Surface Transportation Innovations Newsletter

Surface Transportation Innovations #66

Topics include: protecting the public interest in public-private partnership toll road deals; fresh thinking on highway design; role for tolling ; automating HOT lanes enforcement; greenhouse gas regulation; and other news.

In this issue:

Protecting the Public Interest in Long-Term Toll Concessions

Three reports have appeared in recent weeks offering advice to state transportation policymakers on how to protect the public interest when choosing to use long-term PPP arrangements in transportation infrastructure, especially for toll roads. One comes from the National Cooperative Highway Research Program (NCHRP), another from the Pew Center on the States, and the third from the Public Interest Research Group. I commend two of these reports to your attention.

The broadest and most comprehensive is NCHRP Synthesis 391, “Public Sector Decision Making for Public-Private Partnerships.” NCHRP is an ongoing research program of the Transportation Research Board, in cooperation with state DOTs. Its synthesis reports are generally well-done, and this one is no exception. Written by Jeff Buxbaum and Iris Ortiz of Cambridge Systematics (authors of a previous, briefer, report on this topic for the Keston Institute at USC), it provides a comprehensive primer on long-term concession PPPs. It describes how governments should make decisions about when and how to use this approach, provides extensive discussion of public-interest issues and how best to deal with them, and offers a useful discussion of misperceptions about PPPs-including the three biggies:

  • That rigid non-compete provisions are always involved;
  • That PPP deals always mean tolling and the likelihood of windfall profits;
  • That the public sector loses control of the facility.

Unfortunately for such an important report, it is marred by some errors of fact and interpretation, several of which TRB has told me are in the process of being corrected.

The Pew report is “Driven by Dollars: What States Should Know When Considering Public-Private Partnerships to Fund Transportation.” It was stimulated by last year’s controversy over the proposed lease of the Pennsylvania Turnpike. It’s an even-handed look at what Pennsylvania did right and wrong in pursuing this transaction. Though focusing mostly on the lease of existing toll roads (a small fraction of the likely highway PPP transactions over the next several decade), its guidelines and lessons learned are pretty much on-target. (Alas, this report also contains a few factual errors, such as the claim that most US concession terms are 75 to 99 years, which is falsified by 35-year terms in California and 50-year terms in Texas.)

The PIRG report, “Private Roads, Public Costs: The Facts About Toll Road Privatization and How to Protect the Public,” is a sweeping attack on PPP toll road concessions, similar to the group’s previous report (which I reviewed in Issue No. 59, September 2008). As you might guess from the title, it tries to portray this ongoing trend as primarily about the long-term lease of existing toll roads, even though only four such leases have taken place in the United States, while numerous “greenfield” projects to create new capacity have been financed or are in various stages of procurement. It also pretends that most PPP toll road deals are 75 to 99 years and involve large up-front payments, neither of which is the case.

And it also claims that “any private deal must demonstrate that it saves money compared to what public authorities could generate by borrowing against the same toll hikes.” This comparative-cost-of-capital argument (beloved by PPP critic Dennis Enright) is off-base in several ways. First, it ignores the availability of tax-exempt Private Activity Bonds, which brings the cost of debt capital for PPP deals very close to what prevails on tax-exempt public-sector debt. More important, it ignores the value of the large risk transfers likely to be involved in PPP toll roads (which is a big factor in the Public Sector Comparator model I wrote about last issue), as well as potential savings in life-cycle cost via the PPP approach. It’s long overdue that this bogus cost-of-capital argument be put out to pasture.

So read the NCHRP and Pew reports to learn a lot, and if you have time, take a look at the PIRG report to know what kinds of arguments opponents will be making. Sources for these reports:

» return to top

Thinking Outside the Box on Highway Design

One of the reasons it’s assumed that “we can’t build our way out of congestion” is that the costs of expressways and even urban arterials are so high. But what if we could rethink the factors that lead to such high costs per lane-mile? That’s what Chen Feng Ng and Kenneth Small have done in their recent paper, “Tradeoffs Among Free-Flow Speed, Capacity, Cost, and Environmental Footprint in Highway Design” (

They point out that the AASHTO Highway Capacity Manual’s standards for expressway design are based largely on two underlying assumptions: these roads must be safe for travel at high speeds, and they must all be able to carry mixed traffic, including large trucks. They then ask the logical question: if urban expressways are congested much of the day so that only a small fraction of their daily traffic can operate at high speed, should we still design to standards based on those high speeds? And also, should we design every major roadway to accommodate large trucks?

In the paper Ng and Small then explore the trade-offs involved in narrower lane and shoulder widths (which result in lower design speeds) to make possible an additional lane in each direction, on both expressways and major arterials, within the same total width. For approximately the same construction cost, the expressway or arterial with more lane capacity wins out in most cases where peak-period traffic volumes are much greater than off-peak volumes. The “narrow” designs are strongly favored in all cases in which there is appreciable queuing-i.e., serious congestion. That’s because the slightly lower free-flow speeds of the narrow designs pale in comparison to their much greater ability to handle traffic before flow breaks down.

The other important comparison Ng and Small make is between a full-blown expressway and a super-arterial with grade separations (overpasses) instead of signalized intersections. Here they compare roadways with equal capacity but different costs to construct. They find that the unsignalized arterial is more cost-effective than an expressway of equal capacity under most scenarios, “because the difference in travel-time cost is relatively small while the difference in construction cost is much higher.”

Because many highway engineers will be concerned about the safety implications of either narrower lane widths or of building super-arterials instead of expressways, Ng and Small review the relevant literature on safety and find it ambiguous, and they suggest that better speed control might mitigate any adverse safety impacts.

An adaptation of the super-arterial idea, developed independently by highway engineer Chris R. Swenson, appears in the latest policy study from the Reason Foundation’s Galvin Mobility Project, “Reducing Congestion in Lee County, Florida” ( Faced with a fast-growing suburban county with only one expressway (I-75) and no likelihood of adding any more, but a good network of six-lane arterials, Swenson proposed converting some of those arterials into a new kind of managed-lane facility. “Queue-jump” overpasses would be constructed at major signalized intersections, with four lanes on the overpasses and two retained at grade (plus turn lanes). Electronic tolls would be collected from those using the overpasses, but no one else. Thus, the concept would not violate the acce pted Florida principle of charging tolls only for new capacity, not on existing roadways.

Swenson developed cost and traffic estimates for one such corridor that would involve building six queue jumps. Preliminary calculations suggest that the costs could be recovered via toll-revenue financing. The report goes on to suggest a network of such queue jump corridors. Because the throughput of a queue-jump corridor would be significantly greater than that of the existing arterial, the network of such corridors would add the capacity for 3.6 million vehicle miles of travel per day.

This is the kind of outside-the-box thinking that we need more of, to get the maximum value out of always-limited transportation funds.

» return to top

How Much Will Tolling and PPPs Do?

One of the questions I’m asked most often by infrastructure investors and reporters is: “How big a slice of the pie will toll-based PPP projects account for in coming years?” There is no simple answer, since this depends in part on public policy decisions at both federal and state levels of government. Still, there is some data available. The most recent take on this question is a set of reports and databases developed last year by Ben Perez and Steve Lockwood of Parsons Brinckerhoff for the Federal Highway Administration. You can find this very useful compilation of information at: It includes a white paper, four databases, three summaries, and various sortings of the data by different categories.

The main focus of the information is on the role of toll finance in the U.S. highway sector. It covers the period of 1992 to 2008, i.e., from enactment of the ISTEA legislation (which began reducing federal restrictions on tolling of federal-aid highways) to the present. One of its most important findings is that during this period, of the average of 150 centerline miles added per year of expressway-standard highway, between 50 and 75 miles (i.e., one-third to one-half) were financed based on toll revenues. To be sure, toll revenues are still a modest 5.4% of total highway revenues (with the vast majority coming from federal and state fuel taxes). But the trend is clearly upward, and focused on the high-end portions of the system that are generally the most difficult for the public sector to amass funding for.

The survey identified new toll road activity in 33 states and territories during 1992-2008, involving 235 new-capacity projects. Overall, the projects break down as follows:

Stage Value ($B) Lane-miles
Completed $40.0 3,900
In construction 17.8 1,734
In design/finance 19.4 1,887
In enviro. review 23.3 2,266
In planning 60.5 5,900
Total: $161.0 15,687

If you go to the FHWA table that sorts the projects by “Private Sector Development & Financing,” you find 24 projects worth $36.3 billion as having “possible” private involvement and another 24 projects worth $21.4 billion defined as having private involvement. That’s 36% of the $161 billion total. On one hand, that sounds pretty low, but remember that a number of states have only passed enabling legislation in the last year or so; most projects they (e.g., California) might do as PPP toll roads have not yet been defined as such. Plus, the current recession may shift a number of large projects from the public sector to the PPP side. So my guess is that the data in this FHWA database understate the likely magnitude of PPP toll road activity we can expect in coming years.

» return to top

Automating HOT Lane Enforcement

The continued growth of HOT (high occupancy toll) lanes beyond very simple initial ones is leading to increased concern about how to enforce occupancy requirements. Most early HOT lanes, with an entrance at one end and an exit at the other, had room to provide an enforcement zone somewhere in between, in which those HOVs which qualified for reduced-rate or zero tolls moved into a separate lane from the other cars paying tolls via transponder. In the “declaration lane,” visual enforcement is the rule-essentially, counting heads by patrol officers, either overhead or alongside. But larger and more complex networks of HOT lanes, especially when retrofitted into existing urban freeways with constrained rights of way, don’t have room for what would have to be numerous declaration lanes. Moreover, the idea of a single location where tolling (and declaration) takes plac e is also no longer applicable to a network of priced lanes, where people will be charged either by total distance driven (from entry to exit) or for each link in the system that they traverse.

This dilemma surfaced recently in the Seattle area, site of one of FHWA’s Urban Partnership Agreement projects: using electronically collected congestion-priced tolls to manage traffic on, and help finance the replacement of, the SR 520 floating bridge. A study for Washington State DOT by IBI Group, released in January, evaluates five occupancy-enforcement alternatives:

  • Transponders for SOVs plus visual enforcement of HOVs;
  • Transponders for SOVs plus declaration lane for HOVs;
  • Transponders for all plus declaration lane for HOVs;
  • Transponders with declaration feature for all;
  • Transponders for SOVs, license-plate registrations for HOVs.

The fifth of these is what is currently in use on the new I-95 Express Lanes in Miami, the first Urban Partnership project to go into (Phase 1) operation.

The report rejects the first option unreliable, especially on a bridge. And it rejects the second and third due to no room on the bridge for a declaration lane. It rejects the Miami alternative due in part to the potential administrative burden of carpool registration and concern about enforcement. So it recommends going with a more-costly transponder than those currently in use, featuring the addition of a button by which vehicle occupants self-declare how many people are in the vehicle at that time. It claims this would provide “automated enforcement.”

I guess IBI Group and WSDOT have more confidence in people’s honesty than I do. Why would a solo driver running late crossing the bridge be any less likely to divert into the faster HOT-3 lane if he has to press a button claiming there are three people in his car than a similar solo driver who illegally moves into a regular HOV lane? This is not automated enforcement at all. The enforcement is simply the same old unreliable visual enforcement by a patrol officer. If WSDOT is counting on financing several billion dollars worth of SR 520 bridge construction costs via tolling on the honor system, I’d say good luck and good-bye if I were a potential investor.

An alternative to this honor system approach is what I recommended to Florida DOT for the I-95 Express project: registered carpools with transponders recognized by the toll system’s software as qualifying for a special rate during peak periods and the regular rate at all other times. Occupancy enforcement would be the responsibility of the ride-sharing agency that does the registration, which would work with participating employers to audit the system on a regular basis to verify that those claiming to be commuter carpools are still, in fact, in operation. Under this approach, all enforcement would be electronic toll enforcement: Is there a functioning transponder, what toll rate applies at this time of day, and what is the status of the customer’s account?

FDOT accepted the registered carpool idea, but not the transponder enforcement approach, preferring to go with a decal on each registered vehicle plus the license plate number entered into the system software. Hence, actual enforcement still requires a state trooper to be there and count heads.

I fleshed out this idea in a paper and poster session for the 2009 Transportation Research Board annual meeting. It’s paper #09-0385 (“Automating Managed Lanes Enforcement”) on the 2009 Annual Meeting CD-ROM, if you attended and have one. If you don’t have the CD and would like a copy of the paper, just send me an email request and I’ll send it to you.

» return to top

How Not to Regulate Greenhouse Gases

On Earth Day, April 16th, the Environmental Protection Agency is widely expected to announce an “endangerment finding” about CO2. That would be a legal declaration, under which EPA would then be required to prepare a proposal to regulate CO2 like other “criteria pollutants” that come out of factories, refineries, dry cleaners . . . and motor vehicles. That may not sound like a big deal on first hearing, but the consequences could be hugely negative for transportation on our highways.

It was unclear for a long while whether the EPA had legal authority under the Clean Air Act to regulate greenhouse gases (GHGs). But in Massachusetts vs. EPA, the US Supreme Court held that it does, if it chooses to exercise it. So last fall the Bush EPA put out an Advanced Notice of Proposed Rulemaking (ANPR) laying out how it might go about regulating GHG’s under the Clean Air Act (CAA). The comments filed in response to Docket # EPA-HQ-OAR-2008-0318 make for chilling reading.

In order to regulate GHGs under the CAA, the EPA must establish National Ambient Air Quality Standards (NAAQS) for CO2 and other GHGs. And since the sources of GHGs are global, and concentrations produced today remain in the atmosphere for decades, under the CAA’s “transportation conformity” provisions, the entire country would be a non-attainment zone, and would probably remain there for decades or centuries regardless of what regulatory actions are taken here. That means federal transportation funds would be withheld unless and until states produced implementation plans showing the attainment of conformity-very likely impossible for the next few decades. As the US DOT’s comments made clear,

“Such a finding would reach beyond power plants and other installations to include vital transportation infrastructure such as roads, bridges, airports, ports, and transit lines. At a time when our country critically needs to modernize our transportation infrastructure, the NAAQS that the draft rule would establish-and the development of the implementation rules that would follow-could seriously undermine these efforts.”

You can read a lot more of the gory details in the docket submission by my former Reason colleague Marlo Lewis, PhD, former staff director of the House Government Affairs subcommittee on national economic growth, natural resources, and regulatory affairs. (

It’s clear to me that using the Clean Air Act in this way would be disastrous for the economy as well as for transportation. It’s the wrong tool to use against GHGs. As I’ve written elsewhere, the least-bad approach is a revenue-neutral carbon tax, which would accomplish what a cap-and-trade system aims to do (put a price on emitting CO2) but without either (a) creating huge political gains and losses as various sectors jockey for allocations of free permits or (b) producing windfall revenues to be allocated to politically favored purposes.

The transportation community should line up in favor of a revenue-neutral carbon tax, because the alternatives-especially EPA regulation under the CAA-would be so much worse.

» return to top

Upcoming Conferences

I don’t have space to list all possible transportation conferences of interest; those listed here are only ones that a Reason colleague or I will be speaking at. As you can see, we’ll be busy!

» return to top

News Notes

Congressional Budget Office on Congestion Pricing. The CBO has released a well-written overview on “Using Pricing to Reduce Traffic Congestion,” dated March 2009. I visited with lead author Scott Dennis and his colleagues at CBO last month when I was in DC for a conference. These guys get it! This is perhaps the best 25-page overview of how congestion pricing works, its benefits and challenges, and policy options for making greater use of it in this country. Go to:

Court Questions Port/Union Deal. A package deal in which the Port of Los Angeles combined its (justified) clean-diesel trucks mandate with a requirement that all port drayage be carried out by fleet operators instead of the current owner/operators violates federal pre-emption under the 1980 trucking deregulation act. So ruled the 9th Circuit Court of Appeals, sending the case back to a lower court for disposition, as requested by the American Trucking Associations. The package deal had been proposed by an unholy alliance of the Teamsters’ union and various environmental groups.

New Transportation Program at MIT. My alma mater has announced Transportation@MIT, a joint effort of the School of Engineering, School of Architecture and Planning, and the Sloan School of Management. It will initially operate on a two-year pilot program basis, addressing the intersection of transportation and environmental policy. Go to

Mary Peters Rejoins the Private Sector. Former Secretary of Transportation Mary Peters has become a senior advisor to one of the leading U.S. firms in the transportation PPP field, Zachry American Infrastructure. Peters will advise Zachry and its partner, Hastings Fund Management, as the Zachry Hastings Alliance pursues PPP opportunities in the transportation, energy, water, aviation, and ports sectors.

Supply and Demand at Work on 91 Express Lanes. Those who thought the value-priced toll rates on Orange County, California’s 91 Express Lanes on SR 91 could only go up are learning more about economics. Since traffic levels have dropped nearly 15% over the past eight months, congestion during peak periods has decreased, thereby triggering the pricing algorithm. As of April 1, 2009, the toll rates will decrease by 50 cents. This is the third price decrease in the past year. The highest rates, for selected PM peak hours in the eastbound (peak) direction, are in the $6 to $9.50 range. By contrast, the AM peak tolls in the westbound (peak) direction are in the $3 to $4.35 range.

Legal Issues in Highway PPPs. The National Cooperative Highway Research Program has released “Major Legal Issues for Highway Public-Private Partnerships,” as Legal Research Digest 51. Not being an attorney, I cannot attest to the quality of its legal scholarship. But from a policy standpoint, it gets the issues mostly correct, discussing both “brownfield” and “greenfield” projects. I did find more than half a dozen errors of fact or interpretation regarding the projects used as examples, which suggest an unfortunate lapse in the peer review process for a document of this importance.

» return to top

Quotable Quotes

“By providing a financial incentive for drivers to switch to times, routes, or modes of transportation that are less congested, [congestion] pricing encourages drivers to use the existing highways more efficiently. The resulting revenues both inform future investment decisions-by providing data about the value of expanding highway capacity at the places and times where the funds are collected-and can provide the funds to pay for doing so.”
–Scott Dennis, “Using Pricing to Reduce Traffic Congestion,” Congressional Budget Office, March 2009 (URL provided above under News Notes)

“There is no doubt that the technology for distance-related road user charging schemes on a small or large scale is available and mature in relation to component development. However, road user charging is not primarily a question of technology. It is a question of policy, of trust in the credibility of politicians by the public, of political objectives, and of acceptance by the various stakeholders. It is a question of the real transaction costs in relation to the revenue, the use of the revenue, additional burden for the road user, or a compensation on the tax side within the frame of a systematic paradigm shift in financing the transport infrastructure.”
–Andreas Kossak, “The Politics of Tolling,” Traffic Technology International Annual 2009

“To summarize, the private provision of highway services offers benefits in many situations. It deserves a prominent place in highway policy, especially given the potential to provide new capacity in a fiscally constrained and congestion-challenged environment. In order to realize these benefits and to prevent undesirable outcomes, however, the public sector must be sophisticated in setting the terms under which private firms operate.”
–Kenneth A. Small, “Private Provision of Highways: Economic Issues,” Policy Study No. 17, Show-Me Institute (

“[T]ransportation facilities are likely to be heterogeneous in terms of their current and expected future traffic flow, competition, and amount of maintenance required, among many other dimensions. No one concession length is appropriate for facilities that vary so greatly. If, for example, traffic demand is low and uncertain on a particular facility, a longer concession may be necessary to attract private capital to the project. . . .[T]he key lesson here is that a flexible approach is warranted, and a one-size-fits-all policy that would artificially limit concession length is likely to result in social harm.”
–Rick Geddes, Cornell University, “The Future of Surface Transportation Funding in America,” Tollways, Autumn 2008. (

» return to top