Surface Transportation Innovations Newsletter

Surface Transportation Innovations # 60

Topics include: creating a HOT Network while minimizing capital outlays; megaprojects cut congestion, add economic value; rethinking the "costs of sprawl"; unjustified exemptions from gas taxes and tolls; centralized dredging fund-another "tax-and-grant"

In this issue:

A New Approach to Creating HOT Networks

Five years ago when Ken Orski and I introduced the concept of HOT Networks, the reactions were mixed. A lot of people were excited by the idea of a seamless, region-wide priced-lane network offering reliable, uncongested travel not only to motorists but also to vanpools and express buses. But many balked at the very high capital costs, ranging from $3 billion to $11 billion, since most of the network would be new capacity (though of course converting any existing HOV lanes to HOT lanes). The original study is available online at

But a breakthrough paper was released last month, offering a way to create HOT Networks at far lower cost. It’s by Patrick DeCorla-Souza, the indefatigable head of the Federal Highway Administration’s Value Pricing office and co-chair of the Transportation Research Board’s Congestion Pricing Committee. The basic concept calls for converting the left-most lane of a freeway (whether it’s HOV or general purpose[GP]) to HOT, and simultaneously converting the right shoulder to a peak-period-only travel lane. Thus, no “free” lanes would be taken away, and the HOT lane would represent a net addition of peak-period capacity. To provide for the needed buffer space between HOT and GP lanes, the lane-widths would be reduced from 12 ft. to 11 ft., via a federal waiver (as Miami is doing for its I-95 Express Lanes).

For a 10-mile segment of a six-lane freeway, DeCorla-Souza estimates the benefits (just in time and fuel savings) at $42 million per year, and the toll revenue (assuming tolling only during peak periods) at nearly $6 million. He draws on data from studies of a possible network for the Seattle metro area (300 route-miles) to estimate the implementation cost (including a lot of technology for overhead lane controls, dynamic shoulder management, and tolling infrastructure) at about $3.1 billion. Estimated peak-period revenues are $170 million per year (which would certainly increase over time), and benefits are $1.3 billion per year. The benefits estimate suggests people would pay a lot more than $170 million to realize those benefits. The good news is that even with all this technology, the system could well be self-supporting.

The paper is going to appear shortly in FHWA’s Public Roads magazine. But meanwhile you can find it online at

Two other points you should be aware of as you read this path-breaking paper. First, the revenue assumptions assume that only registered vanpools and buses get free passage; all other users pay the value-priced toll. That’s a recommendation Ken and I made in our 2003 study, and it’s one of the keys to generating enough revenue to make such networks approach self-supporting status. Second, it turns out that the vast majority of existing urban freeways have 12-ft. lane widths, so they have the potential (with FHWA approval) to apply this lane-narrowing/dynamic shoulder usage concept.

Finally, let me note that while this approach requires little or no widening of existing freeways, it would require the creation of flyover ramps to inter-connect the HOT lanes at freeway interchanges. The cost of those flyovers does not appear to be included in DeCorla-Souza’s notional Seattle network, but it was part of the Orski and Poole paper-and really should be part of any serious HOT Network plan.

» return to top

The Benefits of Megaprojects: Two Examples

One of the premises of Reason Foundation’s work on urban transportation is that there is often real value in adding capacity to congested urban roadway networks. For the past 25 years or so, traffic (as measured by vehicle miles of travel) has grown at least 10 times more than the growth in urban lane-miles, which tends to discredit the notion that if you don’t build it, they won’t come. With that thought in mind, let’s consider two recent studies of urban roadway megaprojects-one in Boston and the other in Sydney.

The Boston project is the infamous Big Dig, known formally as the Central Artery/Tunnel project. I only recently discovered that the well-respected consulting firm Economic Development Research Group did a study of the completed project in 2006. Volume 1 of “Economic Impact of the Massachusetts Turnpike Authority & Related Projects” covers the Big Dig’s transportation impacts, while Volume 2 assesses land-use and economic development impacts. ( Two useful summary/discussions are by Peter Samuel ( and Dan Baxter (

The project put a key stretch of I-93 underground, extended the I-90 turnpike underground to the seaport area and linked it to a new (third) tunnel to Logan Airport. The project also carried I-93 north across the Charles River on the world’s widest cable-stayed bridge. But as Baxter and Samuel both point out, this is not primarily a highway; rather, it is America’s largest interchange reconstruction, removing numerous bottlenecks that were notorious congestion causes, even way back when I was at MIT in the 1960s. In addition, the project added a lot of new capacity-one-third more airport-tunnel access, the extension of I-90, and on I-93 itself 8 to 10 lanes where there used to be six.

Critics said the Big Dig would be obsolete the day it opened, on the premise that latent or induced demand would quickly eat up the new capacity. But as EDR Group documented, they were very wrong. The project has cut congestion dramatically, with daily vehicle hours of delay dropping from over 38,000 in 1995 to 12,800 in 2005 (the latest year covered in the study). And this was despite VMT growth of 13% during that decade. EDR Group estimates the direct annual savings to motorists at $167 million per year, most of it in time savings. Another beneficiary is Logan Airport, which had been losing ground to other New England airports because it was so hard to get to. Some 800,000 more people (a nearly 50% increase) now live within a 40-minute drive of Logan.

Then there are the real estate benefits from the combination of removing the former elevated Central Artery and improving access to downtown via roadway. Thus far, $7 billion in new private investment in office, retail, and residential has been generated downtown, about 10 million sq. ft. of an estimated 16-21 million that planners estimate will occur just in the South Boston Seaport district. That will generate loads of economic activity, while increasing property tax revenues.

There are two questions you should be asking now. First, do these benefits justify spending nearly $15 billion of mostly federal and state highway user tax revenues on this one project? Second, how sustainable are the congestion-relief (and therefore access) benefits? On the second, unless the Turnpike Authority can somehow implement congestion pricing on the project, those gains will erode over time. On the first, the costs exceed benefits-but if this project had been developed as a toll concession megaproject from the outset, it’s highly unlikely that its costs would have ever gotten so far out of control (from a very early estimate of $2.2 billion to nearly $15 billion).

And that brings me to the second megaproject, Sydney Australia’s urban tollway network. It is the subject of a detailed assessment by Ernst & Young, “The Economic Contribution of Sydney’s Toll Roads” ($file/Toll-roads.pdf). Its purpose is to quantify the net benefits (NPV of benefits minus NPV of costs) of the six toll roads and short interconnecting freeways that make up the Sydney Orbital Network, which was completed last year with the opening of the Westlink M7. E&Y first went back and reviewed the original benefit-cost analysis for each tollway, and then updated them using actual rather than projected data. In addition, to estimate network benefits, they used a general-equilibrium model of the regi onal economy.

The results were pretty dramatic. Expressed in 2007 Australian dollars, the cost of the network was $7.5 billion (which was 33% higher than originally forecast). User net benefits totaled $30.8 billion (travel time savings, lower operating costs, and accident savings). After accounting for externalities and operations and maintenance costs, the overall net benefit was $22.7 billion, well above initial projections. All of these measures are for the period 1986 to 2020. Data from the general-equilibrium model, comparing toll road and no-toll road scenarios, suggests that the toll network adds about 0.55% to the gross state product of New South Wales-by making the economy work better in various ways.

E&Y noted, but was unable to quantify, the benefits to NSW from the use of private finance and private concession companies to develop and operate these toll roads. Let me suggest that the relatively modest cost over-run on this megaproject, compared with megaprojects generally and the Big Dig in particular, is one such benefit. And the fact that these are all toll roads means that when they start becoming congested, two remedies will be readily available that the Big Dig currently does not have. First, they can shift to congestion pricing, to manage demand and keep traffic flowing smoothly. Second, they will have funds available to add capacity, where it is feasible to do so. In that sense, the Sydney improvements should prove to be far more sustainable than the Big Dig, however impressive the latter’s current benefits.

» return to top

Rethinking the “Costs of Sprawl”

Last month I criticized the very tenuous connection between restricting the growth of low-density suburbs and reducing greenhouse gas (GHG) emissions-the idea underlying California’s SB 375. Alas, Gov. Schwarzenegger didn’t agree, and signed the bill into law. The widespread belief that low-density suburbs (a.k.a. “sprawl”) are bad has been around for a long time, and GHG production is just the latest stick being used to beat up suburbs. Two earlier ideas were (1) that such suburbs don’t pay their way and hence are subsidized by taxpayers in other parts of the metro area, and (2) that residential energy use is much higher in those suburbs. Here are two relatively recent works that challenge those ideas.

On the first, one of the main sources for the idea of subsidized suburbs is a pair of reports from the Transit Cooperative Research Program, “Costs of Sprawl-2000” and “Costs of Sprawl Revisited” (2002). Wendell Cox and Joshua Utt set out to test this empirically by looking for three kinds of data. If the “low-density suburbs entail higher public-sector costs” premise is true, then comparative data on municipal expenditures should reveal that lower spending per capita is correlated with (a) higher population densities, (b) lower rates of population growth (since it’s the fast-growing metro areas that typically exemplify sprawl), and (c) older municipalities. They carried out an econometric analysis using data on 700 municipalities from the 2000 census government finance database. The results are available at

Cox and Utt created three econometric models: one for general government expenditure, another for water systems, and a third for wastewater (since the latter two are usually government-provided, as opposed to developer-provided streets and commercially provided electricity). The general government model therefore excluded utilities (and also K-12 education, since that is provided separately by school districts). The 13 variables, including density, growth rate, and age of community, explained only 29% of the variation in per-capita costs. Density alone had a very small effect: a 10% increase in density would produce only a 1.46% decrease in municipal cost per capita. The water and wastewater models also found a small impact from density-e.g., a 1,000 person per square mile difference (quite large) was correlated with only a $4 per capita annual difference in water charges (and $6 annual difference in wastewater charges). And since those charges are typically paid by homeowners via utility bills, they do not represent added costs to other taxpayers. These results certainly call into question the “subsidized sprawl” model.

Another much-cited critic of low-density suburbs is Reid Ewing of the National Center for Smart Growth Research and Education at the University of Maryland. Early this year, Ewing and Fang Rong published “The Impact of Urban Form on U.S. Residential Energy Use” in Housing Policy Debate (Vol. 19, Issue 1). In the same issue there is a critique by my Reason Foundation colleague Sam Staley, “Missing the Forest through the Trees?” I commend this debate to you, and can only give you the briefest summary here.

Ewing and Fang also do econometric modeling, using several data sources to compare household energy use by county, with each county assigned a “sprawl index” developed in previous work by Ewing. They find that, after controlling for household size and income (which are positively correlated with size of house) that an average household would use about 20% less residential energy if it lived in a compact county rather than a sprawling one.

Staley finds no fault with the econometrics, but takes issue with their recommendation that public policy should discourage sprawl and encourage compact development. He notes the rather different findings on residential energy use in a study by the Australian Conservation Fund and the University of Sydney (which I noted in Issue No. 56). He also points out that if the main concern is reducing GHG emissions (on which Ewing and Rong base their case), there are alternative ways to pursue that goal (such as shifting to much greater use of nuclear power for electricity) that would not deprive families of their preferred choice of housing. Their analysis also fails to take into account how changing prices affect technology and therefore the choices people have, and that these effects are far more direct than land-use policies. Staley also cites another Australian study whi ch found that high-rise apartment buildings had twice as much GHG emissions per capita as single-family detached homes, though town homes and villas emitted less than single-family homes. I think you will enjoy this spirited exchange.

» return to top

Exemptions from Fuel Taxes and Tolls

In recent months several news items have come my way in which an enterprising reporter has discovered that some subset of public officials has been given special electronic toll transponders allowing them toll-free passage. The most recent example was in New York City, where it was discovered that MTA Bridges & Tunnels had been providing no-toll E-ZPass tags to board members, employees, retirees, and numerous city, county, and state vehicles. In late September, citing a budget crunch, the MTA board voted to change the policy. Henceforth, all city, county, and state vehicles’ E-ZPass tags will now pay the regular toll, like everyone else’s vehicles. Left unchanged, however, according to the New York Times story on Sept. 25, was toll-free E-ZPasses for 1,813 MTA employees and 1,035 retirees. Maybe that’s considered part of their compensation package, like airline employee passes.

At any rate, this got me thinking about the long-standing (and seldom-questioned) exemptions from federal and state fuel taxes. Like tolls, fuel taxes are supposed to be the way we pay for using roads. Every single user takes up space, contributing (at times) to congestion and adding to wear and tear on the road. I can’t see any good reason to exempt any vehicle from paying for its road use. But you’d be amazed at who gets exempted, and how much revenue is foregone in the process.

Fortunately, the American Transportation Research Institute has done some homework on this issue. There are four principal categories of highway vehicles exempt from federal fuel taxes:

  • Those used exclusively by state or local governments, or the District of Columbia;
  • School buses;
  • Qualified intercity and local transit buses;
  • Those used exclusively by nonprofit educational organizations.

ATRI was unable to estimate the last of these, but the first three add up to at least $570 million per year that would otherwise go into the Highway Trust Fund.

At the state level, the laws vary from state to state. A number exempt or refund fuel taxes for state, county, and city vehicles, which ATRI estimates as cutting state fuel tax receipts by $155 million per year. Most or all school buses are also exempt at the state level, costing another $126 million per year. Some 36 states exempt federal fleets from paying state fuel taxes (even though federal vehicles do pay federal fuel taxes-go figure!), for another $29 million in lost state revenue. And 47 states and the District of Columbia exempt US Postal Service vehicles from their fuel taxes, costing them another $27 million per year (and giving USPS an unfair advantage over Fedex and UPS).

Altogether, the federal and state exemptions add up to $907 million per year, and if unknowns (such as charitable organizations) were also factored in, ATRI estimates the total would exceed $1 billion, and might be $1.5 billion per year.

When the new Congress begins work on reauthorizing the federal program next year, and as legislatures convene to cope with massive transportation funding shortfalls, isn’t it long overdue to rethink these exemptions from fuel taxes and tolls, and make everyone who uses roadways pay their fair share?

» return to top

Centralized Dredging Fund-Another Case for Devolution

Various thinkers and pundits have been making the case against the centralized “tax-and-grant” system of infrastructure funding for many years. As Peter Samuel explained in a Reason policy study last year (“The Role of Tolls in Financing 21st Century Highways,”, this model distributes money via political rather than economic (return on investment) criteria, short-changes maintenance (thereby increasing life-cycle costs), and fails to provide enough of the kinds of road capacity that specific categories of users would be willing to pay for.

But Congress loves the idea of centralized infrastructure funds, pretending that this somehow allows them to master-plan a system and allocate resources where they are really needed. So in 1986 they decided to extend the model to harbors. Now we have a Harbor Maintenance Trust Fund, aimed at providing funds to keep harbors and waterways dredged to proper depths. The money comes from a federal Harbor Maintenance Tax, and the Army Corps of Engineers does the dredging.

How is this working in practice? There is “a dredging crisis in many parts of the country,” Jim Weakley of RAMP (Realize America’s Maritime Promise) told the House Transportation & Infrastructure Committee in September. In FY 2007, the feds collected $1.4 billion in harbor maintenance taxes, but Congress only appropriated $751 million for dredging. The rest just sits in the Trust Fund (current balance $4.7 billion) making the federal budget deficit look smaller.

RAMP’s solution is for Congress to create a “firewall” like the one that saved the federal Highway Trust Fund from bankruptcy-or was supposed to. Far better would be to devolve the user fee/user tax authority to each port and waterway, and allow them to contract with whoever is available and qualified-not just the Army Corps but any private dredging company. That way, resources would flow to those dredging projects that urgently need work, and in amounts necessary to get the job done. But I won’t be holding my breath for that approach to be adopted.

» return to top

Upcoming Conferences

Just a reminder that I don’t have space to list all possible conferences of interest; those listed here are only ones that I or a Reason colleague will be speaking at.

Port & Intermodal Finance & Investment Summit, Houston, TX, Oct. 20-22, 2008, The St. Regis Hotel. Details at

Partnerships in Transit, National Council for Public-Private Partnerships and Federal Transit Administration, Dallas: Oct. 22-23, 2008, Radisson Hotel Central Dallas. Details available at

Toll Roads South, Orlando, FL, Oct. 27-28, 2008, Crown Plaza Orlando Universal. Details at

The California Infrastructure Summit, Anaheim, CA , Dec. 2-3, 2008, Sheraton Garden Grove-Anaheim South Hotel. Details at

IBTTA Transportation Finance Summit, Washington, DC, Dec. 7-9, 2008, Fairmont Hotel. Details at

» return to top

News Notes

New Deloitte Report on Road User Pricing. My former Reason colleague Bill Eggers is one of the authors of a useful new report from Deloitte Research, “Changing Lanes: Addressing America’s Congestion Problems Through Road User Pricing.” In addition to providing a good overview of how and why pricing works and the various ways to use pricing, it also has some useful pointers on how to go about implementing it.

Latest on Shift from Cars to Transit. Several issues ago, I did some back-of-the-envelope calculations estimating that only a single-digit percentage of reduced auto trips (as indicated by declining VMT) were shifting to transit. New figures are now available at for second-quarter 2008. Comparing that quarter with the equivalent period in 2007, transit passenger miles increased from 14.28 billion to 15.02 billion. Urban roadway travel decreased from 791.04 billion passenger miles to 766.4 billion. So transit’s gain of 0.74 billion passenger miles is about 3% of the 24.6 billion passenger miles lost to highways from marginal trips.

Innovators in Action 2008. A fascinating new publication from Reason Foundation consists of essays by and interviews with 10 of the most creative public officials we could find-including Transportation Secretary Mary Peters, the late chairman of the Texas Transportation Commission Rick Williamson, and Denver Regional Transportation District CEO Cal Marsella. Go to

Cautionary Article on Hydrogen Highway. I was an early skeptic on the idea that hydrogen would be the fuel of the future, and that hydrogen fuel-cell cars would soon be displacing gas-guzzlers. A thoughtful essay on what’s happened to this vision appeared recently in The Economist. “The Car of the Perpetual Future” appeared in the magazine’s Technology Quarterly for Sept. 6, 2008. Well worth reading.

Reason Anniversary Celebration. Reason magazine, the flagship publication of Reason Foundation, celebrates its 40th anniversary this year, with a gala event in Los Angeles, Nov. 14-15. “Reason Goes Hollywood” takes place at the historic and glamorous Hollywood Roosevelt Hotel and includes both a conference and a black tie banquet. Details are at Hope to see you there!

» return to top

Quotable Quotes

“To ensure that we were giving customers [on the 91 Express Lanes] what they wanted, we did an evaluation of traffic volumes. We found that somewhere between the first 3,200 and 3,400 cars per hour [in two Express Lanes], the speeds become unstable . . . . When that happens, we actually make a refund to the customers, because when they aren’t happy, we aren’t happy. We established a toll policy that says when volumes get to a certain level we will raise tolls. Every quarter we reevaluate traffic volumes, and because of that we now have tolls that are $10 one way on Thursday and Friday afternoons [eastbound]. Tolls at other times of day are as low as $1.50 . . . . Most customers only use the toll [lanes] one or two times a week. They use it on the days they need the speed. . . . In fact, the two toll lanes, during peak, have higher volumes than the parallel free lanes.”
–Art Leahy, CEO of Orange County Transportation Authority, Metro Investment Report, September 2008

“The program Eisenhower created 50 years ago was well-defined and well-suited to its time. The goal was clear: build the Interstates and connect the country-and we did. Since that mission was accomplished a quarter of a century ago, our federal surface transportation program has lost its sense of direction. It has become a breeding ground for earmarks and burdened by a proliferation of special-interest programs, goals, and requirements. . . . The time has come to move beyond superficial discussions of how much money the federal government is going to spend on transportation and lay out a policy foundation that fits our current circumstances. . . . If lawmakers in Washington turn their backs on reform and content themselves with figuring out the funding formula-how to divvy up among the states what is left after the set-asides and earmarks-they will have failed.”< br /> Mary Peters, Secretary of Transportation, in Innovators in Action 2008 (

» return to top

Robert Poole is director of transportation policy and Searle Freedom Trust Transportation Fellow at Reason Foundation. Poole, an MIT-trained engineer, has advised the Ronald Reagan, the George H.W. Bush, the Clinton, and the George W. Bush administrations.