As Congress argues over the size of the infrastructure bill and how to pay for it, very little attention is being devoted to one of the most perplexing problems: Why does it cost so much more to build transportation networks in the US than in the rest of the world?
In an interview in early June, Transportation Secretary Pete Buttigieg acknowledged the problem, but he offered no solutions except the need to study it further.
Biden’s original infrastructure proposal included $621 billion for roads, rail, and bridges. His plan is billed not only as an infrastructure plan but one that would help respond to the climate crisis. A big part of that is making it easier for more Americans to travel by mass transit. The Biden plan noted that “America lags its peers — including Canada, the U.K., and Australia — in the on-time and on-budget delivery of infrastructure,” but that understates the problem.
Not only are these projects inordinately expensive, states and localities are not even attempting to build particularly ambitious projects. The US is the sixth-most expensive country in the world to build rapid-rail transit infrastructure like the New York City Subway, the Washington Metro, or the Chicago “L.” And that’s with the nation often avoiding tunneling projects, which are often the most complicated and expensive parts of any new metro line. According to the Transit Costs Project, the five countries with higher costs than the US “are building projects that are more than 80 percent tunneled ... [whereas in the US] only 37 percent of the total track length is tunneled.”
There’s no one simple reason for this. Experts are clamoring for the government to collect more data, but the complexity of the problem lends itself to complex solutions. Still, the stakes are high. NYU researchers noted the massive economic stakes, pointing to studies that show that building dense urban transit networks could increase aggregate economic growth by roughly 10 percent.
The goal is bigger than a single light rail line or a trolley; it’s about creating transit networks that allow people to navigate from home to work to play easily enough to give up their cars, thereby reducing pollution and congestion. It’s not enough that the nation’s richest states can afford to build one or two transit projects every couple of decades. Bringing down costs is paramount to achieve the climate and economic benefits that accompany transit projects.
So, how bad is the transit and road costs problem?
America’s infrastructure cost problem isn’t just confined to transit, it’s also the country’s highways.
Research by New York Federal Reserve Bank and Brown University researchers reveals that the cost to construct a “lane mile of interstate increased five-fold” between 1990 and 2008. New construction — widening and building interchanges and building new sections of road altogether — is where the bulk of the problem lies, says one of the researchers, economist Matthew Turner. (The cost of “heavy maintenance” like resurfacing increased as well, but Turner said that’s due almost entirely to the rise in the price of certain paving materials.)
Leah Brooks, an economist at George Washington University, also studied the highway problem. Her research finds that states spent nearly three times as much to build a mile of highway in the 1980s as they did in the early ‘60s.
“Big picture for the interstate highway system is that we saw relatively reasonable cost increases [from the mid 1950s] to about the early ‘70s and then pretty substantial cost increases thereafter,” Brooks explained.
And when it comes to mass transit, looking closer at the major projects states have attempted in comparison with their international counterparts is clarifying. The Eno Center for Transportation painted a dismal picture in its recent report: On a per mile basis, America’s transit rail projects are some of the most expensive in the world. In New York, the Second Avenue Subway cost $2.6 billion per mile, in San Francisco the Central Subway cost $920 million per mile, in Los Angeles the Purple Line cost $800 million per mile.
In contrast, Copenhagen built a project at just $323 million per mile, and Paris and Madrid did their projects for $160 million and $320 million per mile, respectively. These are massive differences in cost.
Paul Lewis, vice president of policy and finance at the Eno Center, broke it down: “The US has a pretty big premium for tunnel projects across the board and a slight premium for [above-ground] projects.”
Researchers have ruled out some of the more obvious potential explanations for why these projects cost more in the US.
As transit researcher Alon Levy writes in a report for the Niskanen Center, “This is not about our wealth: there is no correlation between a country’s GDP per capita and its subway construction costs. Nor is it about geological factors: the biggest factor behind a project’s cost is what country it is in, and costs are fairly consistent even across different geologies ... This is purely institutional.”
And when it comes to roads and the bulk of the high costs — “the new construction bit” — Turner says even though he has “no idea why those prices are increasing,” he can “eliminate a lot of things.” Turner explains that common theories like unions or the way we’re building roads or where we’re building them (for example, in more urban areas) are not supported by statistical evidence.
Time is money, money is power, power is delaying transportation projects
I moved to Montgomery County, Maryland, in 1999. Officials had been thinking about building a rail line across the county for more than a decade at that point. More than 20 years later, the Purple Line is only 40 percent built and has run hundreds of millions of dollars over budget, according to the Washington Post.
As the Action Committee for Transit, a local pro-transit organization, documented, residents of the wealthy DC suburb of Chevy Chase have led a decades-long crusade against the light rail project, which will benefit the entire region, by claiming that a “tiny transparent invertebrate” might be at risk. “When no endangered amphipods were found,” the detractors turned to other arguments. However, repeated references to the potential harm to the Columbia Country Club and also a public comment disparaging the needs of people in less affluent communities makes clear that much of the stated concern was likely never environmental or financial.
Gustavo Torres, executive director of a local immigration rights advocacy group CASA told the Baltimore Sun: “Right now, for a domestic worker to go to Bethesda to work, it takes two hours because they take three buses,” he said. “On the Purple Line, it is going to take between 15 and 20 minutes.”
Any major project will incur harms on some group of people — construction can impede traffic, new structures can obstruct a view from your porch, things that reasonable people would agree are annoying or costly. But transportation projects have to go somewhere, and one only has to look at the housing market to see the costs of allowing individual citizens to derail projects due to real or fabricated harms.
Brooks, the George Washington University economist, explained that the case of the Purple Line might explain many of the cost increases we’ve seen over the decades. What does she call this phenomenon? “We call this the rise of citizen voice.”
Brooks ruled out the “perfectly reasonable explanations” like paying workers more or the increased cost of highway materials as driving much of the cost increase. Instead, she says, judicial, statutory, and administrative changes — in particular the passage of the National Environmental Policy Act (NEPA) in 1970 — have led to increased power for citizens.
This isn’t inherently an issue — while it raises costs to engage with lawsuits, if it stops the government from taking harmful action, that could be a good thing! Sometimes costs are rising because we’re paying for something valuable, for instance higher safety standards and accessibility infrastructure like elevators.
But often, we’re just paying for wealthy individuals to exert their preferences over everyone else.
NEPA required environmental impact statements (EIS) for “major federal actions” that could “significantly affect” the environment, write the Niskanen Center’s Brink Lindsey and Samuel Hammond. Lawsuits have turned what were once reasonable requirements into behemoths:
An EIS could be as short as 10 pages ... [but] under the pressure of litigation the law’s demands grew ever more onerous: Today the average EIS runs more than 600 pages, plus appendices that typically exceed 1,000 pages. The average EIS now takes 4.5 years to complete; between 2010 and 2017, four such statements were completed after delays of 17 years or more. And remember, no ground can be broken on a project until the EIS has made it through the legal gauntlet – and this includes both federal projects and private projects that require a federal permit.
“Once the construction process starts, people complain. And those complaints lead to lawsuits,” NYU professor and transit researcher Eric Goldwyn explains. Even in New York City where in 2018, 62 percent of all reported trips were made by transit, biking, or walking, the Second Avenue Subway project was plagued with disruptions by locals.
SPUR, a California public policy nonprofit, argues that its state’s environmental protection law (CEQA) “is often used inappropriately to delay or stop transit and sustainable transportation projects that would have significant benefit to the environment.” As city planner and UCLA housing researcher Nolan Gray documented in the Atlantic, CEQA lawsuits have “imperiled infill housing in Sacramento, solar farms in San Diego, and transit in San Francisco. The mere threat of a lawsuit is enough to stop small projects—especially housing—from starting in the first place.”
“There are egregious examples of this,” SPUR’s transportation policy director Laura Tolkoff told me. “One person sued the San Francisco bicycle plan over parking losses which basically held up 34 miles of bike lanes for over four years. Over 2,000 individuals were injured during that same time period due to collision while riding their bicycle.”
In Tolkoff’s example, the state’s environmental protection law was being used to defend a parking lot.
Delving deeper: The case of Boston’s Green Line Extension
To try and figure out the root causes of America’s transit costs problem, NYU researchers Goldwyn, Alon Levy, and Elif Ensari looked to Boston, where the Green Line Extension (GLX) has been in the works for three decades. The 4.3-mile light rail project was estimated to cost $1.12 billion in 2012 — by 2015, that number had nearly tripled. Unsurprisingly, the project was put on hold.
“One of the things that is so vexing about this issue is that there’s not just one obvious thing,” Eric Goldwyn, an assistant professor at NYU and a member of the Transit Costs Project, told Vox.
But the researchers were able to identify a few reasons for what happened to the Green Line: Jockeying between two different understaffed agencies with little experience managing large projects and consultants, a laissez-faire approach to allowing stakeholders’ expensive ideas to be added to the project scope “even if impractical,” and public pressure for more as the project dragged on and the demand for transit options increased.
While Boston’s cost overruns were extreme, even for the US, the lessons learned in this case study are ones repeated by the experts Vox spoke with.
One reason the US isn’t very good at building transit cheaply is that it doesn’t practice.
“If you look at Paris or Seoul or Shanghai, they’ve been building [transit] pretty much nonstop for decades now. New York, on the other hand, built its subway at a breakneck pace until 1940 and then cooled it,” Goldwyn explained.
Agencies aren’t routinely in charge of building new things, so every time they do, it’s back to the drawing board.
“There’s a learning curve, almost with every city, when they’re introducing rail, because you don’t have a local knowledge base on how to do this,” explained Ethan Elkind, director of the climate program at UC Berkeley’s Center for Law, Energy & the Environment. “Whether that’s on the construction side or the oversight side or public management side. You do see a lot of cities stumble out of the gate when they’re trying to get their first projects done.”
Then there’s the complexity of building across multiple jurisdictions. The federal government often provides funding for a project that requires multiple cities or counties to coordinate, all to complete a multibillion-dollar project unlike one they’ve probably ever accomplished before, often without a clearly defined leader — it’s like the most dysfunctional group project ever.
Elkind noted that even when the federal government works with a single jurisdiction, it can get messy, pointing to “the existence of both Farragut North and Farragut West stations [in Washington, DC] — there was supposed to be one station but the federal government wouldn’t hand over this parcel that would have provided for one unified station.”
Further, these agencies are often understaffed, creating bottlenecks at every step of the process: “The nature of these projects is that you get questions about like ‘should I use this CCTV wiring or that one’ — you need people who can respond quickly and be expert enough on an array of topics [to keep the project moving],” Goldwyn explained.
Levy highlighted the failure of American institutions to turn to other parts of the world to learn what they are doing better: “Getting more of these planning managers to say we need to be more German, more Italian, more Swedish, Korean — that’s really hard,” they said. “The United States is so bad at learning from other countries. If something in America doesn’t work [and] in other countries it does work, people don’t really course-correct.”
Transit agencies and local governments also sometimes try to preempt public outcry, creating more work for themselves in the process. The Boston Green Line, for instance, proposed limiting construction to between 10 pm and 7 am and required that traffic keep flowing. But as the transit costs researchers noted, “laborers are less productive when they have to spend the first hour of their shift setting things up and the last hour breaking things down to mask the fact that there is a major construction project underway. ... [I]n Istanbul, transit construction projects run 24 hours a day.” In the span of seven years (the expected duration of the Green Line’s 4.3-mile project), Istanbul built over 12 miles of subway.
“Here in the United States, we as a country are not very accepting of disruption,” Goldwyn told Vox. In Los Angeles, workers constructing the Purple Line could only work weekends — then Covid-19 hit, and stay-at-home orders made it possible for them to work 24/7. The result?
“They completed the project seven months ahead of schedule.”
America needs to spend more — not less — on transportation
After all this talk of cost overruns and inordinate spending on various projects, you might think that the US is spending way too much on transportation. In fact, according to a report by the Brookings Metropolitan Policy Program, the nation’s transit spending “fell by $9.9 billion in inflation-adjusted terms” over the last 10 years. In comparison with similar countries, America spends a relatively small amount of its GDP (1.5 percent) on public infrastructure, while the UK spends 2 percent, France 2.4 percent, and Australia 3.5 percent.
The problem is fundamentally that the US is getting very little for what it builds.
“Don’t miss the bigger prize here, which is, are we building the right infrastructure, in the right places, for the right reasons. We’re not lacking money here,” Adie Tomer, Brookings Metropolitan Policy Program fellow, argued. “We need to somehow talk about infrastructure efficiency and project delivery, but there’s also a bigger story here too.”
Step one is more data. The research we have on cost is limited by the fact that very little data is collected to figure out what each portion of a project actually cost. If a project is getting federal money, requiring that it provide an itemized list of costs to the federal government is a first step.
“The first is like the most benign thing, which is you should tell us how much a project costs when it’s done,” Brooks told Vox. “At the moment, the Federal Highway Administration doesn’t know how much it costs to build any individual section of the highway.”
This isn’t a new idea. As Matt Yglesias reported for Vox in 2017, “someone killed a congressional inquiry into America’s sky-high transit construction costs” with no explanation.
There are also myriad ways the US needs to streamline the process for developing transit projects. Lewis explained to me that European regulators were often shocked that American transit agencies have to go through their own process to get authorization to shut down a street or prepare an area for construction.
“A lot of the [processes] that we use here in the United States are too slow or too cumbersome and outdated. We need to make it easier to build more and better transit projects,” Lewis explained.
But cutting down on bureaucracy doesn’t mean slashing government budgets — while simplifying the rules and regulations that go into developing projects, American transit agencies need to be staffed up in-house to reduce reliance on expensive contractors and build up institutional knowledge.
Turning to other countries can show how to accomplish some of the country’s transit goals (environmental protection, for instance) without exorbitant costs — Levy writes that while the US paves the way for expensive lawsuits, Italy’s equivalent federal environmental law “is enforced by the administrative state.”
There is no magic button or single regulatory fix that will put the US on the path to a future with accessible, plentiful, and cost-efficient transit. But public fascination with seemingly futuristic forms of transportation like Elon Musk’s hyperloop dream or a network of high-speed rail crisscrossing the country proves the value of pushing through the bureaucratic and legal morass.