Here's What $4 Trillion In Infrastructure Spending Could Fix

The day before President Joe Biden's inauguration, Senator Joe Manchin, a democrat from West Virginia, said on a local news program that while he's not opposed to $2,000 stimulus checks, he'd prefer to support a massive infrastructure program. And by massive, I mean almost incomprehensibly massive. Say, $4 trillion massive.

"The most important thing? Do a big infrastructure," Manchin told the news program Inside West Virginia Politics on the local CBS affiliate. "Spend $2, $3, $4 trillion over a 10-year period on infrastructure. You want to put everybody back to work? There's a lot of people who lost their jobs and those jobs aren't coming back. They need a place to work."

Putting aside whether this is a wise job creation program for a recession that has impacted the service economy more than the construction sector, it's still big news. Without a doubt, the U.S. has a lot of infrastructure needs. Democrats only have a majority in the evenly-split Senate, thanks to Vice President Kamala Harris's tie-breaking vote, so moderate Democrats like Manchin will have enormous influence in Biden's first two years. When Manchin says he supports up to $4 trillion in infrastructure spending, it's a big deal.

This is obviously a staggering amount of money, so much money it's difficult to even wrap my head around how much money it is, in the way that it's hard to contemplate the number of stars in the sky (astronomers estimate there are up to 400 billion stars in the Milky Way, so we're actually talking about ten times more dollars than the number of stars in the Milky Way). As the Eno Center for Transportation's Jeff Davis put it over email, this is "just a stupendous amount of money."  So let's break this down a bit. 

Let's start relatively small by imagining four million dollars. For you and me, this is probably a difficult amount of money to imagine possessing in its own right (if it's not, then good for you) but I can look up what four million dollars buys with relative ease. A luxury Manhattan apartment, for example, or two Bugatti Veyron supercars. Now imagine you could buy one million of those big Manhattan apartments, or two million Bugatti Veyrons. That's $4,000,000,000,000. 

But Manchin isn't talking about buying Manhattan apartments or supercars. So what does $4 trillion buy in infrastructure these days?

The Biden team called for $2 trillion in spending in his first term—which, on per-year basis, is actually more than Manchin's $4 trillion over ten years—as part of his clean energy promises, including massive expenditures on public transit, the auto industry and the power grid, constructing 1.5 million housing units, weatherizing two million existing homes, and more. Manchin floated the idea of spending double that amount, albeit over a longer period. 

Of course, how much $4 trillion actually buys in infrastructure improvements depends on a great many factors, including what infrastructure you're actually buying. Fortunately, the federal government has been spending lots of money on infrastructure for decades, so we have a fairly good idea of how much things cost.

Davis pointed me to a 2018 Congressional Budget office report that looked at how many tax dollars—from federal, state, and local governments—have been spent on transportation and water infrastructure since 1956, normalized into 2017 dollars. Although it fluctuates over time, Davis said $4 trillion would be "equivalent to all combined federal-state-local spending on transportation and infrastructure for about 9 years at current spending rates." 

On some level, this is a humbling statement because we basically already spend $4 trillion every decade or so on just two infrastructure needs that don't even include housing or energy infrastructure. But there are two different types of infrastructure spending. There is the cost to construct new things like rail lines, wind turbines, buy new buses, put in new sewers, and so on—known as "capital" costs—and then the cost to keep those things maintained and working like paying the bus drivers or fixing the train signals and unclogging the pipes—known as "operating" costs. 

Traditionally, big infrastructure stimulus programs are focused on the types of stuff that fall under "capital" costs because they're more expensive and also flashier projects for the politicians announcing them (although I for one would vote for any politician who holds a ceremony for a stimulus program that pays city bus drivers a living wage). If we're looking just at capital spending, $4 trillion would be, according to Davis, "roughly equivalent" to all of the government spending on capital costs for infrastructure in the 21st Century so far. 

So, compared to how much the U.S. government at all levels has been spending on infrastructure so far this century, we're talking about a huge increase. But still, this is all pretty abstract stuff. Time to get specific.   

I asked Yonah Freemark, a senior research associate at the Urban Institute and a longtime chronicler of transportation costs, how much we could get for this money. Obviously, not all of the $4 trillion would go solely to transportation. So, Freemark said, let's imagine a quarter of it went to all transportation needs, including things like electric car subsidies and public charging stations. And then a quarter of that $1 trillion went to public transportation specifically, so $250 billion overall. Freemark said this would be enough to buy about 2,500 miles of at-grade light rail service, meaning no tunnels or overpasses. Or, we could get about 400 miles of subway (these assume costs in line with the Eno Center for Transportation's cost database). By comparison, the Washington D.C. Metro has 117 miles of track and the Los Angeles light rail system has 83.6 miles of light rail. 

If we sunk all the money into intercity rail instead, that would easily be enough to finish the California high speed rail project and add a new, high-speed Northeast Corridor, Freemark said. All of these figures reflect the U.S.'s extremely high construction costs relative to international peers. We could get more if we got costs under control.

The vast amounts of money also means cities could think completely differently about what they want to build. 

"From a transportation infrastructure point of view, I think the biggest thing is that it opens up cities/transit agencies to think at a network scale rather than at the scale of single route or extension, which is how thinking is now," said Eric Goldwyn, a researcher at New York University working on a multi-year project about transportation construction costs around the world. While a few cities like Seattle and Los Angeles do have relatively comprehensive plans, they have to slow-walk them in part due to limits in federal funding. Other cities don't even try to put together big ideas that could change the way people get around, instead opting to build short stump lines through the touristy sections that are more about image enhancement than being actually useful for its residents. This one-at-a-time approach is a big departure from the metropolitan-wide thinking that dominated how San Francisco's BART, Atlanta's MARTA, and even DC's Metro systems were conceived.

In this sense, the biggest question about any prospective infrastructure stimulus is less about how much money it includes but what it is intended to do and what we want the country to look like as a result. 

"A $4 trillion reinvestment...is a huge number, and something that could transform our infrastructure system," said Carnegie Mellon University associate professor Costa Samaras. "But if we're using $4 trillion to mainly just widen highways and add overpasses, we will have a transportation system ready for the 1960s instead of for this century. And if we don't build our infrastructure to be resilient for the climate of the future instead of the climate of the past, we would be throwing away our money."

Samaras says that he'd spend the money on quadrupling the frequency of our public transportation services so they're actually useful. Electrifying vehicles. Investing in protected bike infrastructure and sidewalks. Making all public transportation accessible. Getting our water infrastructure into working order, including but hardly limited to removing lead water lines. Then there's upgrading federal buildings and public schools into working order, too, and decarbonizing them. And that's not even mentioning all of the needs for a 21st century economy like decarbonizing the grid, getting every U.S. resident broadband access, greening the freight industry, and on and on it goes. 

In fact, as Samaras and I tallied up what $4 trillion could buy, he quickly came to the conclusion it probably isn't nearly enough. If $4 trillion is a "stupendous" amount of money, as Davis from the Eno Center put it, then the U.S. has a stupendous amount of problems.