More details about President Donald Trump’s infrastructure plan should come to light this week as he plans to share his vision for a proposed $1 trillion investment over 10 years in the nation’s roads, bridges, waterways and other structures.

Francis Fukuyama, a senior fellow at the Freeman Spogli Institute for International Studies, and Raymond Levitt, professor of engineering and director of the Stanford Global Projects Center, talk about how those funds could be used and which projects should be prioritized.

Highways intersecting

Stanford scholars Frank Fukuyama and Raymond Levitt discuss how and where federal dollars might be allocated to enhance the nation’s aging and distressed infrastructure. (Image credit: Thomas Northcut of Getty)

The American Society of Civil Engineers (ASCE) awarded the nation’s infrastructure a D+ on its recently published report card. How can the federal government best spend the new investment dollars?

Fukuyama: Infrastructure provision is primarily a state and local business – both in terms of the economic benefits and in terms of funding. Federal spending over the last 40 years has remained at roughly the same level when adjusted for inflation and most of that spending has been oriented towards new capital projects. Meanwhile, state and local spending has steadily increased over time, especially for operations and maintenance. That is not necessarily bad.

Levitt: New federal investment should be oriented towards those regional and national megaprojects that cross state lines or have national economic benefits. Other federal spending on infrastructure capital costs should be oriented towards projects that leverage significant state or local funding, and/or private investment, to maximize the benefits flowing from a new federal infrastructure program.

How can the federal government ensure the new funding is directed towards the most nationally beneficial projects or renovations?

Levitt: In a recent white paper published by the Global Projects Center, we cited Australia as having developed a unique model to prioritize federal spending on infrastructure. Historically, the national government in Australia has spent considerably less than the United States as a percent of total investment in infrastructure. To maximize the benefits of federal spending, the government created a central agency, Infrastructure Australia, as an independent, professional body to prioritize federal spending on infrastructure projects. Among other functions, the agency develops and maintains a prioritized list of key national-scope infrastructure projects and supports it with a rigorous analysis of the costs and economic, environmental and social benefits of each project.

The national government isn’t required to make investment decisions based exclusively on Infrastructure Australia’s priorities, but the program publicizes its priorities and this facilitates more rational – rather than partisan – decisions by legislators.

The administration mentioned the use of private investment, via Public-Private Partnerships (P3s), for infrastructure development. How can P3s facilitate additional investment and to what sectors or situations are they most applicable? 

Fukuyama: The U.S. relies considerably less on P3s for infrastructure projects than many other developed economies globally. The delivery model has primarily been used for major transportation projects in the U.S., but P3s have also been used for some water and wastewater projects and some social infrastructure projects such as justice complexes, city administrative buildings and military base housing.

P3s leverage private investment and expertise to complete a “life-cycle” procurement for infrastructure service delivery, rather than having the government finance and procure an infrastructure asset that will be operated – and hopefully maintained – by the government. P3 infrastructure service delivery requires the private investors not only to finance, design and construct the project, but also to operate and maintain it at preset performance standards over a long-term period–typically 30 years or longer.

Levitt: Several decades of experience in mature P3 markets such as the U.K., Australia and Canada suggest that roughly 10 to 15 percent of all infrastructure projects can be delivered via P3 concessions. The public sector – and especially the private sector – transaction costs of P3 tendering and procurement can cost tens of millions of dollars for each of the shortlisted bidders to prepare and submit a P3 proposal, primarily due to the significant design and legal costs involved. So P3 concessions can only be used economically for significant-sized projects whose estimated capital costs exceed about $100 million. Smaller projects can be bundled together and packaged into larger concessions to make them P3-eligible. This has been done for district-wide schools, statewide bridge repairs and collections of water projects.

What federal policy changes could increase the use of P3s for infrastructure in the U.S.?

Levitt: To start, the Office of Management and Budget’s scoring requirements effectively prevent some federal agencies, such as the Corps of Engineers, from being a party to a P3 contract. They effectively score all of the costs of a long-term project in the first year of the project, which is prohibitive for long-term infrastructure concession contracts like P3s. An exception to those budget scoring requirements could be made for major capital projects.

Tax exemption for privately financed infrastructure bonds would level the cost of capital for traditionally procured versus P3 infrastructure projects. This would make sense since the tax exemptions on interest payments for municipal bonds were designed to support state and local infrastructure investment irrespective of the delivery model used to procure a given project.

Fukuyama: There is potential for federal P3 incentive grant programs, which have gone a long way to develop the industry in other countries. The programs provide small grants to P3 projects developed by state and local governments if they meet certain criteria. They are beneficial for several reasons: first, they leverage small amounts of federal funding with other investments; second, they enable the federal government to set criteria for well-structured P3s – this helps ensure state and local governments follow best practices; and third, they help foster the standardization of P3 contractual terms nationally – this helps foster competition in the private sector and lowers transaction costs, especially the legal fees associated with preparing a P3 concession in different states.

The Obama administration allocated hundreds of millions of dollars to new infrastructure projects via the American Recovery and Reinvestment Act (ARRA), touting plans to allocate the funding to projects deemed “shovel ready” or that could begin within 90 days. This was not the case in practice, with President Obama later lamenting that many of those projects were not shovel ready. President Trump has remarked that he is also interested in projects that can begin quickly. Do you think “shovel ready” is a good criterion for the allocation of resources for the new program?

Levitt: In their book, Economic Policy Beyond the Headlines, George Shultz and Kenneth Dam argue that major projects are almost never truly shovel ready. They argue that using investment in infrastructure as a fiscal stimulus tends to occur too slowly to stimulate a sagging economy and that it can often spur a subsequent round of inflation. So, selecting projects that are alleged to be “shovel ready” is not necessarily a good way to allocate scarce federal funds to nationally important infrastructure projects.

Fukuyama: We argue that prioritization of federal funds for infrastructure projects should be based on two major criteria:

  • A rigorous and nonpartisan evaluation by a professional commission to assess and rank the expected economic and other benefits versus costs of each project.
  • A minimum level of participation by local government entities or private firms. Funding the project through a combination of user fees and local taxes ensures that projects that receive federal funding have strong local support versus merely political support by the area’s federal representatives.

Levitt: Some cities have begun crowdsourcing community input and crowdfunding up to 5 percent of capital costs for smaller projects, like bikeways, as a way to gauge the strength of local support for a given local project. Clearly some communities are wealthier and better able to crowdfund their preferred projects. The infrastructure prioritization process can account for this by weighing the benefits of local economic growth and job creation heavily as part of its assessment.

Part of the problem with project delays during the ARRA stimulus was that projects took a very long time navigating the planning and approval process. What policy reforms could help projects get from conceptual approval to construction more quickly?

Fukuyama: Much of the delay comes from the decentralized nature of the federal government with at least a dozen federal agencies – and often an even greater number of state and local ones – having input into the approval process. For example, the National Environmental Protection Act (NEPA) requires Environment Impact Statements (EIS) and Environmental Assessments – evaluations that detail whether a project will have significant environmental effects. These requirements are often duplicated at a state level. No one doubts the need for such evaluations, but the process has become complex and burdensome, with the average EIS taking six years to complete and often numbering thousands of pages.

Also, agency reviews are often done sequentially rather than simultaneously and there is no overall authority to resolve inter-agency disputes. Both federal and state law grants much broader standing to parties who want to block infrastructure projects than in other developed democracies. Resulting litigation delays the approval process, increases costs and scares away potential private sector finance.

Levitt: We therefore recommend a single federal “Department of Infrastructure” to be created for a fixed term of 10 years that would have the authority to both select and prioritize infrastructure projects and to oversee the approval process. This agency would also have the capacity to manage complex P3 projects, like similar offices in Australia or Canada. There may also need to be statutory changes to broad regulations like NEPA and their state-level counterparts to streamline the environmental review process.

Media Contacts

Francis Fukuyama, Freeman Spogli Institute for International Studies:

Raymond Levitt, Stanford Global Projects Center:

Mike Bennon, Stanford Global Projects Center:

Milenko Martinovich, Stanford News Service: (650) 725-9281,