Skip to main content
America May Finally Be Ready To Fix Its Infrastructure. Too Bad The Timing Stinks.

Decades of underinvestment have left America with aging airports, buckling bridges, stalled subways — a veritable plague of inadequate infrastructure. On this, President Trump and his Democratic opponents in Congress actually agree. Which is why there is some hope for Trump’s appeal, during the State of the Union address, asking “both parties to come together to give us the safe, fast, reliable and modern infrastructure our economy needs and our people deserve.”

But fixing America’s infrastructure will cost trillions, which is one reason Washington has been kicking this can down the unpaved road for years. And while inking an infrastructure deal is tricky under the best of circumstances, now is a particularly bad time — because the economy is just too strong.

Costly repairs make more sense when the economy is faltering and Americans are desperate for work. In that environment, infrastructure spending has a supersized impact: Not only does it improve America’s bridges and transit systems, but it also provides jobs for people whose skills might otherwise go to waste.

However, in today’s economic climate, where unemployment is nearing a 50-year low, even a massive infrastructure bill would likely generate only a trivial number of new jobs. Instead, the government would have to fill its construction crews by poaching private-sector workers, which could potentially create an inflation-generating war for scarce workers and neutralize many of the economic benefits commonly associated with large-scale government spending.

Of course, that still leaves one perfectly good reason to support such spending: It improves America’s crumbling infrastructure. But skeptics can point to the economic risks as a reason to say, yet again, “Let’s fix our infrastructure some other time.”

When the economy is strong, government spending doesn’t pay off

At the nadir of the Great Recession, when home prices were collapsing and unemployment was spiking, infrastructure projects came with a free lunch. Every dollar of direct government spending could generate roughly $1.50 in economic activity.1

But one of the keys to this windfall was high unemployment. When the federal government hires an unemployed worker, it sets off a chain reaction of benefits. Not only does the worker suddenly get a paycheck to spend at local stores, but the store owners get more money, too, which they can use to invest in their business, hire more workers or otherwise pay forward in this cascade of useful spending.

By contrast, when the government hires people who already have jobs, there’s no real infusion of money, just a different name atop the paychecks. And that seems to be the situation that government programs face at the dawn 2018.

Jason Furman, who served as a top economic advisor to President Obama, estimates that even a substantial infrastructure program would create approximately zero new jobs. Along the same lines, Harvard economist Gabriel Chodorow-Reich thinks each dollar of infrastructure spending in today’s economy would likely produce less than $1 in economic output in the short term because the benefits of government spending are more than offset by the costs of taking that money out of the hands of the high-performing private sector, whether through taxes or borrowing.

It’s not quite enough to say that this is because unemployment is nearing historic lows. Plenty of economists insist that there are still underutilized workers, pointing to alternative data like middling wage growth or the still-underperforming share of 25- to 54-year-olds with jobs.

What’s critical, though, is that the Federal Reserve disagrees. According to its published projections, Fed members think the current unemployment rate is already below its stable, long-term level of roughly 4.6 percent. Any infrastructure plan that threatened to push the unemployment rate further below this sustainable level would only exacerbate concerns some Fed members have already expressed that the economy risks overheating. And that would likely bring faster interest rate hikes explicitly designed to slow the economy and stabilize unemployment.

More infrastructure spending means less spending on something else

Even without job creation, pouring money into infrastructure could still prove transformative — though not necessarily in a good way.

By setting its own spending priorities, the federal government would essentially be pulling resources away from activities preferred by businesses and individuals. Construction crews may end up heading to highway work sites, for example, instead of to residential home-building jobs, or engineers may find themselves working on subway projects instead of private-sector research and development.

Whether that’s a useful shift depends on the relative value of these various activities. But there’s no avoiding the trade-offs. And that’s true regardless of how the government funds its infrastructure plans, whether through direct spending or by partnering with states and private businesses, as the Trump administration seems to prefer.

That is not to suggest that such policy details are irrelevant: Effective public-private partnerships require a tricky balance between allowing private companies to profit and ensuring that rich and poor alike are given fair access to the resulting infrastructure; new government borrowing could make it more expensive for private companies to borrow — which would mean fewer productive investments.

But in both these approaches, the only way to ultimately break ground on new infrastructure projects is to divert people and machines away from their current use — because in an economy at or around full employment, there isn’t a reserve army to draw on.

Now, this competition for limited resources isn’t all bad, not for workers anyway. If the federal government has to vie with U.S. companies to hire qualified workers, it could start a bidding war, driving up pay and benefits. This is far from guaranteed, given how sluggish wage growth has been despite our tight labor market, but it can’t be ruled out.

But if pay did start to rise, odds are inflation would, too — which could prove self-defeating. Rising inflation would likely trigger an aggressive response from the Fed, which has an explicit mandate to keep inflation under control. That would mean faster interest-rate hikes as part of a concerted effort to blunt inflation and moderate those wage gains.

Infrastructure has to be done carefully

Even with all these economic risks, America still needs repair. One assessment, from a nonprofit launched by the G-20 group of major economies, found that the U.S. would have to increase spending by $1.2 trillion over 10 years to address the backlog of infrastructure needs. The American Society of Civil Engineers has put that number at $2 trillion.

Infrastructure spending comes with well-established, long-term benefits: Improved roadways and bridges make it easier for businesses to transport goods, well-designed school buildings may enhance educational opportunities, and transit systems help workers travel more efficiently to find the jobs best-suited to their skills.

It might even be possible to design an infrastructure program for those groups that are still missing out on economic gains. A commitment to remove lead contamination from struggling towns like Flint, Michigan, could create local jobs and make the communities more appealing to newcomers. And leaked reports suggest the Trump administration would dedicate a substantial share of available funding to rural projects.2

But even a targeted plan could bump up against some significant constraints. Traditionally, for instance, infrastructure spending has provided a particular boon to workers without a college degree — because the jobs these programs provide are tilted toward blue-collar fields like construction and transportation. But employers in those industries are already struggling to find qualified workers, according to a recent analysis from the Federal Reserve. And this is reflected in the fact that the strongest wage gains of the past year have actually gone to low-wage workers and those without a college degree. The arrival of hefty government contracts seems as likely to exacerbate the shortage as it is to create new opportunities.

And then there are the political implications. Congressmen from thriving, urban districts might not love the idea of diverting funds to other parts of the country if it means their constituents get a smaller piece of the funding pie. And that could sink the quest for the bipartisan consensus needed to avoid a filibuster and pass a substantial spending bill.

With political and economic risks like these, it’s harder to argue that now is the time for infrastructure investment — and it wasn’t an easy sell to start with. Maybe the backlog of repairs has finally gotten long enough that politicians can find common ground, not necessarily because it will help the economy, but because they see functioning infrastructure as vital to America’s future.

Or we can put the challenge off yet again.


  1. Exact estimates of this effect vary. The $1.50 number stems from 2009 projection by the White House Council of Economic Advisors, and it has since been reaffirmed by independent economists. The Congressional Budget Office used an average estimate of $1.80.

  2. Regional targeting was incorporated in the New Deal, which made some efforts to align spending with need.

Evan Horowitz writes the “Quick Study” column for the Boston Globe.