If there’s one issue that Democrats and Republicans seem to agree on, it’s that America’s infrastructure is in need of an upgrade. Many of the nation’s highways and bridges are deficient; its airports and waterways are clogged; its power lines and levees are in disrepair.
Both presidential candidates say it’s time for a fix. Donald Trump and Hillary Clinton have each promised to boost infrastructure spending by hundreds of billions of dollars, which would dwarf the investments made under President Obama’s 2009 stimulus bill. And it isn’t just politicians: Many economists also think infrastructure should be a priority.
But peek beneath the surface and that apparent consensus disappears. Democrats, for example, want to borrow money to pay for new investments in infrastructure, while Republicans would rather cut spending elsewhere or rely on the private sector. The real difference, though, is less between left and right than between politicians and economists. Politicians have an incentive to favor projects that provide a quick infusion of jobs that they can point to in their next re-election campaign; economists are more concerned about the long term. Politicians tend to favor impressive new projects they can put their names on; many economists would rather focus on repairing and upgrading existing infrastructure. And there are also disagreements over how much to spend on infrastructure projects and how to pay for them.
Still, infrastructure is a rare policy area in which action seems possible under the next president and Congress. Here are four questions that will help determine whether an infrastructure deal is good economics as well as good politics.
- Short term or long term?
There are several arguments for why we should improve America’s infrastructure, and they sometimes get conflated in the public debate. Some policymakers, particularly on the left, see infrastructure investment primarily as a jobs program, like the New Deal’s Works Progress Administration. In this framing, the goal of infrastructure spending is to provide a short-term stimulus that jump-starts the economy — more jobs lead to more spending, which leads to more hiring, and so on.
Many economists embraced that logic during the depths of the recession in 2008-09 and in the early stages of the recovery, when the economy badly needed a boost. But now, with the unemployment rate at 4.9 percent, the need for an immediate stimulus is less clear. Some economists, especially conservative ones, are skeptical that government spending does much to boost the economy outside of recessions.
“The public discussion cherry-picks examples — ‘look at the internet, look at the Hoover Dam,’” said Andrew Warner, an economist at the International Monetary Fund who has studied large infrastructure booms in other nations. “It’s just not obvious that past infrastructure surges have had these big effects as advertised.”
Instead, many economists see infrastructure spending as a long-term investment that enhances productivity. By eliminating bottlenecks in energy, transportation and communications, they say, an infrastructure-spending boost would make the whole economy more efficient, which in turn would let it grow more quickly. (In economics jargon, it would raise the economy’s potential growth rate, which has been lagging.) In this framing, the economic returns to upgrading our infrastructure won’t materialize quickly, but will be realized over decades. The Congressional Budget Office recently estimated that it often takes 20 years for the full effects of federal infrastructure investment to be felt.
“I think of it as a long-term economic growth issue,” said Keith Hennessey, a lecturer at Stanford and the former head of President George W. Bush’s National Economic Council. “If it increases short-term employment, great — but that’s not as important.”
Short-term employment is, however, important to politicians, who want results that will help them win re-election. That difference in emphasis can have practical implications: It gives politicians an incentive to push for projects that create the most jobs quickly, rather than ones that offer the best payoff in the long term.
Not all economists see a tension between short-term and long-term goals. Larry Summers, a Harvard economist who has become a prominent advocate of increased infrastructure spending, said that part of what makes such investments attractive is that they offer both immediate and longer-lasting benefits.
- Build new projects or repair old ones?
Another decision that policymakers face is whether to focus on building new projects or maintaining old ones. The nation’s roads, bridges and airports have a massive backlog of deferred maintenance. Many economists see that backlog as a top infrastructure priority: A 2011 report from the Brookings Institution recommended redirecting all revenue from the federal gasoline tax to “repair, maintain, rehabilitate, reconstruct, and enhance existing roads and bridges.”
Economists like spending on repairs and maintenance because it’s cheaper in the long run to maintain existing infrastructure than to let it deteriorate and be forced to start over. Maintenance also takes out the guesswork — we already know which bridges and highways are the most valuable. “One of the benefits of an emphasis on deferred maintenance is that it’s done on the infrastructure that’s heavily used,” said Summers, who served in the Clinton and Obama administrations.
But politicians might not play along. A new bridge or expressway is more marketable than filling potholes — hence the “bridges to nowhere” that occasionally make headlines. Some research has suggested that political incentives lead governments to direct resources to overly expensive or unnecessary projects and to underinvest in maintenance.
In the end, the new president and Congress may not have much control over which projects are chosen. A large share of federal infrastructure spending is executed by block grants to state and local governments, which decide how the money gets spent.
Some economists consider that a good thing. “Let them spend it on what they consider most important,” said Sherle Schwenninger, the founder of the World Economic Roundtable and a director at New America, a left-leaning think tank. “There will be some waste and pet projects, but a lot of them will start repairing potholes and replacing water pipes, moving power lines underground, etc.”
- How big?
Perhaps no issue is more contentious than the scale of an infrastructure package. The topline numbers vary a lot across proposals. Clinton’s plan calls for $250 billion in direct federal spending over five years (with another $25 billion allotted to the creation of a national infrastructure bank — more on that later).1 Bernie Sanders called for a much bigger plan during his primary campaign: $1 trillion over five years. And Trump has said he would “at least double” Clinton’s package.2 Republicans in Congress, meanwhile, aren’t exactly happy with this bidding war, so if the GOP retains control of Congress, it’s not clear how big a package the two parties will be able to agree on.
But how much should it be? Summers said one way to answer that question is to look at total public infrastructure investment (at all levels of government) as a share of the overall economy. By that measure, it looks like the U.S. has been underinvesting in its infrastructure for decades. Gross government investment in structures — taken here as a close approximation to infrastructure3 — has fallen from a peak of around 3.5 percent of gross domestic product in the late 1950s to 1.5 percent in mid-2016. That is nearly the lowest mark in nearly 70 years. (The historical average is about 2.3 percent of GDP.)
Summers recommends a big boost to infrastructure spending to make up for what he considers years of underinvestment. In an interview, he suggested the U.S. should spend an additional 1 percent of GDP per year for 10 years, in addition to the roughly 1.5 percent of GDP the U.S. spends now. As a result, total spending over the next decade would rise by around $2 trillion — more than either candidate is proposing.
Some experts put the price tag even higher. The American Society of Civil Engineers recommends $3.6 trillion in investment by 2020. ASCE is hardly unbiased: Its members would benefit from increased infrastructure investment. Asked about this potential conflict of interest, Brian Pallasch, the managing director of government relations for ASCE, said: “Engineers are in the best position to comment on this; obviously we know a little bit more about it. You’re not going to a lawyer for a health care checkup.” Although this number may cause sticker shock for some, the Economic Policy Institute, a left-leaning think tank, took ASCE’s “infrastructure gap” seriously enough to outline a scenario of investment to close it.
- How do we pay for it?
Historically, most government infrastructure projects at both the state and federal level have been paid for with borrowed money, via government bonds. Economists often warn about the risks of too much debt. But in this case, some economists see borrowing as an argument in favor of infrastructure spending, not against it.
This argument is a bit technical, but the short version is this: The U.S. and other rich nations have been stuck in a prolonged period of low economic growth. That pattern, and the low interest rates that come with it, have discouraged private investment — why build a new factory or invest in a new project if you won’t get a decent return on your investment? Infrastructure spending, because it relies on long-term borrowing, would probably drive up interest rates, encouraging more private investment while improving the long-term productive capacity of the U.S. economy. Infrastructure investment, Summers said, “enables us to have more financial stability with the same level of demand at a higher interest rate.”
Not everyone buys Summers’s theory. But even if he is wrong, low interest rates mean that it is remarkably cheap for the U.S. to borrow money right now. That means the government can afford to borrow more than it could during a period of higher interest rates. (Of course, interest rates could always rise.)
Many politicians from both parties, however, are wary of increasing the national debt. Clinton has pledged repeatedly not to “add a penny” to the debt. Trump has likewise pledged not to increase the debt; his advisers argue his plan to use tax credits to spur private infrastructure spending would pay for itself. (Many outside experts are skeptical.)
Democrats have proposed an alternative financing option called a national infrastructure bank. It would work like this: The federal government provides some initial capital — in the billions of dollars — to establish the bank, which then leverages that money to finance infrastructure projects (done by the public or private sector) at many times that amount. The idea was a key piece of President Obama’s 2014 infrastructure proposal that stalled. Clinton’s plan is to leverage $25 billion in federal funding to enable lending at nearly 10 times that amount.
But Republicans are likely to oppose such a body. Hennessey, the former Bush advisor, called the idea “terrible.” He thinks an infrastructure bank would become “captured” by the interests that would benefit from government subsidies, leading it to make poor decisions; he cited the housing giants Fannie Mae and Freddie Mac as examples of how it could go wrong.