequent and new readers), Rubinstein’s piece is a succinct look at an issue that New York City must solve if it is to meet the demands of its population. Without a handle on costs, the money to expand nets fewer and fewer improvements.

Rubinstein frames her piece around the idea that transit agencies have a rich history of low-balling costs to get money to start a project only to return, cap in hand, for more to finish. Robert Moses deployed this strategy to great effect throughout the city, and the MTA and Port Authority have essentially done the same with their recent construction binges. Think, after all, on how Phase 1 of the Second Ave. Subway should have cost $3.5 billion or how the Port Authority WTC station was originally budgeted at under $2 billion. Once shovels are in the ground, it’s hard to stop, especially if federal grants are involved, and local politicians are forced to fork over the dollars.

What I found even more intriguing though was this excerpt that shows how few people are engaged in this issue:

How New York City’s megaprojects compare in cost to those in similarly developed countries around the world is a question that is, somehow, very rarely studied. Stringer’s spokesman said the comptroller relied for his numbers, in part, on a mathematician named Alon Levy, who’s now completing his post-doc at the Royal Institute of Technology, and who notes, in his blog Pedestrian Observations, that, mass transit is a “side interest” for him and “entirely unrelated to my work.”

The experts at the Regional Plan Association, who are looking into the problem of megaproject cost overruns as part of their latest survey of regional infrastructure, directed Capital to a blog post by Levy, too. The post, from 2011, reported that the Toei Oedo Line in Japan cost $560 million per mile. The Berlin U55 cost $400 million per mile. The Paris Metro Line 14 cost $368 million per mile. New York’s construction costs blew all of that away, the study found. The Second Avenue Subway is coming in at $2.7 billion per mile. The 7 train extension to the far West Side? $2.1 billion per mile.

David Schleicher, an associate professor at George Mason University School of Law, has analyzed Levy’s numbers and says that his analysis basically confirms Levy’s. Barone, of Regional Plan Association, said, “The question is always why, why, why is it so expensive?” said Barone. The answer always seems to come back to a limited universe of issues, in varying combination: labor costs, work rules, managerial incompetence, the spaghetti of infrastructure tangled beneath Manhattan’s streets, a political firmament without incentive to tackle hard issues.

I’ve never met anyone who’s had reason to doubt Alon’s numbers (and you can read the post in question right here on his site). What’s surprising is how few comparative studies have been done to highlight these cost disparities. For its part, the MTA talks about a design-build process that’s supposed to mitigate costs, but working hand-in-hand with the parties responsible for the high costs (that is, the contractors) won’t lead to meaningful reform.

Meanwhile, it’s Chris Ward, a former head of the Port Authority, who has seized on this issue. “It is time to recognize that the delivery model for big projects is broken and fiddling on the margins will not build the kind of projects the region needs,” he said to Rubinstein. Without a better handle on costs, the MTA’s request for $15 billion in capital funding is a tough one to stomach, and future megaprojects are doomed to an expensive limbo at a time when the city and its current and future residents need them the most.

You may also like


D in Bushwick April 2, 2015 - 11:55 am

There’s always an excuse and never a solution to why NYC is the land of corrupt transit.

Brandon April 2, 2015 - 11:57 am

Are there similar high costs for ongoing work, such as station rehabs, track work, CBTC installations, or is it just new construction?

I had looked up some CBTC numbers for a line update in Paris and for the L recently and it definitely seemed that they were out of line by about 3x, IIRC.

I’m glad to hear this is finally getting some attention at a high level. The capital program may be underfunded but we could be making it go so much farther. We could have had the 2nd Ave subway completed with the money we’ve already spent on phase I if we had our costs under control. Its depressing really.

Bolwerk April 2, 2015 - 4:18 pm

They gave a statistic amounting to ~$20 million/mile for track replacement in the new capital program, but they didn’t state whether that was bi-directional or not. If it’s not, that seems pretty high. If it is, it actually seem pretty good.

anon_coward April 2, 2015 - 12:12 pm

there is a lot of underbidding for political purposes. NYC pols would most likely try to kill the ESA if they knew the full cost because it will take money away from the subway once it opens.

don’t know about other places, but a lot of the costs in NYC is buying the real estate for market value. for the 7 line they had to buy a parking lot on 11th and 26st along with other pieces of land.

and for the SAS they are not only paying for expensive RE, they paid for people’s hotel rooms, fixed up damaged buildings and had to tunnel through bedrock.

al April 2, 2015 - 8:16 pm

Speaking of ESA, its costs is approaching that of a new nuclear power plant. How fitting for Long Island commuters.

Christopher April 2, 2015 - 12:29 pm

It’s not just the underbidding that’s the problem, it’s the “we’ll still pay for it when the costs go up” that’s the problem. I’m thinking of a situation in SF with the new Transbay Terminal. It’s become clear that the original cost estimates were low. So what are they going to do about the over-runs? NOT BUILD THEM. It’s such a novel idea that NYC has never tried this. When things go over budget, you figure out how to get them back under budget.

I know a lot of people don’t think much of the Beame era of NYC mayorship, but I’m not sure how many people know that his work before being mayor was building all those sort of bland branch 1950s branch libraries. He was obsessive about keeping costs down (probably too obsessive) but it would be nice if someone took that kind of leadership again. Years ago in Time (or was it the NYT?) there was a long article by Michael Dukakis on the need for really aggressive cost management (and the ways that the Big Dig spiraled out of control after his administration). I wish I could find it again, it was really well written and relevant now more than ever.

adirondacker12800 April 2, 2015 - 3:27 pm

. It’s such a novel idea that NYC has never tried this. When things go over budget, you figure out how to get them back under budget.

42st and 10th was dropped from the 7 extension. The Calatrava dinoskelton was supposed to open up to the sky on fair weather days. There were parts of the Fulton Center that were supposed to be more spectacular and were scaled back.

Justin Samuels April 2, 2015 - 8:13 pm

72nd Street on phase one of the Second Avenue Subway was supposed to have three tracks and two island platforms.

APH April 3, 2015 - 2:35 pm

They killed the radome on the new 1WTC which saved a few million $ – meaning 1WTC – meh…

Webster April 2, 2015 - 12:42 pm

To me, the missing link is the actual bid, themselves.

We can talk about controlling costs by spending less, but the core problem won’t be solved until we understand why the actual bids/contracts are so much higher (why do contractors ask for more to build similar infrastructure and why does the MTA not expect to build that infrastructure for less).

Corruption is a great buzzword, and feels great to talk about as the magic bullet, but I think it’s something more fundamental than all that.

I’d also suspect that the construction METHODS themselves might be slightly different than in other cities (e.g. I doubt there was very much blasting of rock for the RER Express line in Paris or Crossrail in London).

Certainly, there are some extenuating circumstances, but not enough to warrant the discrepancies, but likewise find it foolish to think that the MTA has no interest at all in trying to get more for less money. At least, I’d hope so.

*On a side note, I think it’s weird to keep referring to the 7 extension as $X for Y number of stations rather than the actual length of the line, since you use that as the metric for the cost ($X per Z number of miles).

anon_coward April 2, 2015 - 12:54 pm

it’s still $2.1 billion for about a mile of track. the tunnel goes down to around 18th street and they only finished one station. But i think they did excavation for another few stations as well. they also added some huge fan at 26st to ventilate the tunnel. they still aren’t done with that either

Eric April 3, 2015 - 10:45 am

That doesn’t explain why things that are built the same way everywhere, like signaling systems or Stegosaurus station heads, are vastly inflated here.

Webster April 3, 2015 - 11:30 am

I guess what I’m getting at is whether it makes sense to only focus on cost controls when it seems like the real problem is with the prices being set: contractors offering bids that are a lot higher than they ought to be or over-engineering during the design process.

I don’t personally buy the entire “low balling” argument as the sole explanation, either, since – even when they do so – the projected costs are still almost always higher than they are for similar projects elsewhere.

For example, I don’t understand why the stations for 2nd Ave., the 7 line, ESA, etc are all virtual caverns, that are mined/blasted out of the rock…it seems unnecessary and over-engineered.

The same, I’d suspect, with the Fulton Center and PATH Hub: $2 and $4 billion for a head house sounds ridiculous, because it is, which would lead me to believe that money is being spent elsewhere.

No amount of cost creep could inflate the budget of that thing so high…and until we get their documents, we won’t really know, I guess.

Jon April 9, 2015 - 3:11 pm

The Port Authority’s spectacular cost estimate for replacing the PABT includes over $3 billion for design, engineering, and legal consulting alone. That’s not the only part of the problem, but when you’re spending more on consultants than any other place spends on actually building the thing, you’re off to a bad start.

APH April 2, 2015 - 2:56 pm

The insane costs are not even for mega capital projects – I believe I read the MTA has budgeted $110 million to rehabilitate the 5 or so outdoor N/Q stations in Astoria. Firstly, I should point out that this was supposed to have been well underway by now and slated for completion in 2015. They haven’t even really started as far as I can tell except for repairing a few staircases. So once again their timetables are way off. Secondly, I know nobody bats an eye at $110 million these days, but that comes to over $20M per station. Unless they are rebuilding the iron platforms from scratch (which I think they are not), what the hell costs so much? Am I an idiot for even questioning why it’s so expensive.

Jonathan R April 2, 2015 - 3:02 pm

I think Ben is hiding the lede here. The population of budget analysts at all levels of government with nominal interest in these costs must be at least 1000: US DOT, US Senators’ offices, US Reps’ offices, governor’s office, state legislature, Port Authority, MTA, state DOT, mayor’s office, public advocate, controller, city council, plus all the not-for-profits who write about this issue, and it takes ALON LEVY of all people to connect the dots and assert with evidence that it costs too much to build transportation projects in NYC?

Bolwerk April 2, 2015 - 11:49 pm

The cost thing is an open secret at best. The people who seem best able to communicate it tend to be European, I’d say mainly French and German speaking, railroaders who can understand both their own countries’ practices and the USA’s. It probably helps that French and German-speaking countries tend to be more competent in English than most other countries with advanced rail networks as well as, well, Americans. Here‘s a usenet post about costs for HSR from 2004 (the quoted text is the interesting part).

The part that gets me isn’t that accountants, bureacrats, and politicians don’t care. Why don’t mainstream journalists? This problem isn’t even limited to transit.

Alon Levy April 3, 2015 - 6:51 am

It’s surprising, but I just don’t think there’s a tradition in the US of looking internationally. It took until ten years ago and Ezra Klein for people to notice that the US spends twice as much on health care per capita as the average rich country. This despite the fact that this gap was already large in the 1990s, and health care was a key national controversy then, if not earlier.

(I bring this up because Ezra’s The Health of Nations post was one of the two big influences on my decision to start looking around at how other major transit cities do things.)

Eric April 3, 2015 - 10:47 am

What was the other?

Alon Levy April 4, 2015 - 5:53 pm

Israel has a huge cultural cringe about its politics. There’s even a phrase, “unbroken country,” contrasted with Israel. Tell Israelis that a certain feature of Israeli society or politics is unique in the developed world and they’ll will perceive it as negative; should they want to defend this feature, they’ll try to argue that it exists in multiple developed countries. Then I moved from that environment to the US, where this mentality barely exists – it exists in some quarters, but for the most part, saying the US is globally unique about something is a positive as often as it is a negative. Between that and seeing Americans constantly say certain things are obvious (or impossible) when in other countries they aren’t, I got the “let’s look at how other countries do health care/education/infrastructure/whatever” bug.

SEAN April 2, 2015 - 3:31 pm

Rubinstein frames her piece around the idea that transit agencies have a rich history of low-balling costs to get money to start a project only to return, cap in hand, for more to finish. Robert Moses deployed this strategy to great effect throughout the city, and the MTA and Port Authority have essentially done the same with their recent construction binges.

When the federal government has new starts funding for transit, agencies are forced to lowball ridership numbers to obtain funds. This tactic applies to cost projections as well & it’s part of the game of politics. But some how Portland’s light rail lines come under budget as noted with the new orange line that’s opening this September 12.

Alon Levy April 3, 2015 - 6:58 am

In recent years, US light rail lines have generally met cost and ridership projections. It didn’t use to be like this, but there was a great improvement in forecast accuracy in the 1990s. That said, this was achieved by revising budgets upward rather than revising final costs downward. There are light rail lines in the US, such as Portland’s Milwaukie extension, built for not much less than the European average for a subway.

That said, subways still experience major cost blowouts, and commuter rail lines experience very large ridership shortfalls (see e.g. the Greenbush line and Rhode Island’s South County commuter rail extension). ESA and ARC are outliers and there need to be corruption investigations; in contrast, the 30%-ish cost overruns for SAS and the 7 extension are issues subways have to deal with. Just be thankful you didn’t get the factor-of-3.5 overrun Barcelona L9/10 did. It’s inherently less accurate to forecast costs underground than at-grade, and optimism bias and strategic lying being what they are, this leads to overruns rather than underruns. But there’s a big problem with $1.3 billion/km budgets blowing to $1.7 billion/km ones rather than $130 million/km budgets blowing to $170 million/km ones.

lawhawk April 2, 2015 - 4:41 pm

First off, kudos to Alon for getting recognized in the Rubenstein piece, and for putting the pieces together that others should have done before him (namely anyone/everyone involved in making the decisions that put shovels in the ground).

This site is a daily must-read for me because anyone who lives/works in NYC metro relies on its mass transit system, and I know that without a functioning and expanding transit system, economic growth will come to a standstill and the traffic will become overwhelming.

Along the lines of how and why things cost several times as much as other cities, let us not forget that cities like London, Paris, and Rome have comparable underground infrastructure, to say nothing of the historical/archaeological maze running underneath the surface. Paris is littered with an untold number of caverns and mines, plus catacombs, along with more modern infrastructure. They have unions, work rules, and yet still get things done for a fraction of what we spend in NYC.

If NYC got costs down to the Parisian level, the $3.5 billion from Phase 1 could have gotten nearly 10 miles of subway done. Think about that. Heck, even if we were at twice the Paris cost level, we’re talking about 5 miles of subway done, which would be the equivalent of 5 out of the 8.5 miles envisioned (or roughly Phases 1-3).

I still have yet to see a clear reason why the costs are so out of whack compared to Europe though I think some of it might have to do with some companies deciding to profit from US rules in ways they are restricted in Europe.

al April 2, 2015 - 9:27 pm

Some of it has to do with the limited number of pre-qualified contractors available for bids. This limits competition.

Some of it has to do with the shenanigans that goes on during design phase. For example, NJ Transit design work progress deadlines and benchmarks on the cancelled ARC project was redefined by the very team designing it to receive bonuses.

There is bad design that increases the amount of work.

There is also the fact that parent multinationals who own local contractors look at the MTA, NY gov and PANYNJ as cash cows.

Uncle Moishy April 2, 2015 - 9:09 pm

And yet we all think Chris Christie made the wrong call killing ARC when he did, right?

Alon Levy April 3, 2015 - 7:08 am

I think he made the wrong call killing rather than changing it. But the project as designed was terrible and deserved to die. American commuter rail projects have ridiculous costs, but they’re compounded by bad design decisions, like the ARC and ESA caverns, which turn them from $3 billion projects to $10 billion boondoggles.

Fool April 2, 2015 - 9:18 pm

Notice how even in this context the focus is still not actually on lowering costs, just underestimating costs.

BoerumHillScott April 3, 2015 - 8:24 am

One thing to keep in mind is that most European systems have shorter and in many cases narrower trains than NYC, which directly impacts the most expensive aspect of subway construction: the stations.

This does not cover the majority of the cost differences, but comparing costs only on a per-mile basis ignores where the real expense is.

Bolwerk April 3, 2015 - 10:31 am

That’s a good point.

Also, NYC builds mega stations. Europeans are usually satisfied with a platform, which may not be much more than a slab of concrete on an at-grade line.

Alon Levy April 4, 2015 - 5:56 pm

The Paris RER has longer platforms than the New York City Subway. The cost estimate for the westward RER E extension is around the same as the actual cost of Metro Line 14 – if I remember correctly it’s something like $300 million per km vs. $240 million.

Also, the Stockholm Metro has trains that are only marginally shorter than IRT trains – 140 meters vs. 155. Current cost projection for the extension project: $110 million/km.

Bolwerk April 5, 2015 - 1:00 pm

Aren’t RER deep stations rather notoriously expensive? Of course, nowhere near NYC prices.

Larry Littlefield April 3, 2015 - 9:37 am

“Over the years, observers and experts have blamed everything from stringent federal regulations regarding emergency access, a costly and litigious environmental review process, corruption in the construction industry and the uncertainties of digging up old New York City streets.”

None of this affects the simple replacement of existing infrastructure, as in NYC Transit’s signal replacement program. There the cost was doubled between the 1990s and 2000s, after adjustment for inflation.

I believe that the MTA, which is required to use union contractors, is now carrying most of the pension underfunding crisis for the construction multi-employer pension fund. There is a conspiracy to wreck the transit system not only to pay for the public employee pension increases and underfunding, but that of the private sector as well, so that Generation Greed can safely make it to Florida.

Also, the MTA makes its estimates public, and that becomes the point at which the contractors work their way up from. It is an oligopoly of collusion, with the MTA itself helping to create the collusion by its estimates. In the case of Railworks it isn’t an oligopoly, it is a monopoly.

What could the MTA do? Slash the estimates in half, and make them not estimates. Make them the bid price, for a design-bid process with the preliminary design done. No one meets the price? Fine. Wait a year. Two years. Five years. The money is there when someone meets the price.

Again, I believe there is a conspiracy with the politicians. They want the MTA to pay more to make up for pension underfunding, for the union workers and former executives (who got big bonuses because of the “profits” created by low pension funding). Who do the multi employer pension funds for NY cover? How much of the state or country is the MTA paying for?

Bolwerk April 3, 2015 - 10:43 am

The psychology and politics of cost/deficit scolding don’t help things. Say something to the effect that costs are too high and you get lumped in with some of the stupidest people in society. On top of that, an onslaught of propaganda from groups like the less stupid, but more ideological, Reason Foundation seem to have successfully convinced many nominal transit advocates to eschew the issue, or at least reduce it to some easy fix like throwing buses at the problem.

Eric April 3, 2015 - 10:52 am

Actually, Tea Party members are better educated that the average American…

Larry Littlefield April 5, 2015 - 8:34 pm

Health care and infrastructure construction! What do they have in common?

Borrowed money taken from future citizens (while they aren’t looking and are not informed) by the government, and private contractors.

The private sector is more efficient. One of the things it does efficiently is rip off the government.

fotowoltaika April 6, 2015 - 2:13 pm

The report will enhance your decision making capability in more rapid and time
sensitive manner. The other type is thermal solar power, which uses the sun. ” When combined with another Greek word, grafikos – meaning “image” or “inscription” – we
get the word photograph, which means “image from light.


Leave a Comment