A National Infrastructure Bank Can Provide Important Benefits If Mission and Scope Are Defined Narrowly

Prepared for presentation to:
Subcommittee on Select Revenue Measures
U.S. House Committee on Ways and Means
1100 Longworth House Office Building
Washington, D.C.

Chairman Neal, Ranking Member Tiberi, members of the Subcommittee, thank you for this opportunity to discuss the potential benefits and limits of a national infrastructure bank (NIB) in meeting U.S. infrastructure needs. I am Sam Staley, Robert W. Galvin Fellow and Director of Urban and Land Use Policy for Reason Foundation, a non-profit think tank that advances the practical application of free market principles to improve the quality of life for all Americans.

We are at a crossroads on how to fund long-term investments in the nation’s roads, bridges, ports, airports, water, sewer, and other infrastructure. Our growing metropolitan areas need significant investments in major physical infrastructure, whether a toll tunnel in Atlanta, freight rail modernization in Chicago, or the expansion of port capacity in Los Angeles. Many of our older metropolitan areas need to upgrade or retrofit existing infrastructure at levels that require hundreds of millions of dollars. An infrastructure bank, if properly defined and implemented, can provide a meaningful role in meeting these needs.

However, a NIB may also side track important national priorities and undermine economic competitiveness if its mission and programs are not properly defined and implemented. My testimony will focus on the characteristics and criteria necessary for ensuring an infrastructure bank works effectively and avoids the inefficiencies inherent in a politicized funding environment.

My testimony today will focus largely on the implications of a NIB on transportation infrastructure, but I think these comments are equally applicable to other forms of infrastructure including water, sewer, stormwater management and other utilities.

1. What is an Infrastructure Bank?

In an ideal world, the private sector would be able to identify water, sewer, utility, roads, bridges, and other physical infrastructure projects necessary to meet key community need, tap private capital markets to finance them, and fund them through market-based user fees. In practice, this approach has been outside the U.S. framework for funding and financing core infrastructure although the extensive use of public-private partnerships in other countries (e.g., China, Australia, France, and elsewhere) have brought many countries closer to a market-driven environment. Instead, we have relied on a system of government grants and ad hoc funding measures to bridge an increasingly large infrastructure funding gap. In the absence of effective markets for large, capital intensive projects, infrastructure banks have stepped in an attempt to fill some of this gap. The question now before the U.S. Congress is whether the nation needs to create a national infrastructure bank to take us a step closer to meeting our core public service needs and adopt new ways of thinking about and financing these projects.

Serious proposals for some form of a national infrastructure bank emerged midway through the last decade as the gap between infrastructure needs and revenues became increasingly obvious. Current audits of the nation’s roadways alone suggest our annual spending falls about $27 billion short of funding levels needed to simply keep our existing transportation network in a state of good repair.

While the nation’s current transportation funding system has a source of revenue-the federal gas tax-for the short-term, a porous political process and rigid distribution formulas too often misallocate funding away from key projects. (The last transportation bill, for example, included thousands of discretionary earmarks.) Government budgeting hampers strategic decisionmaking since funding for long-term projects is contingent on annual revenue distributions. Thus, a long-term capacity need may not be met because in any given year annual revenues might not be sufficient to finance the project. This is a problem endemic at the state level as well as the national level, as the public concern over lagging reauthorization of federal transportation funding showed at the end of 2009.

A national infrastructure bank could provide an objective and transparent framework for filling the large gaps in the nation’s physical infrastructure as long as its mission is defined narrowly and decisions on infrastructure investments are politically independent and transparent. In theory, a NIB would allow infrastructure decisions to be made in an environment more closely resembling a private sector bank since it is more suitable to matching investments with long-term rates of return and other benefits. The key is to provide a rigorous framework for attempts to allocate increasingly scarce federal funds to their highest, best, and most productive use based on objective assessments of potential public benefits against costs.

If an infrastructure bank, however, is used to simply consolidate existing grant programs, these benefits are not likely to materialize. By their nature, loans to public and private customers are based on expectations of paying the bank back. They include assessments of uncertainty and risk that balance the costs of providing capital to fund long-term investments. In principle, these revenues should become part of a “revolving” fund, in which revenues from previous loans are used to underwrite future investments.

At least in principle, physical infrastructure projects lend themselves to this type of financing because they provide long-term benefits, assets can be depreciated over long-periods of time, and their values remain relatively stable. The key is to ensure that a sustainable and stable revenue stream is tied to the project to finance the loan. In many countries, these revenues are tied to user fees (e.g., tolls, water usage fees, wastewater volume, etc.) or explicit commitments of future tax revenue to paying off the bonds used to pay for the facilities. On the local level, user fees are common for water and sewer systems in the U.S. Tolls on the other hand, are a very small source of funds for U.S. roads despite their widespread use in nations ranging from China to France. Nevertheless, their potential to become a significant source of funds in U.S. urban areas is substantial.

Of course, debt is not a substitute for revenue; it merely changes the timing of revenue payments allowing large amounts borrowed up front to finance a bridge, road, or sewer plant to be spread out over the economic life of the facility. In the case of highways, tunnels, and bridges, the typical economic “life” of a project or asset typically extends to 50 years or more. Water, sewer and other public-health infrastructure may have shorter life spans depending on advances in technology and changing health (and environmental) standards.

Grants, in contrast, work on fundamentally different principles. A grant reflects a one-time payment without an expectation of repayment. Grants often carry more risk to the taxpayer and government because once the funds are dispersed they are difficult to reclaim if the project fails. Grants, for example, do not have revenue streams tied to the project since the nature of the grant is to provide free and clear funding without future financial obligations. While an infrastructure bank might have responsibilities for making grants of some types, grants are not loans and should not be confused with the fundamental nature and character of a bank which is to incur debt and loan funds to underwrite the construction and sometimes maintenance of infrastructure facilities.

Thus, a national infrastructure bank should be thought of primarily as a lender of public funds, not a grant maker. To the extent that a national infrastructure bank might include grant making responsibilities, explicit rules and regulations should be adopted that recognize the fundamental differences between the financing instruments, the criteria needed to evaluate proposals, and evaluating the projects receiving the funds. And whenever possible, grants should be tied to loans to provide a greater level of accountability for the projects.

Fundamentally, however, a NIB should be constituted to issue debt, not grants, and use the proceeds from these loans to underwrite capital projects.

1. Potential Benefits of an Infrastructure Bank

A well structured infrastructure bank could have several potential advantages over the current system if structured properly and its scope sufficiently narrow to avoid political abuse and mismanagement. Potential benefits include;

• Gap financing. The raison d’être of a government loan is to make up the difference between existing revenues and the amount required to underwrite the project. When the private sector is either unwilling or unable to fund a project and the project has significant public benefits-it is a “public good” in the terms of economists-public financing might be justified. Indeed, if financing can be obtained from the private sector (and the U.S. has yet to tap much of the private capital available for infrastructure investments), no compelling reason exists for putting taxpayers at risk. At times, public funding can be useful to bridge the period between funding the construction of facility and the time revenues come on line (e.g., a toll road or bridge).
• Transparency. Loans could be made (and debt issued) through a consolidated entity, avoiding the “alphabet soup” of agencies and programs that currently provide funds through a diverse array of government departments. By consolidating national loan programs, loan (and grant) success and failure can be tracked and coordinated more easily. This facilitates audits of project performance as well as providing public accountability. Efficiencies might also be achieved by consolidating different loan programs for different types of infrastructure under one entity so that financing staff and expertise can be shared among organizations. However, there are limits to this type of aggregation when projects are sufficiently diverse to warrant different evaluative procedures. The decision to consolidate programs must be made carefully.
• Accelerated funding. A key goal of an NIB would also be to more effectively align funding with project benefits (and revenues). Banks are able to make decisions that span decades because they have a long-term focus. Their decisions also tend to be more strategic than focused on more temporary conditions. Currently, most governments allocate funds to specific projects on an annual basis. In some cases, state governments have established five- or ten-year infrastructure plans, but this planning fails to adequately capture the need to finance and manage projects that extend over decades. Project commitments are driven by annual budget decisions and limited by current-year tax revenues. As a result, projects can be assessed and approved more efficiently in a bank-like setting, allowing state and local governments to commit to long-term projects more quickly based on expectations of future revenue (either through user fees or tax revenues).
• Improved strategic decisionmaking. By adopting performance criteria for loans, whether rate of return or objective measurements of public benefits, an infrastructure bank might help maximize public benefits. Loans have an inherent advantage over grants in that a “willingness to pay” criteria helps ensure benefits are aligned with costs. This is the case with true user fees, such as tolls or water rates that are priced to reflect the actual costs of using the facility. For toll roads, for example, the willingness of drivers to pay a toll is essential for securing the loans necessary to build the facility. Risk is not eliminated, but user based pricing allows state agencies and private companies to gauge the priority and importance of different projects based on the response of users. Unless the public (through taxation) or customers (through user fees) are willing to cover the costs of the loan (borrowing plus debt service), the project may not be justified or qualify for funding through a NIB because the users are simply not willing to pay for the projects-although there may be public benefits that still justify funding the project.
• Diversifying revenue streams. Since a primary criterion for making loans involves ensuring a revenue stream exists to pay the loan back, a NIB can encourage state and local governments to identify and implement more diversified revenue streams. As a practical matter, this would include tolls for roads and bridges or quantity-based usage fees for water and sewer systems. Because user fees tie costs to specific benefits, they can be both practical and sustainable alternatives for raising revenue compared to general taxes (which also often have sunset provisions).

1. Potential Limits and pitfalls

Theory, however, doesn’t always square with reality.

First, an infrastructure bank is not a substitute for other forms of public funding. On the contrary, the debt issued by the bank is a liability for the agency, not an asset. At best, an infrastructure bank can supplement federal funding for infrastructure by leveraging projects through borrowing, but loans or government guaranteed bonds do not supplant federal funding. The potential benefit is in realigning financing costs with more realistic timetables for paying off the loans (or bonds). Debt must be paid back, either through future tax revenues or through user fees, so the funds are not “free” in an administrative or economic sense.

Second, public debt is also not issued in a vacuum. It must compete for private dollars in a global market place. If a NIB lends $1 billion for a new road, those funds are diverted from the private sector, either from the general public or private investment funds. Issuing too much debt, which often occurs at lower interest rates because of the implicit government guarantee, or funding projects with few benefits, will crowd out private investment in other parts of the economy that may be more productive. Debt is not a free fiscal lunch.

Third, a poorly structured infrastructure bank runs the risk of political manipulation, much like what was experienced with Fannie Mae and Freddie Mac when best business practices for mortgage lending were subverted by political goals and objectives. For those projects that can be funded through user fees, this risk of political manipulation is relatively small because the criteria are straightforward: users value the project at sufficiently high levels they will pay for the entire project. The loan simply bridges the gap between the funds needed to construct the facility and the time the facility generates revenues to pay for the cost. For projects not fully funded by user fees, the risk is higher because achieving social goals becomes an important justification for public financing of a project.

Thus, absent self-funding user fee projects, infrastructure banks are susceptible to making investment decisions that are highly distorted by political interference. This is particularly true if its mission is interpreted broadly, or the criteria for providing the loans or grants are loose, poorly defined, or hard to measure. Loans must be made using objective criteria directly linked to the benefit expected. Investments in roads, for example, should have meaningful impacts on mobility and travel times in order to justify loans or other bank investments.

Fourth, an unanticipated outcome of a NIB might be to weaken the authority of state and local governments in setting policy and investment priorities. This might be more likely if a NIB is established without a clear national or federal project mandate incorporated into its mission and purpose. Currently, states and local governments are given deference in funding since they are often in the best position to evaluate the potential benefits of infrastructure investments. A NIB that has wide discretionary authority over funding may well undermine this implicit recognition of the efficiencies provided by local knowledge of needs and requirements.

While a national infrastructure bank has several potential benefits, policymakers run the risk of asking the bank to do “too much” or create a management organization that is simply incapable of efficiently and objectively evaluating the viability of projects or their public benefits. Infrastructure banks must have a clearly defined role and its activities must be directly tied to specific activities and projects that have measurable outcomes.

1. Characteristics of a High-Performing Infrastructure Bank

If a national infrastructure bank were established by Congress, what would its fundamental characteristics look like?

• Independence. Infrastructure banks must be as insulated as possible from political manipulation to be effective. This requires a management structure that is independent of the day-to-day policy concerns of Congress and the White House and a management structure that is focused on a bottom line with a clear bottom line to judge success and failure. In practice this will be very difficult to achieve, but it should still be an important goal.
• Objective loan criteria. Bank viability is rooted in sound loan management, and the same criteria should be applied to government funded infrastructure banks. In the cases where user fees are not fully capable of covering the costs of the loan, performance criteria must be tied to the loan agreement to ensure public benefits are maximized. Moreover, these benefits must be measurable, directly tied to the project, and objectively evaluated. For example, a new road should significantly improve travel times, increase mobility or reduce congestion. A new water treatment plant should improve public-health outcomes. In contrast, general social goals and planning objectives such as improving “livability” or “enhance quality of life” are difficult to measure and evaluate, leading to inefficiency and ineffective grant making.
• Well-defined mission. The infrastructure bank should not be seen as a catch all for funding for public projects. The bank should have a clearly defined mission that constrains the types of loans and grants it can make. The NIB should not be considered a source of “free” money, or become a slush fund for favored projects. A NIB should be limited to making loans for bona fide physical infrastructure projects, and these projects should have measurable outcomes tied to them.
• Well-defined federal role. In the case of a national infrastructure bank, projects must have a clear federal priority and justification, either because the project is of national importance or the project involves interstate or international cooperation beyond the scope of state and local governments.
• Loans are restricted to capital projects. A fundamental principle of public and private finance is that debt should not be issued to cover ongoing operations and maintenance. Stable, steady revenues should be used to offset these expenses. Loans and their associated debt are used to finance long-term capital projects.
• Loans require sustainable revenue sources. All projects selected for funding should have sustainable revenues sources to ensure the loan will be paid back in a timely way. As mentioned earlier, these revenue sources could include dedicated tax revenues although user fees would probably provide a more reliable, stable, and sustainable source. This is crucial for sustaining a NIB since it also protects the viability of the revolving loan function of the bank.

Many of these criteria are summarized in the table below .

Unfortunately, the current infrastructure bank proposals before the Congress fall short on many of these criteria. The most detailed proposal, H.R. 2521, the “National Infrastructure Development Bank Act of 2009,” envisions a complex and diffused management structure that includes 9 executive officers appointed, fired, and compensated by a five member Board appointed by the President of the United States (with the advice and consent of the Senate). The Board also appoints two standing committees that include four additional members each to a Risk Management and separate Audit Committee. The criteria for qualifying loans are extremely porous, including criteria that are more appropriately classified as social goals. Presumably projects that meet these social goals, which include workforce training, reducing poverty, job creation, and Smart Growth, would qualify for funding even if they do not provide adequate or efficient physical infrastructure. Indeed, these goals have little application to providing efficient or productive infrastructure, reflecting political considerations and policy tradeoffs.

Criteria for Determining Eligibility of Projects for Funding From a National Infrastructure Bank	Should the Project Be Funded?
Selected Key Criteria	Yes	No
The project is national in scope	ü
The project crosses state boundaries	ü
The project is international in scope or impact	ü
The project provides significant national or multi-state public benefits	ü
User fees can fund the project	ü
Private financing is unavailable	ü
Private equity and funding sources have been tapped	ü
The project is too large to be funded by state or regional governments	ü
Federal funding provides “bridge” financing	ü
Federal assistance pre-empts state or local funding		X
The project fills an investment gap	ü
Sustainable revenue sources are unavailable		X
Project funds operations or ongoing maintenance		X
State and local funding sources have been exhausted	ü
Loan is tied to specific performance criteria (unless reimbursed by user fees)	ü

The White House’s proposal to create a National Infrastructure Innovation and Finance Fund (I-Fund) is less well developed, so specific comments on its operation, mission, and potential programs are speculative at best. For example, it’s unclear how nesting the I-Fund in the US DOT will create the independence necessary to follow through on a rigorous and objective analytical process, or what criteria will be used to determine the merits of varying infrastructure projects. The primary objective of the I-Fund appears to be consolidating federal programs that fund various forms of infrastructure (breaking down “silos”). While consolidation may have value, a national infrastructure bank would need to have clear criteria for assessing risk and the potential rate of return for investments in different projects. In fact, one possible outcome of consolidating federal funding programs might be less accountability, as a rigorous evaluation of investments in different infrastructure projects becomes difficult to assess without clear objectives or performance criteria.

1. Conclusions

On the surface, the creation of a national infrastructure bank is an attractive option for the federal government. The possibility of consolidating programs, putting government loan programs on a more objective basis for evaluating performance, and streamlining the approval process for key infrastructure facilities holds promise.

These potential benefits must be balanced with a clear understanding of the limits of infrastructure banks and the forces that could lead to even more inefficiency and political manipulation. Infrastructure banks work best when they have a clear, focused mission, their operations focus on loans, and transparent performance criteria allow their operations to be monitored, evaluated, and held accountable.

Thank you for your attention and this opportunity to address the Subcommittee on this important national issue.