President-elect Donald Trump has promised to boost infrastructure investment by $550 billion-$1 trillion because there is supposedly an “infrastructure crisis.” Never mind that the number of structurally deficient bridges has been steadily declining for 25 years while pavement roughness has been improving. If there is a transportation “infrastructure crisis,” it exists due to a failure to maintain mass transit facilities and municipal streets, neither of which can be fixed by increasing federal capital spending (as a general rule, maintenance projects and non-National Highway System road segments are ineligible for these funds). A sketch of this plan by Trump advisors Wilbur Ross and Peter Navarro can be found here. The Trump transition website touts the $550 billion figure. House Minority Leader Nancy Pelosi has pledged to “work together [with Trump] to quickly pass a robust infrastructure jobs bill.”
I won’t critique the financing specifics here, many of which are dubious. Instead, I’ll focus on the assumption underlying the assumptions: that more public infrastructure investment is a surefire way to boost the economy. Contrary to the dominant political narrative from members of both parties, which is parroted uncritically by most of the press, there is little evidence that these public works projects promote long-run economic growth.
As Bent Flyvbjerg, Nils Bruzelius, and Werner Rothengatter document in their indispensable Megaprojects and Risk, the large development effects claimed by project boosters rarely materialize (or are so small and diffuse that they are undetectable to researchers), operational viability is generally overstated, cost overruns of 50 to 100 percent are the norm, and demand forecasts are frequently off by 20 to 70 percent. These problems exist worldwide at all levels of development.
As economist Andrew M. Warner noted in a 2014 International Monetary Fund paper evaluating public infrastructure investment in developing countries, “Overall it is difficult to find a clear-cut example that fits the oft-repeated narrative of a public investment boom followed by acceleration in GDP growth. If anything the cases of clear-cut booms illustrate the opposite – major drives in the past have been followed by slumps rather than booms.”
Yet, despite the lack of evidence supporting the claim that public infrastructure investment is an important economic catalyst, there is a near-consensus among the professional commentariat that infrastructure is economic magic. In attempting (and largely failing) to critique Trump’s plan, progressive writer Jeff Spross regurgitates the Great Infrastructure Myth: “Any infrastructure plan that isn’t ultimately redistributive — that finances itself entirely from the people that use the infrastructure — isn’t going to serve its social purpose. It’s also unlikely to serve the broader purpose of reviving the national economy.”
A major problem with public infrastructure projects isn’t that the benefits aren’t redistributed to a level Spross would prefer; rather, it’s that the costs of the projects are usually completely divorced from the primary purported beneficiaries—users. (In reality, the primary beneficiaries of infrastructure projects often turn out to be the construction contractors and the politicians they support.) As noted by Flyvbjerg et al., measuring the economic benefits of infrastructure investment is already exceedingly difficult due to the diffuse nature of the benefits distribution. Spross is essentially arguing that project proponents should work harder to make these benefits even less detectable—and more illusory.
None of this is particularly controversial within the development, transportation, and urban economics research communities. The infrastructure-as-productivity-booster narrative has been around for many years—see this ungated 1994 paper by Douglas Holtz-Eakin and Amy Ellen Schwartz if you want to see how stale this “debate” is. Microeconomists who focus on infrastructure generally reject this narrative. The macroeconomists such as Larry Summers who promote the Great Infrastructure Myth are being fooled by their unreliable macro models.
Instead of promoting expensive infrastructure investment plans that are unlikely to produce their claimed benefits, Congress and President Trump should work to reduce the large misallocation of infrastructure resources and improve the efficiency of existing infrastructure networks. This would involve prioritizing maintenance over expansions, restoring the users-pay/users-benefit principle, removing regulatory barriers to construction, and adopting road pricing. Unfortunately, the Great Infrastructure Myth is unlikely to die as long as politicians place ribbon-cutting ceremonies before the truth.