Skip to main content
HC-0267[1]
Up Front

Larry Summers v. Edward Glaeser: Two Harvard economists debate increased infrastructure investments

Anna Malinovskaya and David Wessel

As politicians debate the merits of increased federal spending on infrastructure, the Hutchins Center on Fiscal and Monetary Policy asked two prominent economists—Harvard University’s Lawrence Summers and Edward Glaeser—about the economic case for stepped-up infrastructure spending and their thoughts on how to spend any additional money most wisely. Here are the highlights of the conversation.

Larry Summers: 4 reasons to drastically increase infrastructure investment

“I think there is a strong case for a substantially more ambitious national infrastructure investment program, perhaps on the order of one percent of GDP each year going forward,” Mr. Summers said.

  1. Increased infrastructure investment can lead to faster economic growth. “Infrastructure permits, in substantial part, larger interchange and reduces effective distances, thereby facilitating trade and agglomeration,” Mr. Summers said. As a result, the benefits to society almost certainly exceed the sum of all the private benefits. Moreover, in a world where private capital, private companies and ideas are increasingly mobile, a nation’s infrastructure is “distinctively local and distinctively defining of its strength.” Summers added that this is an especially good time to prioritize infrastructure investment because the return on infrastructure investment is high compared to the government borrowing rate, which is currently close to zero, adjusted for inflation.
  1. The importance of investment in maintenance is often overlooked. Infrastructure maintenance projects can have very high rates of return. “Depending on just how you do the calculation, it is between $.50 and $1 tax on gasoline that motorists pay each year in extra automobile repairs because of poorly repaired roads,” Mr. Summers said. Plus, maintenance projects can have a high social value. As Mr. Summers pointed out, fixing paint that is chipping off the walls in public schools may not yield a high return in financial terms, but it would definitely show that our society values education. Finally, there’s the fiscal prudence argument. Deferred maintenance is “a debt burden on the next generation”, which “cumulates and compounds… We have no mechanism for measuring, let alone limiting it, let alone containing it,” Mr. Summers noted. 
  1. Improved safety and option values should be included in social benefit calculations. There may be instances where investing in new infrastructure is important for safety considerations. Mr. Summers used New York’s air traffic control system as one example of a system so obsolete that only a complete overhaul of the system would ensure the highest safety standards. Plus, as Mr. Summers noted, some new infrastructure projects may create a high option value. That is, they may not be used intensively now, but could become extremely valuable in the future. “It’s hard to believe that we are not awfully grateful that people built a subway in Manhattan 100 year ago… It seems to me when we think about our obligations to posterity in a world where things become much, much more costly to do, we need to recognize the creation of those option values as a value,” said Mr. Summers.
  1. User fees may be an effective way to pay for infrastructure whereas tax credits are a bad idea. “There is no reason why people who use infrastructure more heavily should not pay for it,” Mr. Summers noted. But providing private contractors undertaking infrastructure projects with tax credits, as advocated by Mr. Trump, “would enrich those who are the recipients of the credits, in many cases for projects that would have been undertaken anyway, rather than bring about a necessary infrastructure revival,” Mr. Summers said.

Edward Glaeser: 5 reasons to proceed with caution

“[J]ust as America’s infrastructure needs its full throated advocates it also needs its skeptics, because Americans are not going to get the infrastructure they deserve unless we both propose it and then critique it and then argue over it,” Mr. Glaeser said.

  1. Infrastructure investments should be evaluated using a microeconomic approach. As Mr. Glaeser explained, “If you have a focus on jobs and macroeconomic effects, it leads to infrastructure in the wrong place. The places that need more infrastructure are largely America’s growing successful metropolitan areas, it’s not West Virginia, it’s not the rustbelt. Detroit was built for 1.85 million people; it now has less than half of that amount. It is not Detroit that needs new roads. It is San Francisco, it is New York.” Plus, a focus on macroeconomics “pushes toward the wrong funding model”, added Mr. Glaeser, as “it is an outrage that we expect voters and taxpayers in Montana to pay for the quite wealthy people who take New York’s airports.” 
  1. User fees are usually, but not always, the best funding model. Glaeser suggested that expecting users to pay market prices for public infrastructure generally will to more rigorous cost-benefit evaluations of prices and to a better allocation of infrastructure toward its highest benefit. But, he noted, in cases where people other than users benefit, we should try to come up with creative funding models. “I’m a big fan, for example, of Hong Kong’s mass transit system, which funds itself not with higher user fees, but by building skyscrapers atop new subway stops and turns a tidy profit as a result… Something that ties more closely to property tax revenues near the infrastructure to the benefits, like tax increment financing, is not crazy,” Mr. Glaeser explained.
  1. Infrastructure investment will be most effective when it is paired with the right institutions. This means, according to Professor Glaeser, privatizing some public infrastructure assets—perhaps airports—and operating some public assets jointly with private agents, for example in the form of public-private partnerships where it improves efficiency. It also means not going overboard with community involvement in the planning process so that projects can be accomplished in a reasonably timely manner.
  1. Maintaining our current infrastructure likely has the highest return, and we shouldn’t ignore it because new technologies may emerge. Glaeser notes that “Maintenance is job one, it is the most important thing” and, given the current system, it needs to be financed with public funding. But that doesn’t rule out also investing in new technologies like autonomous vehicle lanes or autonomous vehicle truck lanes, and finance these new technologies entirely with user fees.
  1. “Bus good, train bad.” Glaeser calls buses the “forgotten children of American transportation.” He explains, “there’s not a lot that you can do with a train in an urban setting that you can’t do if you have a bus on a dedicated lane… The beauty of the bus…is that it’s flexible.  You can change the technology as the technology improves, you can make it autonomous, you can reroute the thing.”

More

Get daily updates from Brookings