Be very, very wary of public-private partnerships

Assuming that it’s possible for Connecticut to impose tolls on roads constructed, reconstructed or maintained in part with federal dollars, how can the investment required to install toll gantries and cameras be funded before any tolls are actually collected?

Bill Cibes

The answer is simple and straightforward: issue state revenue bonds, to be repaid from the future revenue stream generated by tolls. State revenue bonds issued for a public purpose are tax-exempt. Accordingly, they would likely bear a lower interest rate than any loan granted by a private entity looking for a rate of return that would not only cover the cost of any capital that it borrows, but also provide a profit to the entity.

Revenue bonds for specific transportation purposes – analogous to revenue bonds issued to support the Clean Water Fund or Bradley International Airport improvements – also would not come under the state’s bond cap. In that respect, they would align with Gov. Ned Lamont’s proposed “debt diet.” This financing mechanism – distinct from general obligation bonds – has been productively used in Connecticut since at least 1987, with total authorizations of over $4 billion for just the two programs cited above.

Alternatively, the fertile minds of financiers and consultants have created a more complicated mechanism for developing public projects. Used for years in foreign countries, “public-private partnerships” (known by the acronym “P3s”) of widely varying scope and magnitude have been suggested to not only use private money to finance the installation of toll devices, but to employ private companies to build transportation projects – and perhaps to own and operate them. The final report of the Commission on Fiscal Stability and Economic Growth, issued in November 2018, outlined such a strategy.

A newly elected state senator has gone further, advocating the use of an extensive P3, the “securitization” of the revenue stream from tolls, to “unlock this enormous pot of private-sector money that we are not accessing now.” She suggests that it could amount to as much as $7 to $9 of private-sector money for every $1 the state pledges. Such a model would presumably be based on securitization agreements like those involving parking spaces on Chicago streets and the Indiana Toll Road, which produced, respectively, $1.16 billion and $3.85 billion in upfront payments from private investors in return for turning over parking fees and tolls to those investors for 75 years.

Lamont has also expressed interest in P3s, although he has wisely recognized that the details still need to be developed, and has rejected giving away ownership and/or control to private investors.

No matter how much a P3 is gussied up, at its heart it is just a loan that must be repaid with interest. Since a private entity’s financing will not be tax-exempt, that factor alone undercuts the use of a P3 to finance a project. And when ribbons, bows, bells and whistles are added to the basic loan without adequate analysis, even more unfavorable consequences can follow.

They include:
1) Obscure, supposedly innocuous provisions can be included in the contract that limit what the government can do in the future.
2) The calculation of the revenue and the return on investment – heavily dependent on assumptions about inflation, fee increases and usage – can severely disadvantage the government over the long term.

No matter how much a P3 is gussied up, at its heart it is just a loan that must be repaid with interest.

An excellent example is Chicago’s leasing of its 36,000 parking meters in 2008. Chicago leased its parking meters to a private company for 75 years in exchange for a payment of $1.16 billion. The next year, the Inspector General of Chicago, after reviewing the process of negotiating the deal as well as its substance, concluded that there had been no independent analysis of the terms of the deal before it was finalized, and no meaningful opportunity for public input.

Instead of subjecting the provisions of the proposal to open analysis before the deal was concluded, details of the agreement were announced by the mayor on Dec. 2, 2008 and the lease was approved by the city council two days later, without time to subject it to expert review. Such a review might have revealed a provision (partially renegotiated by a later administration) that obligated the city to reimburse the investing company for lost revenue if a parking space was unavailable for any reason.

In 2012 alone, that reimbursement totaled $27 million – which was more than the total revenue the city received from parking meter fees in the year before the P3 was signed. These annual payments were in addition to the parking meter revenue directly received by the company.

The financial penalty associated with this provision also effectively foreclosed public policy options that the city might have wanted to undertake in the future – such as removing parking spaces in order to add bicycle lanes or bus lanes.

Moreover, assumptions undergirding the calculations of both the value of the revenue forfeited by the city and the value of the revenue expected by the private company were not rigorously evaluated.

The Inspector General found in 2009 that “the city was paid, conservatively, $974 million less for this 75-year lease than the city would have received from 75 years of parking-meter revenue had it retained the parking-meter system under the same terms that the city agreed to in the lease.”

As it turned out, however, the 2009 report proved to have been far too conservative in underestimating the revenue loss to the city. A 2018 audit by KPMG showed that parking meter revenue to Chicago Parking Meters LLC increased to $134.2 million in 2017, “putting private investors on pace to recoup their entire $1.16 billion investment by 2021 with 62 years to go in the lease” – a stunning revelation.

Another example of this type of partnership backfiring was the Indiana Toll Road agreement. The U.S. Government Accountability Office prepared a report on Highway Public-Private Partnerships in 2008, in which the GAO pointed out both the benefits and costs potentially associated with such partnerships. Such P3s involved either the sale or long-term lease of existing highways for an upfront payment in return for the right to collect tolls on the road, or the right to construct a new highway with the right to collect tolls on the project when completed.

One example the GAO examined was the 75-year lease of the Indiana Toll Road (originally constructed with very few federal dollars, which were repaid before the lease began) in 2006 for $3.8 billion, giving the private company the right to collect tolls on the road. Although the GAO agreed that there were some potential benefits to the arrangement, it also noted some potential costs, which foreshadowed the experience of the Chicago Parking Meter fiasco:

a) “There is no ‘free money’ in highway public-private partnerships. Rather, this funding is a form of privately issued debt that must be repaid.”
b) “It is possible that the net present value of the future stream of toll revenues (less operating and capital costs) given up can be much larger than the concession payment received.”
c) “Non-compete clauses” could prevent the public from building competing facilities within a certain distance of the road in question, or improving surrounding roads if high tolls diverted traffic to those facilities

Another example is the use of P3s for some California toll roads and courthouses. According to one reviewer, “In a famous case, the California Department of Transportation used a P3 to build and operate express lanes that opened in the center of California State Route 91 in Orange County in 1995. When the government wanted to expand parts of the roadway to alleviate congestion, it was blocked by a “non-compete” clause in the 35-year contract. Following litigation, the government ultimately bought out the private partner. Just seven years after the express lanes opened, the county’s transportation authority paid $207.5 million for the $130 million project.”

The reviewer of the deal concluded: “Governments actually may take on all kinds of new risk they didn’t face before — like the implications of entering into long-term deals that can constrain lawmakers’ policy-making options for decades.”

In 2012, the Legislative Analyst’s Office (LAO) in California examined the P3 involved in the construction of a courthouse in Long Beach, which had been lauded as a project which came in on time and under budget. The LAO was more skeptical: “Our analysis indicates that utilizing a different set of assumptions than those [used to justify the project] (such as excluding the assumed federal tax adjustment and leasing costs) would result in the cost of the Long Beach courthouse project being less—by as much as $160 million in net present value terms – in the long run under a traditional procurement approach than the chosen P3 approach.”

Advocates of P3s argue that implementation and execution of P3 projects is far superior than projects undertaken by public agencies. The author of a McKinsey and Company report in 2017 maintains that in a well-structured P3, an owner can clearly specify performance standards, responsibilities, rewards and penalties.

In my experience, the ability of a government agency to draw up sufficiently specific contract requirements that anticipate future problems and control for them is sadly lacking.

The contention is that P3 private-sector partners can then engage in innovative problem-solving, align the various segments of project delivery, and manage and mitigate deviations from the agreed-upon plan. The overall conclusion of the report is that P3s can “consistently deliver better schedule and cost performance.”

However, any advantages of P3s depend on the competent upfront delineation of requirements and effective analysis of a potential agreement by government. If government is so bad at implementation and execution of projects, why does McKinsey believe that it will be that much better at the specification of contract details and careful scrutiny of the degree to which potential concessionaires meet required standards?

In my experience, the ability of a government agency to draw up sufficiently specific contract requirements that anticipate future problems and control for them is sadly lacking. It’s better to recognize that the public agencies can more effectively respond to future problems when they arise, as long as they are not limited by ill-considered contract provisions. Even the McKinsey report concedes that “a poorly executed contract can put a government in a risky position should the private partner fail to deliver.”

If electronic tolling is to be initiated in Connecticut, the simplest and most straightforward – and least expensive – way to finance the cost of constructing gantries and auxiliary devices like cameras is to issue revenue bonds. To undertake further steps – such as the design, construction, operation and maintenance of transportation projects supported by toll revenue – using public-private partnerships is unnecessary and potentially damaging to the public interest.

Bill Cibes is Chancellor Emeritus of the Connecticut State University System; was formerly Secretary of the Office of Policy and Management under Gov. Lowell P. Weicker; and is a former board member of the Connecticut News Project, publisher of the Connecticut Mirror.  

CTViewpoints welcomes rebuttal or opposing views to this and all its commentaries. Read our guidelines and submit your commentary here.