In this issue:
- What to make of the National Commission recommendations
- New Jersey confuses taxing and tolling
- Seattle area ready for road pricing?
- What now for Dulles Rail?
- CO2 and traffic congestion
- Japan's amazing highway sound walls
- Upcoming Conferences
- News Notes
Major controversy has erupted over last month's final report of the National Surface Transportation Policy & Revenue Commission. While most of the initial negative reaction came from conservatives over the potential tripling of the federal fuel tax, at least as much heat came from liberals who lambasted Mary Peters and the other two dissenting members for their minority report. Among the most biting of these was Washington Post editorial page editor Fred Hiatt's piece, "She Brakes for Ideology." The thrust of his piece seemed to be that simplistic right-wing, anti-tax ideology had forced the dissenters to disagree with the majority's proposals for a needed major revamping of the federal role in surface transportation.
If that is your assessment, I urge you to read the minority report, since it makes a far more sophisticated argument that the majority's prescription for change is misguided (see www.transportationfortomorrow.org/final_report/pdf/volume_1_minority_views.pdf).
But rather than defend that specific document, let me summarize my own concerns about the road map the majority laid out. I elaborated on these points in my January Public Works Financing column, at www.reason.org/commentaries/poole_20080100.shtml.
Let's set to one side the gas-tax increase proposal, since in my view that's a side-show. It's almost certain that we need to invest a lot more in transportation infrastructure, but my concern is that we make sound investments, not simply do more of the same. There are several crucial questions that must be asked and carefully answered before we can figure out how much is needed, and for what, and who should be doing the project selection.
My first question is: What is the appropriate role for the federal government, in this post-Interstate-building era. The Commission's answer is: Everything you can think of. Instead of looking carefully at which transportation problems are local, which ones are regional, and which ones truly national, they produced a set of 10 mega-programs that would greatly expand the federal role (but because there are only 10 instead of the current 108, they call this streamlining).
Next, they proposed demolishing the user-pays principle, by converting the highway user tax into an all-purpose transportation tax. The new transportation trust fund would be the equivalent of a giant martini glass, with every imaginable spending constituency having a straw. You think I'm kidding? If this proposal were implemented, the gas taxes we depend on to ensure modernized and expanded highways would be available for waterway projects, freight rail projects, high-speed intercity rail, urban transit projects in villages and burgs of virtually any size, and large new commitments of transportation funds to energy and environmental projects (as if we didn't already have DOE and EPA). Good-by user-fee highways; hello massive politicized public works.
Third, to arrive at the transportation investment need, they added up wish lists from all 10 of these spending constituencies. But that's not all. They then decided, on no basis other than tradition, that since federal funding has averaged 40% of highway and transit spending in recent decades, the feds should pay for 40% of this massively expanded agenda. No wonder they defined a "need" for surface transportation spending of $225 billion per year, compared with about $90 billion now.
And finally, realizing that the federal program has a growing credibility problem thanks to the explosion of earmarks, the majority called for the creation of a central planning agency to set priorities and make the spending decisions, rather than leaving this to the normal political horse-trading in Congress. That will never happen, but I'd oppose it even if by some miracle Congress decided to cede this enormous power to an independent agency. That's because central planning of the country's entire surface transportation system is an impossible dream, no more feasible than central planning of national economies. In the real world, such efforts are always subverted by interest groups who have far more to gain than the average voter or taxpayer has to lose from the decisions made by the central planners. An entire academic discipline, called Public Choice Theory, explains the incentive problems with interest groups and large-scale public bureaucracies.
I realize that this brief assessment may sound overly harsh. But read Volume I of the Commission's report with this critique and that of my PWF column in mind. I think you will then see that what the majority report recommends is a lot more complex than simply a big gas tax increase—and that the minority had reasonable grounds for dissenting.
I've talked with quite a few reporters over the last several weeks, discussing New Jersey Gov. Jon Corzine's plan to "monetize" the state's toll roads, refinancing them based on huge increases in toll rates in order to bail the state out of its current fiscal hole. In order to get around existing bond covenants, under which all toll revenues must be used to operate, maintain, rebuild, and expand the toll roads, Corzine would create a new "public benefit corporation" which would take over the toll roads. By paying off their existing bonds, the new entity would be free to use toll revenues for whatever its charter called for. Under Corzine's proposal, that would include paying off $16 billion of general state debt.
My problem with this plan is twofold. First, it's grossly unfair to motorists and truckers who use the New Jersey toll roads, 51% of whom (on the NJ Turnpike) are from out of state. They are not responsible for decades of fiscal mismanagement. If anyone should be bailing out the state, it should be the voters and taxpayers of New Jersey, who were asleep at the switch while their public officials spent the state into ruination.
And this gets to my second, related, objection. The extremely high new toll rates—for a truck they would go from $5 in 2008 to $52 in 2033—combine a toll and a tax. The portion that ends up actually spent on maintaining, expanding, and modernizing the toll roads is a toll. But the portion that goes for general state purposes is a tax, plain and simple. Blurring that distinction could greatly harm America's highway system. Converting a toll (a pure user fee) into a tax at the same time as the rates are greatly increased is virtually guaranteed to stimulate opposition to tolls as a method of highway financing.
And that could not happen at a worse time. During the last few years, the global capital markets have discovered the U.S. highway sector as a new investment category. This has occurred just as we're starting to realize the enormity of the investment needs facing us in this sector—for urban congestion relief and for long-haul truck routes, in addition to reconstructing much of our 30- to 50-year-old freeway and Interstate system. The last thing we need is a large-scale backlash against tolling by the motorists and truckers who should be looking to toll finance as a key part of the answer to our highway capital shortfall.
Gov. Corzine decided on the public benefit corporation approach after getting negative feedback to a trial balloon proposal for leasing the toll roads, as Indiana did. Yet it may be that leasing the toll roads to a toll road company with a global track record could raise what the Governor claims to need—without such sky-high toll rates. A source at one such company shared with me some internal number-crunching they have done recently. Remember that the Governor says his plan is expected to generate $37 billion for the state, thanks to the aggressive schedule of toll increases. My source said that when they ran their valuation model using the Governor's toll increase schedule, the value totaled $50 billion. To get the same $37 billion value, their model showed tolls increasing only 3.3 times over the next 15 years, compared with 5 times under the Gov's plan. Alternatively, star ting with an even more modest toll rate schedule (today's rates adjusted annually for CPI increases), the model yields a valuation of $18 billion.
So what's with the Governor's numbers? My source could only speculate, citing their own assumptions about "substantial operational savings" and life-cycle cost savings in the modernization program. In other words, the proposition here is that the Gov's plan requires significantly higher tolls because it assumes business-as-usual in how the toll roads operate. It would be very interesting to see what kinds of bids the private sector might submit, if outrage over the Gov's proposed toll increases leads to a search for alternatives.
Over the last decade there have been all kinds of surveys on tolling and road pricing. While the results are generally positive, they can vary enormously depending on how the questions are worded and how much explanatory information is provided. The most difficult idea for people to grasp seems to be how variable tolling works to reduce congestion.
So I'm pleased to call your attention to the results of a survey carried out last December in King County, Washington. Two major road projects need replacing—the Alaskan Way Viaduct and the SR 520 floating bridge—and both would be multi-billion-dollar projects that are unlikely to be do-able without toll financing (neither has a toll at present).
Not too surprisingly, given the Seattle area's congestion, transportation and traffic congestion were the top choices (from a long list) as the single most important problem facing the region. The survey focused most of its attention on the 520 bridge. With no prompting, 68% either strongly or somewhat agreed that it needs to be replaced. After being given more specifics about its age and capacity, that number increased to 83%. When given a choice between a toll or a sales tax increase to pay for the project, 84% preferred a toll. In a similar choice between a toll or a gas tax increase, 78% opted for the toll.
The survey then introduced the idea of a variable toll, also described as "congestion pricing." Although only 28% had ever heard the term before, after a careful explanation, 75% strongly or somewhat supported the idea, and 63% supported that form of tolling for the new 520 bridge. They also could see the logic of putting such a toll on both Lake Washington bridges (I-90 as well as 520), to even out the traffic flows.
Finally, the survey tested a number of different arguments for variable tolling. Out of 10 such arguments, the four most persuasive were:
- It is popular with people, regardless of income, as found in San Diego. (76%)
- It reduces congestion, by shifting some trips to other times or other modes. (73%)
- It encourages non-commuters to travel at off-peak times. (70%)
- It reduces fuel consumption and vehicle emissions. (70%)
These are important and useful results. If you are even vaguely thinking about a variable-pricing project, you should get the full survey results from www.EMCresearch.com.
Politicians and pundits in the Northern Virginia/Washington, DC region are up in arms about a January 24th letter from Federal Transit Administrator James Simpson to Virginia's Gov. Tim Kaine, suggesting that the $2.6 billion Phase I of the Dulles Rail project is unlikely to meet FTA criteria for the requested $900 million in New Starts funding. The thrust of many of the comments is that, in addition to being unwarranted, the decision is a last-minute reversal. And that there is no good alternative to the ultimately $5 billion project to extend the Washington Metro out to Dulles airport and beyond. All three points are wrong.
Major doubts about the project have existed within FTA and elsewhere for years, but were brought to a head last July by the DOT Inspector General's office. I was unaware of this report until reading Simpson's letter, but it took only about 30 seconds to find it on the OIG website ("Baseline Report on Major Project Monitoring of the Dulles Corridor Metrorail Project," DOT OIG, Report # MH-2007-060, July 27, 2007) [www.oig.dot.gov/StreamFile?file=/data/pdfdocs/Dulles_Rail_Report_Final_072707.pdf]
The OIG report provides in-depth detail on the points raised in Administrator Simpson's letter, including the "significant cost growth and schedule slippages" already experienced, concerns about the project being managed by the local airport authority, and concern over the significant risks to the Washington Metro organization itself.
What I found most significant was the report's focus on the poor cost-effectiveness of the project. I had not realized that FTA has not followed its own current methodology for assessing the user benefits of the Dulles Rail project. Instead, it "grandfathered" this project under a no-longer-used method that includes time-saving benefits on other parts of the existing Metrorail system. OIG recommended that FTA reevaluate user benefits using the current methodology—and implied that this would lead to a "low" assessment of cost-effectiveness (when the project needs at least "medium-low" to qualify for New Starts funding).
As for the claim that there are no alternatives, that is simply not the case. More than a decade ago, some of the original corridor studies called for developing a bus-rapid transit corridor along the Dulles Toll Road right of way as the first phase, and if demand proved to be more than such BRT service could handle, "upgrading" the service to rail as an eventual second phase. That would have been a far more prudent approach.
Variations on this theme have been suggested since then. Peter Samuel recently proposed that Metrorail be extended only to the dense edge city of Tyson's Corner, but that the Dulles corridor itself be served by a transitway open to buses, vans, and taxis. He notes that the capacity of a single transitway lane in each direction could be several times that of the 4,800 people/hour projected as Metrorail ridership: 8,200 people/hour with a mix of buses, vans, and taxis and as much as 30,000 people/hour in an all-bus system capable of 1,000 buses/hour (if that much demand existed). Speed in the transitway would be much faster than on rail, because of all the station stops every train must make (whereas bus stations can easily be off-line, to permit nonstop express trips to key destinations such as the airport). [www.tollroadsnews.com/node/3368]
Another alternative was modeled in a Transportation Research Board paper by William Vincent and Gabriel Roth, "Comparison of Bus and Rail Transit Modes for the Dulles Corridor," Nov. 15, 2005, Paper No. 06-2396, available on the TRB 2006 Annual Meeting CD-ROM. They compared both a busway/BRT option and an express toll lane (ETL) option with the Dulles Rail plan. Using the earlier Dulles Rail cost numbers then available ($4.36 billion for Phases I and II instead of today's $5.1 billion), and making what appear to be reasonable capital and operating cost assumptions for the BRT and ETL alternatives, they found that while the cost per new transit trip generated would be nearly $34 for rail, it would be $9-12 for BRT and about the same for ETL. The disparity would be even greater if the current $5 billion cost had been used for Dulles Rail.
So to repeat, yes, there were plenty of previous warning signs that the rail project would be hard to justify. And there are far more cost-effective alternatives out there that could be pursued. A reasonable compromise would be a short extension of Metrorail to Tyson's Corner plus a BRT/ETL system in the middle of the Dulles Toll Road right of way.
Most discussion thus far on CO2 emissions from surface transportation has focused either on reducing the amount of driving (reducing vehicle miles of travel—VMT) or reducing the carbon-intensity of the vehicle's propulsion system (downsizing the vehicle, using alternative fuels or powerplants, etc.). Yet it turns out that another factor is potentially quite important, and is subject to near-term reductions. As Matthew Barth and Kanok Boriboonsomsin of UC Riverside point out in a paper from the 2008 Transportation Research Board meeting, CO2 emissions from conventional vehicles relate directly to fuel consumption, and the latter is very sensitive to the type of driving that occurs. As they say in the paper, "traveling at a steady-state velocity will give much lower emissions and fuel consumption compared to a stop-and-go driving pattern . . . that is associated with congested traffic ." ("Real-World CO2 Impacts of Traffic Congestion,"on the TRB 2008 Annual Meeting CD-ROM)
Barth and Borboonsomsin (hereafter B&B) more than 10 years ago developed a Comprehensive Modal Emission Model (CMEM), with support from the EPA and the National Cooperative Highway Research Program. It's a microscale model that can predict vehicle fuel consumption and emissions based on various traffic operations; it has been integrated into a number of transportation planning models (such as CORSIM) to permit corridor-level analyses. For this paper, B&B examined the velocity patterns of vehicles operating at various speeds under different level of service (LOS) conditions, using data from Southern California freeways. From this, they found that a plot of CO2 emissions vs. average speed is a parabola, with very high emissions below 20 mph, the lowest emissions between 20 mph and about 50 mph, and gradually rising emissions at higher speeds. Thus, they conclude that serious congestion le ads to much-increased CO2 emissions, but so does very high speeds. Consequently, they suggest that near-term CO2 reductions could be achieved via congestion mitigation measures at the low-speed end and speed-management techniques at the high-speed end. "Ideally, CO2 emissions can be reduced as much as almost 45% if traffic flow is smoothed to a steady-state condition," they estimate.
Amazingly, however, while they focus on such congestion mitigation techniques as ramp metering, and speed-management techniques such as stricter enforcement and variable speed limits, there is not a word in the paper about congestion pricing. While I understand the phenomenon of academic compartmentalization, this omission seems especially odd since UC Riverside is not that many miles away from the Riverside Freeway (SR 91), on 10 miles of which sits the 91 Express Lanes. Those lanes have been using congestion pricing for 12 years to eliminate stop-and-go conditions and keep traffic flowing smoothly at all hours of the day and night. While it is obviously a big step to go from pricing optional-use express lanes to pricing all lanes of a freeway, in an academic paper looking at alternatives you'd think this proven approach would at least deserve a mention.
One of the problems with expanding urban highway capacity—especially with elevated roadways—is the noise impact on surrounding areas. Experience has led us to equate freeways with noise—lots of it. Residential property values decline the closer you get to a major highway.
That's why I really paid attention when an old friend emailed me on his return from Tokyo last fall. Referring to the extensive system of elevated tollways, he and his wife were amazed at how quiet they are. My friends live in McLean, Virginia, over a mile from the Beltway—and they hear the noise from it at night. But in Tokyo, the noise level from the elevated tollways was so low as to be practically inaudible from two blocks away. Observing the sound walls from moving vehicles, they judged them to be of double-wall construction, with what appeared to be insulating material between the two walls, and topped with a rounded edge. Instead of reflecting noise, the sound walls seemed to be absorbing it.
I had to learn more about this, and thanks to a colleague at the Japan International Transport Institute in Washington, DC, I have learned a good deal more. While I was unable to get any statistics on the extent of motorway sound walls, I received several graphics files illustrating the many variations in design—and some information in English. Japan Highway Public Corp (the largest toll road operator) calls their system Unified Noise Guard. Their systems are made up of two types of panel—acoustical panels (which feature high noise absorption) and translucent panels. Some sound walls consist entirely of the former, but others have portions that are translucent. The total thickness of the acoustical panels is 20-22 inches; heights range from 6.5 feet to as much as 26.5 feet, and some of the illustrations show panels that curve partway over the roadway.
I won't go on at any greater length except to say that it's clear that, with Japan's much higher population density, roadway noise reduction has been pursued far more aggressively than it has in the United States. It may also be the case that since most of these expressways are tollways, it has been easier to pay for high-quality noise barriers as a standard design feature. Given our need to expand urban expressway capacity, and the potential of elevated express toll lanes, it seems to me we have an urgent need to get up-to-speed on what our Japanese colleagues have learned about making expressways better neighbors by greatly reducing their noise impacts. It may well be time for the National Cooperative Highway Research Program to organize a study tour, so that much of this technology can be incorporated into U.S. urban highway practice.
(If you'd like to see the graphics files I mentioned above, just send me an email and I'll forward them to you.)
Toll Roads 2008: Successful Strategies for Financing Toll Road Infrastructure Through Public-Private Partnerships, Feb. 26-27, Arlington, VA.
I'm speaking at this day and a half-long conference, at the Hotel Palomar Arlington. Register online at www.worldrg.com or call 800-647-7600.
Investing in Infrastructure Assets, Americas 2008, April 16-18, New York, NY
My Reason colleague Shirley Ybarra will be taking my place on a first-day panel at this event, which is focused heavily on infrastructure investors, including pension funds. Venue is Bridgewaters, atop the Fulton Market Building in Manhattan. Register online at www.terrapinn.com/2008/iiaa or email email@example.com.
For Whom Should Roads Toll? The Future of Toll Roads and Road Pricing in California, May 2, Ontario, CA.
I am one of the speakers at this event, being held at the Hilton Ontario Airport and sponsored by the Leonard Transportation Center at Cal State University San Bernardino. Go to http://leonard.csusb.edu/news/May 22008TransportationForum.htm. Or contact Rusty Thornton at firstname.lastname@example.org.
Synthesis Report on Toll Road Revenue. The National Cooperative Highway Research Program has a new synthesis report on the timely topic of "Estimating Toll Road Demand and Revenue." It's NCHRP Synthesis #364 and is available from the Transportation Research Board (http://gulliver.trb.org/bookstore).
Boycotting Earmarks? It may be too soon to judge it a critical mass, but nearly two dozen members of the Senate and House have vowed not to seek any new earmarks for their districts—a reaction to outrage over recent boondoggles such as Alaska's "bridge to nowhere." Senate members include John McCain (R, AZ), Jim DeMint (R, SC), and Tom Coburn (R, OK) along with Sen. Claire McCaskill (D, MO). I hope their numbers multiply enormously.
PPPs Judged Better in Australia. A study by Allen Consulting Group and the University of Melbourne found that privately financed infrastructure projects in Australia cost less and are completed more punctually than those procured in the conventional way. It compared the cost and schedule performance of 21 PPP projects and 33 conventional projects. On average, the PPP projects were 31% less costly at completion, compared with conventional projects. And in terms of schedule, the PPP projects came in 3.4% early compared with 23.5% late for the conventional projects. (Source: Public Works Financing, January 2008, p. 18)