Elections Finance Infrastructure

Should the U.S. spend $1 trillion on new infrastructure?

» Donald Trump wants to make a big splash by supporting a huge new infrastructure bill. But we don’t want to end up with the construction of massive new highways from coast to coast.

After six years of proposals for significant new transportation funding being proposed by President Obama, and then being shot down immediately by intransigent Republican Congresspeople, infrastructure is suddenly the talk of Washington. Both Donald Trump and Hillary Clinton proposed major infrastructure packages during the campaign, and the Trump transition team includes a proposal for transportation investment as one of its top priorities. As I’ll describe below, this proposal would likely primarily fund transportation projects that exacerbate climate change and encourage exurban sprawl.

We must remember that the primary goal of transit advocates should not be to simply get projects built. It should be to create more livable, less carbon-intensive cities by shifting the country’s transportation system away from personal automobiles.

The nascent administration has offered few details other than a 10-page fact sheet released by a consultant in October. This proposal would release $137 billion in tax credits over ten years to private companies spending on infrastructure; in exchange, those companies would be expected to raise $167 billion in total equity, supporting a potential $1 trillion in total new construction, amounting to the largest U.S. infrastructure binge since the construction of the Interstate Highway System. In contrast to the vast majority of public works in the U.S. today, projects would have to raise user fees to pay back initial loans and corporate equity contributions.

This proposal, though not fully developed, has been widely critiqued.

Most importantly, perhaps, Ben Fried contends in Streetsblog that supporting any Trump Administration proposal is suspect; “There is no moral basis for collaboration on Trump’s infrastructure agenda—because enabling any aspect of the Trump policy platform will grease the skids for enacting the entire Trump worldview. No piece of infrastructure is worth that risk.” Whatever the benefits of new infrastructure, Fried suggests, the Trump Administration is so problematic that we must fight any of its initiatives.

From a financial perspective, the tax credits would do nothing to support projects that don’t make money, meaning all hope is lost for fixing Flint’s lead-in-water problem through this legislation, for example. The plan would provide automatic 10 percent pretax profits to contractors, just as a matter of general policy. Credits could be distributed even to projects that were already planned, meaning that there is no guarantee that the investment would actually increase investment or produce new jobs. Meanwhile, the proposal provides no source to actually pay for those tax credits, meaning they could be “weaponized” to “justify future cuts in health care, education and social programs,” as Ronald Klein has put it.

Writing in The New York Times, Paul Krugman focuses on the proposal’s secondary effect, which would be shifting infrastructure from public to private hands. This is “not a plan to borrow $1 trillion and spend it on much-needed projects,” he writes. “It is, instead, supposed to involve having private investors do the work both of raising money and building the projects—with the aid of a huge tax credit that gives them back 82 percent of the equity they put in.” Krugman’s point is that the proposed tax credits are, in reality, giveaways to companies that will then own and operate infrastructure permanently at little to no costs to themselves. The plan, in this way, is privatizing future public assets.

These are, undoubtedly, important concerns that put into question not only Mr. Trump’s motivations on the development of this infrastructure bill, but also his policy agenda on several fronts.

Beyond the financial mechanisms and the question of whether to invest in public or private construction, though, the issue of what a new infrastructure bill might fund has been little mentioned in the press. And the reality is that, for the future of the transportation system, the types of projects funded may ultimately have more of an impact than the way they’re funded.

At the heart of the infrastructure proposal is the requirement that new projects built should be self-funding in some form. All the loans that government equity-through-tax-credits are supposed to support will have to be paid back through some sort of user fee. What consequence will this have?

For one, as noted above, many of the projects funded will probably have already been planned, meaning that government funding will replace private-sector investments that would have occurred anyway. Two, this package will massively preference projects that are designed to be profitable—not projects that may serve the greater good. In such, it wouldn’t be particularly surprising for the primary beneficiaries of this legislation to be oil and gas pipelines.

In the transportation space, as many have noted, profit-motivated investments will mean, overwhelmingly, toll highways. Given that the vast majority of transit lines—perhaps all of them in the U.S.—are deficitary, and the fact that it is inconceivable to imagine developing pedestrian and biking projects that charge customers to use them, a transportation investment structured on these lines seems likely to be very auto-oriented, at least compared to current federal transportation expenditures, which are distributed 1-to-4 transit-to-highways.

But I’ll take a step further. What if, somehow, the infrastructure bill were focused on transportation and required that investments followed a similar modal ratio as they do today? If the bill guaranteed $200 billion in transit investments and $800 billion in highway spending over ten years,* would it be worth it?

I am skeptical. We must remember that the primary goal of transit advocates should not be to simply get projects built. It should be to create more livable, less carbon-intensive cities by shifting the country’s transportation system away from personal automobiles.

At the heart of the issue is that fact that new transit infrastructure is typically only moderately successful in encouraging people to get out of their cars, while new highway infrastructure is usually very good at getting people into their cars. In other words, the net effect of a tripling of the nation’s expenditures on transportation—even if those expenditures were spent in similar proportion as today—would be not a reaffirming of the status quo; it would represent a dramatic incentive to get many more people driving.

Why am I convinced that a massive increase in overall transportation expenditures would do more than just reinforce existing trends? For one, the reality is that exurban or interurban highways are simply much cheaper to build than urban transit. Take Texas’ SH 130, a four-lane toll highway that was extended by 40 miles in 2012 for a cost of a bit more than $1.3 billion.—that’s about $33 million a mile. For comparison, Denver’s 12.1-mile West light rail line, which opened in 2013, cost $707 million, or about $60 million a mile. Minneapolis’ 9.5-mile Green Line light rail, which opened in 2014, cost $957 million, or about $100 million a mile.

In other words, for the same transportation dollar, an investment in new highways can produce dozens of miles of multi-lane, grade-separated highways in rural areas while an investment in urban transit can mean relatively little new capacity.

No new investment of any sort would likely encourage better use of existing infrastructure, thereby improving the performance of existing transit lines and supporting infill development rather than greenfield construction.  

Two, given the fact that a Trump transportation bill requires projects to contribute user fees, it seems very unlikely to contribute to the reconstruction of the Interstate System, since political and public support for the tolling of existing roads is basically nonexistent. This means funds will go to new projects, not renovations.

What would be the direct consequences of thousands of new miles of grade-separated highways? Massive incentives for increased sprawling, unwalkable development, destruction of greenfield and agricultural land, and disincentives for investment in urban infill. Significantly more vehicular travel generated through induced demand. Massive new carbon emissions.

What would be the direct consequences of hundreds of new miles of new transit investments, funded through the same mechanisms? Potentially significant additional transit ridership—but maybe not; what is more likely in most cases is a redirection of riders from buses to trains. New transit-oriented development might surround many new lines, but given that most public investment is being directed to suburban highways, and the simultaneous political resistance to increased density, there likely wouldn’t be much new development from a relative perspective.**

Where does this leave us? For the good of our environment and for the good of our cities, doing nothing would likely be better than supporting this infrastructure package—even if we ignore the potentially disastrous political and financial concerns about this infrastructure bill noted by others. More transit investment simply isn’t worth it in the context of far more massive new highway spending that would overwhelm any potential benefits being derived from transit projects. Indeed, no new investment of any sort would likely encourage better use of existing infrastructure, thereby improving the performance of existing transit lines and supporting infill development rather than greenfield construction.

Ironically, losing out on the massive stream of potential federal funding doesn’t mean that new transit projects have to grind to a halt. Indeed, the results of local referenda this election reinforced the notion that when presented with a generous transit plan, voters in many cities will jump at the opportunity to tax themselves to pay for it. Indeed, the large majorities that came out to support better transit—and pay for it—in Atlanta, the Bay Area, Indianapolis, Los Angeles, Raleigh, and Seattle*** were voting for more livable, less car-dominated communities. Even though each of those plans rely on federal funding to support full implementation, even without support from Washington, much of what they propose can be completed.

Transit agencies will make the reasonable argument that they need funding to bring their systems back to a state of good repair. But this infrastructure bill won’t be of much use to address that problem, since it focuses on expansion projects. And new highways that suck ridership out of the transit system won’t help much, either.

Suggesting that new infrastructure funding isn’t worth it given what will be funded comes close to heretical in a political environment in which such investment has suddenly been identified by politicians on both the right and left as the solution for whatever it is that supposedly ails America. But we need to think long and hard about what kind of society we want before spending billions of dollars for new transportation projects all over the place.

* This would represent roughly a tripling of federal transportation spending, since the U.S. Department of Transportation currently distributes about $50 billion in surface transportation funding annually.
** Of course there are many benefits of investing in public transit projects, and many transit projects will in fact result in many additional riders. But maximizing the benefits of new transit projects—particularly in a short time horizon—is difficult because of the realities of land use.
*** In Atlanta, Los Angeles, and the Bay Area, more than 70 percents of voters expressed themselves in favor of the referenda.

18 replies on “Should the U.S. spend $1 trillion on new infrastructure?”

Excellent, thought-provoking post. Assuming that federal transportation funding remains 80/20, would you therefore advocate for federal cuts in transportation spending on the assumption that localities will pick up the slack with a more favorable (20) split? (Setting aside the implied impact on gas tax and the regressive nature of many local taxes)

Wow, almost sound like the CATO institute. Imagine, local governments paying for local infrastructure. Can you also work on these local governments up-zoning their zoning areas so people can actually afford their homes.

I don’t think that toll roads pave the way for sprawl in the same way as free roads do. There are already some toll roads. And traffic volumes on these road are fairly subdued. Developments along such corridors are likewise restrained. Large scale road tolling will in fact work against auto-driven sprawl. It does increase the cost of driving and this will inevitably force many Americans to change their transport habits.

This is a needlessly pessimistic post. Yes, requiring private investment probably isn’t going to subsidize “starter” streetcar lines that operate in mixed traffic. But there’s one big, big project category you’re overlooking.
High. Speed. Rail.
Numerous private proposals exist. There’s Texas Central’s plan to use N700 Shinkansen to connect Houston and Dallas. There’s All Aboard Florida’s plan to construct 110mph trackage in the median of Route 528. And while we all agree that the Bakersfield-Fresno-Modesto alignment being pursued by CHSRA is superior from an urban planning perspective, let us not forget that an SNCF-led consortium offered to foot the entire bill provided they could run it up the median of I-5.
Indeed, talking about how great China’s trains are (and the sorry state of ours) was an applause line in several of Trump’s speeches. And HSR fits Trump’s personality – it’s big, beautiful, fast, YUGE.
When Barack Obama proposed what was essentially the Heritage Foundation’s 1993 healthcare plan, we called out the Republicans for hypocrisy in opposing it. If we now exhibit the same hypocrisy, if we oppose even good infrastructure projects because we don’t like the guy proposing them, then we deserve every mile of outer-suburban toll highway that gets built.

HSR fits Trump’s personality except in one crucial bit: it wasn’t invented in the US. The attempts to Americanize it have led to disaster, e.g. Amtrak’s spending double on trainsets in order to allow Schumer claim credit for creating a few hundred jobs Upstate. The new FRA rules are not kind to foreign manufacturers except the big European ones, which already have plants in the US. Amtrak involvement leads to obscene costs even in infrastructure – just look at the Gateway cost trajectory. It’s legit to say “a country with 5% unemployment should take a hard look at costs and benefits before borrowing hundreds of billions of dollars to spend on infrastructure.”

Meanwhile, tech libertarians tend to hate HSR (and through Thiel, they influence Trump more than rail advocates do). HSR can’t be built by a tech entrepreneur. HSR won’t let any single individual make many billions of dollars and say “I built that.” The technology for it can be imported to the US, but the business culture for it is not American. The tech boosters talk about self-driving cars, about Uber, about Hyperloop, about literally everything that they can pretend is totally unlike anything previously built.

The problem with high speed rail in the US is that it’s also high-cost rail. Amtrak from NYC to DC is a good example. The train costs anywhere from $69 to $144 (depending on how far in advance one books) to about $220 to $280 one way for the Acela (it is literally less-expensive to fly). This does little to alleviate the use of autos, given that one can drive there for about $40 (about $16 in gas (at $2.25/gal. — easily found in NJ even after the recent tax hike) and $23.55 in tolls)! Even if one doesn’t own a car, one could actually *rent* one and drive it to DC and back for less than the Acela one-way.
Given the interest in high-speed rail by the private sector, one can assume that this reality will not change and may actually get worse. This is, at least in the US, decidedly *not* transit for the masses, exactly *because* we are talking about private-sector investment for privately-owned transit options, that will generate guaranteed profits. The push to privatize — or, as you suggest — only private projects going forward — will mean that only the wealthy will use these systems. Meanwhile, the masses will continue to seek the least-expensive options which, unless we change what gets subsidized by governments, will, unfortunately, continue to be the car.
But here, we have a perverse addition to the subsidy nut — subsidizing *private*, profit-generating projects. This is just another way to shift wealth from the poor and middle class to the wealthy, using taxes to fund privately-owned projects. But this is what privatization has always been about.

Yours is the second argument (after Alon’s) of the form “Amtrak sucks, ergo, private HSR will suck.” I don’t see it.
Texas Central will only turn a profit if it can cannibalize from SWA, presently $95 if booked in advance. All Aboard Florida thinks they can market to the family vacation crowd, which is *very* price-sensitive. Now, maybe they’re wrong. But Japan’s largest banking conglomerates seem to think it can be done, and I’m inclined to give them a shot.

You must be kidding to include 110 mph Florida HSR, when the governor turned down a a safer, more reliable, more frequent 168-185 mph system that Obama authorized grants for the first leg and was prepared to invest more for the second leg.

I agree, H(r)SR is probably the best place to place Trump’s infrastructure genius and it offers several advantages. In addition to the benefits mentioned above, don’t forget that it provides him a great opportunity to complete a project (or at least give it real momentum) that Obama failed at. Offering the opportunity to succeed where Obama failed will definitely inflate his ego and if he was to announce a national H(r)SR initiative, it could be as important as the Interstate Highway system was six decades ago. He could even put his name on it and call it the Trump train; I guarantee that a lot of people will jump on board. The train will also suck up a lot of the trillion, leaving less for sprawling highways. That’s not a bad thing. The only potential setback would be if he insists on building coal plants to power it.

I am worried about Trump’s transit policy. Already he says that he will cancel federal funds to Sanctuary Cities (those who do not cooperate with Fed. immigration operations). They are 10 major cities, including San Francisco, Oakland, Los Angeles, New York, Chicago and Washington, D.C.. If Trump goes though with his threat many transit projects could grind to a halt.

The definition of infrastructure is actually a big deal here. What is in? A giant real estate development project? A logistics center? An industrial park? If all that stuff meets some kind of federal definition for infrastructure (or “the built environment”) then does all that qualify for the generous 82% tax credit on private sector equity investment? I think the general public will be sorely disappointed in what gets built.

I really appreciate this lucid and thorough treatment of both 1st order effects, like the implementation issues that come with relying on tax incentives and negatives associated with retrenching driving as default, and 2nd order effects, long term budgetary crowding out and the privatization of future public works.

I think most urbanists (for lack of a better term, although I am referring to a pretty broad cohort here) assume that short of something revolutionary, a large scale, top down, infrastructure plan from any administration, let alone Trump’s, would have underwhelming if not counterproductive effects. So it makes me glad to be able to base that sentiment on serious analysis rather than just suspicion and bias.

I realize this is probably just beyond the scope of this post, but regarding cost, what alternatives are there besides cost/mile? I know transit systems have cost per unlinked/linked trip, which provides a nice stat that’s easy for even non-experts to understand. While everyone knows public transit doesn’t pay for itself, are there similar measures for driving? Obviously there’s an apple to oranges aspect to the comparison since those transit stats are based on operational costs, but the advantage of mass transit comes from sharing higher up-front capital costs amongst a larger cost base. I’d love to check out something that uses a dcf model across multiple reconstruction lifespans and divides by maybe average annual vehicle (ideally people) traffic.

It would be very cool for $1 Trillion to go towards fixing and “upgrading” our current infrastructure. When I say “upgrade”, I mean for the IOT aka Smart City age. As far as transit spending goes, the models from the current administration’s Smart City Challenge would be a good start. Yes…these are basically learning initiatives at this point but this huge Infrastructure Bill could ramp up the findings in other cities quickly.

These are very much public/ private networks that seek to improve and augment current transportation modes with sensors and automated cars/ vans/ buses. This has the greatest potential for making transit more palatable for suburbanites and better connect the inner city to far flung employment options. Suburbanites are getting more comfortable using Uber/ Lyft for their entertainment trips…automation will drive a further increase in this activity and hopefully include their daily commute.

As much as I like the idea of $1T infrastructure investment, it has to be done in a fiscally responsible way that also boosts productivity. We don’t need tax incentives for more toll roads to no where. Nor do we need poorly planned, gold-plated rapid transit & HSR projects.

If our Washington political cooperation were magically transported to 1973-80, but still facing the Global Warming realities of 2017 and beyond, I believe the President and Congress would agree to a Surface Transportation Bill similar to what Democratic centrist Obama originally proposed and was mostly adopted by Hillary. Their $600B plans actually make sense on many levels.

As much as I like the idea of $1T infrastructure investment, it has to be done in a fiscally responsible way that also boosts productivity. We don’t need tax incentives for more toll roads to no where. Nor do we need poorly planned, gold-plated rapid transit & HSR projects. I initially had high hopes for Trump’s plan, But there is very little long-term good for the public in his transportation plan. It appears to be a windfall for rich developers.

The Trump 10-page fact sheet also feature lies and omissions about infrastructure under Obama. It conveniently blames $10 Trillion of national debt on Obama, though he was handed the Great Recession by Bush II and had to debt-invest (economic stimulus) to lower unemployment and get the economy moving again (15+ million new jobs, 4.6% unemployment & record-breaking Wall Street).

To name another area, Trump 10-page fact sheet never mentions the shovel-ready HSR projects cancelled by Republican governors in Florida and Wisconsin. Nor that Obama’s follow-up transportation proposals were rejected by anti-anything-from-Obama Congressional Republicans, despite Democrat & Republican governors in 37 states wanting most of Obama’s transportation proposals, if only for their state.

Given the many areas of disagreement between Democrats and Republicans, there was plenty of room for Congressional Republicans to “look principled to their base” by fighting with President Obama on other matters without political obstruction to much needed, well-planned transportation infrastructure.

I appears that America will lose again.

Leave a Reply