Reply

Spring 2021 • Regulation
By Eduardo Engel, Ronald Fischer, and Alexander Galetovic

Bob Poole has been a longstanding and influential supporter of public‐​private partnership (P3) highway projects in the United States and we appreciate his interest in our article. His main concern is that our conclusions do not apply to the United States. To the contrary, we argue below that they do apply.

Poole claims that Present Value of Revenue (PVR) contracts are not suitable for the United States because they reduce the demand risk borne by the P3. This would be perceived in the United States as crony capitalism. Bearing demand risk is desirable because “the concession will do a superb job attracting toll‐​paying customers and delivering high value to them.”

Recall however, that in a PVR auction of a P3 highway, firms bid on the present value of toll revenues and the lowest bid wins. Because the concessionaire is selected in a competitive auction, it will receive a fair compensation and the process cannot be described as crony capitalism.

PVR auctions reduce the demand risk facing the concessionaire because the P3 lasts until the bid is collected. Having firms bear demand risk is costly. A firm that bears demand risk requires a higher return to accept a riskier business. Paying a premium is worthwhile if the P3 can manage those risks, such as construction, operation, and maintenance cost risks. But it is wrong to ask the P3 to bear risks it cannot control because then its only response is to ask for a higher return at the expense of users. Because highway demand depends largely on factors beyond the P3’s control — such as the overall state of the economy, the financial crisis or, more recently, the COVID-19 pandemic — it does not make sense to have the concessionaire bear demand risk.

Poole argues that in the United States those highway P3s that overestimated demand (most of them, it turns out) went bankrupt and were not bailed out by renegotiations. It is positive that losses remain private and are not socialized, but an infrastructure system based on occasional huge returns and frequent heavy losses leads, in the best of cases, to expensive finance, with the costs borne by users. Furthermore, as Poole himself pointed out in his 2017 Reason Foundation study “Availability Payment or Revenue‐​Risk P3 Concessions? Pros and Cons for Highway Infrastructure,” the fact that construction companies “are leery of, or are simply unwilling to accept, traffic and revenue risk” explains why, since 2009, half of all new highway P3s in the United States have been availability payment (AP) concessions.

Moreover, contrary to Poole’s assertion that in “the United States there is no history of renegotiation of the deal after the winning bidder has been selected,” several American P3 contracts have been renegotiated. For example, the Dulles Greenway failed to meet demand estimates, went into Chapter 11, tolls were increased, and the concession term lengthened from 40 to 60 years to the benefit of some of the original equity holders. As noted in our 2011 Hamilton Project discussion paper “Public‐​Private Partnerships to Revamp U.S. Infrastructure,” five of the 11 highway P3 contracts that had reached the operational stage at that time had been renegotiated.

Poole argues that governments do not have resources to build new infrastructure, so there is a fiscal reason for P3s. That is not true, as we explain in our 2013 article “The Basic Public Finance of Public‐​Private Partnerships” (Journal of the European Economic Association 11[1]: 83–111). To see the point, suppose that instead of a toll‐​based P3 highway, the transit authority issues a bond backed by the toll revenues from the highway and uses those funds to contract a private firm to build the road. If a private party receives financial support for a toll‐​based P3, the same financial resources are available to the transit authority. That is, for example, how the Golden Gate Bridge in San Francisco and many other U.S. highways were built. When projects are viable and can be financed by a trust financed by tolls, there is no lack of funding for a government‐​sponsored project. The fiscal travails of a state are not a reason for choosing P3s over government‐​led projects. Instead, as we argue in our article, a well‐​designed P3 program can lead to a variety of efficiency gains, including better and cheaper maintenance, screening projects that are white elephants, fewer delays and cost overruns, and avoiding the cost of bureaucracies.

Poole concludes his critique by suggesting that “rather than needing reform, the U.S./Australian P3 approach can serve as a model for other countries.” However, in the period 1993–2016, there were only 13 toll concessions in all the United States (and nine AP concessions since 2009). Contrast that with Chile, whose economy is about 1.3% the size of the U.S. economy and where 40 toll‐​funded highways and tunnels were concessioned in roughly the same period. This suggests that Chile, with its well‐​established conflict resolution mechanisms for P3s, its toll‐​based PVR contracts, and funds provided by institutional investors, is a better model and should be taken seriously, even in the United States.

Download the Regulation Article

About the Authors
Eduardo Engel

University of Chile in Santiago

Ronald Fischer

University of Chile in Santiago

Alexander Galetovic

Hoover Institution at Stanford University