In this issue:
- Time to get rid of toll booths
- Transit system report shows progress—and limits
- Stopping sprawl to "save the planet"
- Federal grant for a toll bridge?
- Cost-effective four-laning
- New congestion measures in Atlanta
- News Notes
- Upcoming Conferences
- Quotable Quotes
For the past five years, I've been urging that owner/operators of toll road and bridges set aggressive deadlines for eliminating toll booths and toll plazas, those relics of 20th-century tolling that motorists hate, for good reasons. Ever since Toronto's Highway 407 opened in 1997, with a combination of transponder tolling and video (license-plate-reading) billing, we've known this was possible. Seven years ago I drove on the then-new Melbourne CityLink in Australia in my Avis rental car, whose license plate number had, at my option, been entered into the toll road's system so I'd be recognized as a customer, not a violator. The Cross-Israel Highway and Santiago, Chile's new urban toll motorways all operate without any toll booths or plazas. What's taking America so long to get with the program?
Fortunately, the last few months have seen a spate of announcements of plans (or at least studies to plan) for phasing out toll booths. Planning for this is already under way at the Florida Turnpike Enterprise and the Miami-Dade Expressway Authority (whose target date is 2012). In August, the North Texas Tollway Authority in Dallas announced that toll booths on its system would be gone by 2010. And studies are under way at E-470 in Denver and the Port Authority of New York and New Jersey to do likewise.
In most places, I expect these transitions to go relatively smoothly. We've had specialized toll roads operating without toll booths for more than a decade in California (the HOT lanes on SR 91 and I-15), and more recently on the Westpark toll road in Houston and the new elevated express toll lanes in Tampa. People love the speed and convenience of "paying tolls" at 65 mph.
But then there's Massachusetts. A recent report by the state's Transportation Finance Commission recommended phasing out toll booths on the Massachusetts Turnpike, and eventually moving to a per-mile charging system for all highways in the state. But the report was immediately attacked by the ACLU of Massachusetts and opportunistic politicians. "It's like Big Brother. It's having government watching everywhere you are going," said Rep. David Torrisi (D, North Andover).
Privacy will always be an issue with electronic tolling. But experience everywhere, thus far, has shown it to be no big deal. Anonymous cash-based transponder accounts have been offered in a number of these systems—with take-up rates at fractions of one percent of all toll road customers. The vast majority of people accept the convenience of the electronic system as worth leaving behind an electronic transaction record—just as they do every day via ATMs, credit cards, and debit cards. And if politicians in one or two states intervene to prevent this shift, it will be one more factor that individuals and businesses use to determine the best places to live and work.
A subtext in the long-running battles over what kind of mass transit makes sense for America's urban areas is that transit proponents are seen as making promises they don't keep. Back in 1990, a researcher at the U.S. DOT's Volpe Center, Don Pickrell, produced a much-cited report on the underwhelming performance of new rail systems. Their costs were mostly way over what had been estimated at the time the decision was made to go forward. Even worse, their ridership numbers were far below projections. None of the 10 federally supported rail projects achieved their ridership goal, and nine of the 10 achieved less than 50%.
That was then, of course, and this is now. For several years I've heard scuttlebutt that a new report along these lines had been completed by the Federal Transit Administration, but it never seemed to appear. Now I'm pleased to confirm that this study exists, and was released to Congress this week; it should be on the FTA website shortly. What you need to look for is something called "Contractor Performance Assessment Report," August 2007. But the real deal is its Appendix, a 191-page report called "Predicted and Actual Impacts of New Starts Projects." And true to what the rumor mill had suggested, the date on this report is September 2003.
Be that as it may, what does the report find? It reviews 21 transit projects that received New Starts funding from FTA, including heavy rail (e.g., Atlanta's North Line MARTA extension, LA's Red Line subway), light rail (e.g., Portland's Westside line, San Jose's Tasman West), people-mover projects in Miami and Jacksonville, and four busway projects. Key measurements were capital costs, operating & maintenance (O&M) costs, and ridership. In each case, the actual (as-built) number is compared with the projected number at the point of (a) Alternatives Analysis/Draft Environmental Impact Study, (b) Final EIS, and (c) signing of the Full Funding Grant Agreement (FFGA). In the case of costs, they were adjusted for inflation, with capital costs adjusted using the Engineering News-Record Construction Cost Index. The idea was to make this as much as possible an apples-vs.-apples comparison.
The good news is that although there is still a strong pattern of cost overruns, it is significantly less than for the earlier rail projects documented in the Pickrell report. The average capital cost of these projects was 121% of what was projected at the time the decision was made to go forward (at the AA/DEIS stage). Five of the 21 actually came in at less than projected, one (San Jose's Tasman West) costing only 72% as much. But for the large majority with overruns, they ranged from 108% of projected (Atlanta) to 172% (Portland).
The news on O&M costs is far more positive. For the 18 projects for which O&M data were available, the average was 98% of what had been projected. Seven of those projects did exceed projections, by a little (102% for St. Louis MetroLink) to a great deal (San Jose's Guadalupe light rail at a whopping 176%).
What about ridership? To me this is really the acid test, because only if transit can actually attract large ridership can it possibly make a meaningful difference on outcome measures such as reduced traffic congestion, energy savings, and environmental impact. Here, unfortunately, the numbers are pretty dismal. In each case, the researchers compared actual average weekday boardings in the transit agency's forecast year with projected boardings for that year, made at the AA/DEIS decision point. The average for all 19 projects for which data were available is 65%. Only three exceeded their projections (by between 1% and 34%), and the range among those falling short is very wide—from a low of 6% (Jacksonville people mover) to many others in the 40%-60% range, with others in the 70-80% range. And even some of those with fairly high percentages only achieved them by aiming low: BART's Colma extension got 86% of what it projected, but that amounted to only13,060 weekday boardings, for a very costly heavy-rail line; likewise for Baltimore's heavy-rail Johns Hopkins extension, averaging only 10,128 weekday boardings.
Whatever the numbers are, it's long overdue that the federal government regularly account in this manner for how it is spending taxpayers' New Starts money. And while it's modestly encouraging that forecasts are better than they used to be, these new data are sure to enliven the continuing debate on whether new rail transit produces meaningful bang for the bucks it's consuming.
Several readers have taken me to task for devoting space to land use and climate change issues in what is supposed to be a transportation policy newsletter. But I was reminded of the nexus among these issues by a headline in the October 1st issue of Engineering News-Record, the journal of record for public works design and construction. "Stopping Sprawl Will Aid Planet," it read. It summarized the message of a forthcoming book from the Urban Land Institute, whose lead author is Reid Ewing of the University of Maryland's National Center for Smart Growth. As of this writing, it's not yet posted on the ULI website, but I was able to download the executive summary from www.smartgrowthamerica.org.
Based on that summary, everyone interested in the future of automobility should take a careful and critical look at this study when it appears. The basic idea is that since transportation is responsible for one-third of CO2 emissions in the USA, we are quite unlikely to meet any goal of dramatic CO2 reductions by 2050 unless we reduce vehicle miles traveled (VMT) significantly. Since there is only so much likely to be do-able about average fuel economy or fuel carbon content, it's essential that most new residential development be of the compact, smart-growth variety, so as to cut VMT.
Without having access to the study itself, I can already see some holes in this argument, and many questions deserving of close scrutiny. First, "transportation" includes not just on-road motor vehicles but aircraft, trains, ships in ports and on waterways, etc. VMT on highways does not account for one-third of CO2 emissions. Second, 2050 is a long ways from now in terms of the evolution of vehicle and propulsion technology. Practical plug-in hybrids, for example, would entirely change the calculations on which these prescriptions depend, since they would radically alter the relationship between VMT and carbon emissions. (Even today's Prius, though a tiny part of the fleet, produces half the CO2 of typical cars.) Third, I'd like to take a much closer look at how they arrive at the finding that "two-thirds of the development on the ground in 2050 will be built between now and then." That sounds highly unlikely to me, given that we'll be adding only 50 million people to the current 300 million (17%).
But getting back more specifically to VMT, it's not clear that there is such a straightforward relationship between compact land uses and reduced VMT, as that on which the authors build their case. This is a far more complex relationship than the executive summary (at least) implies. It's been studied at increasing levels of sophistication over the last 15 years, with mixed results. One of the most methodologically sophisticated papers is by Andrea Sarzynski, et al, in Urban Studies, Vol. 43, No. 3, March 2006. As the title makes clear, "Testing the Conventional Wisdom about Land Use and Traffic Congestion: The More We Sprawl, the Less We Move?" looks at congestion, not VMT per se. But its three measures of congestion (average commute time, average daily traffic per lane, and delay per capita) are certainly related to VMT, and their methodology could be applied to VMT instead of congestion.
Sarzynski and colleagues define seven distinct measures of land use, along with the three measures of congestion. Using data from 50 urban areas, from 1990 and 2000, they build a model that tests for the relationship between congestion in 2000 as a function of land-use in 1990 (on the premise that changes in travel happen a lot more rapidly than changes in land-use)—the first time that such modeling has built in a time lag like this. They first do bivariate correlations, most of which suggest the counter-intuitive result that denser land usage is associated with greater congestion. But since these relationships are complex, their more useful results come from their multiple regressions, in which they can isolate the impact of any one land-use measure. Most of these regressions produce the same general finding. Urban areas with higher density and continuity of development have higher subsequent traffic congestion, and areas with lots of housing near the central business district also have higher subsequent congestion. The only land-use measure that is associated with lower congestion is high housing-jobs proximity.
The point of my amateur critique is not to say that the ULI/Smart Growth analysis is wrong. All I'm saying is that it needs to be looked at very carefully, with the kind of rigor employed by Sarzynski and company, this time modeling VMT in relationship to land-use variables. (Note: the Sarzynski paper is not included in the References listed in the ULI/Smart Growth study's executive summary.)
Personally, I have no problem with compact, mixed-use communities as a choice that developers should be free to offer, and with urban infrastructure policies in which every type of land use pays its true costs (for water & sewer pipes, streets, etc.). But that's not what Smart Growth America and the National Center for Smart Growth want. On the last page of their executive summary, they call for the next federal transportation reauthorization bill to "require transportation conformity for greenhouse gases." That would very likely lead to measures like the "VMT bill" in California that I wrote about in Issue No. 47, denying transportation funds to urban areas that allow new development at low density levels.
Once again, I will repeat my point that if excess carbon emissions are the problem, the fairest and most economically efficient way to deal with the problem is to tax carbon emissions, no matter the source of those emissions. Piecemeal regulations on land-use, housing, fuels, propulsion systems, and driving restrict basic liberties and attempt to do via central planning what people's voluntary choices, in response to price incentives would do naturally.
I was checking on transportation developments in California when I came across a Sept. 29 news item from the San Francisco Chronicle. The Golden Gate Bridge was about to purchase a movable "zipper" barrier to replace the plastic cones currently used to separate northbound and southbound traffic between morning (4 lanes southbound, 2 north) and afternoon (the reverse). Over the years, a number of people have been killed or injured thanks to head-on crashes in this unsafe configuration. So in 1998, the Golden Gate Bridge District board approved the purchase of a state-of-the-art movable barrier from Barrier Systems, Inc.
But this is 2007, a full nine years later. Why hasn't the barrier been installed long before now? Apparently the $25 million cost was beyond the scope of the Golden Gate Bridge's budget. Say what? In case you'd forgotten, this is one of America's landmark toll bridges. The current toll is $5 round-trip ($4 if using Fastrak), charged only in the southbound direction. But despite 100,000 daily users, the Bridge District is running in the red. So the movable barrier purchase has been delayed until now, when the District received a $20 million federal grant for this purpose.
So where does all the toll revenue go? Into the District's other two operations, Golden Gate Transit (buses) and Golden Gate Ferry, both of which operate at large losses. Unlike a normal toll road, bridge, or tunnel, whose business is serving its toll-paying customers, Golden Gate spends its toll revenues subsidizing its transit operations and then has to go to Washington, DC for tax money for capital expenditures. I spent some time poking around on the District's website, and learned that their current capital program assumes (i.e., depends on) 80% federal funding for all major bridge rehabilitation projects.
For several years now, I've been debating people like Prof. Joe Giglio and the Hudson Institute who want to convert the entire highway system to tolls—and then put all that revenue into a giant cocktail glass, with straws for urban transit providers, passenger rail providers, even freight railroads (and oh yes, for the roads and bridges, too). That, I have maintained, would be a disaster for toll roads and bridges. A little preview of that world is on display in San Francisco.
Two-lane highways, with a stripe down the middle to permit passing slow vehicles, are a source of many deaths and injuries due to head-on collisions. Yet many of these highways don't really have the traffic levels to justify four lanes, especially when state highway budgets are stretched thin, and there are many unfunded projects that would yield greater transportation benefits. Yet political reality often requires upgrading such highways, either in response to publicized crashes or for "fair-share of the highway budget" reasons. What's a conscientious state DOT to do?
One promising idea is being tried in Missouri, according to a recent article in Urban Transportation Monitor (Aug. 3, 2007). They call it a "shared four-lane highway," in that it provides most of the functionality of four lanes, but at lower cost. The configuration provides for three 12-foot traffic lanes, in alternating two-lane and one-lane segments, with a 4-foot median between opposing lanes (and 7-foot shoulders on each side). Thus, motorists have regular safe opportunities to pass, without ever having to risk moving into a lane where opposing-direction traffic might be. Rumble strips will be added in the median, as an added safety feature.
MODOT's initial project is 18 miles long, on SR 5 between Lebanon and Camdenton (which connects I-44 to the popular resort area Lake of the Ozarks). Eight miles will be all-new construction, on a new alignment, while the other 10 miles will consist of widening the existing two-lane highway. Total cost is estimated at $44 million. I asked MODOT what a traditional four-laning of a two-lane, 18-mile highway would cost them, and they replied that the cost of a traditional four-lane divided highway at this location would be about double that.
For several years now, the Georgia Regional Transportation Authority (www.grta.org) has been producing an annual Metropolitan Atlanta Performance Report. The 2007 MAP Report, as it is called, was released in mid-summer. Like its predecessors, it includes quantitative measures of mobility, transit accessibility, air quality, safety, and transportation system performance. It's a good example of the kind of accountability motorists and taxpayers should be getting about their transportation systems.
This year's report includes two important new measures of congestion (in addition to the traditional travel time index, which is bad and continues to worsen). Both seek to measure aspects of the highway system's reliability. The Planning Time Index is the ratio of the 95th percentile of travel time to the free-flow travel time. The PTI tells the traveler how much longer it is likely to take during peak periods, to ensure on-time arrival 95% of the time. In 2002, the average morning PTI was calculated at 1.53; by 2006 it had increased to1.80; the afternoon PTI went from 1.83 in 2002 to 2.02 in 2006. The report's Table 2 shows the current PTI for each direction of 16 major freeway segments. In the morning, the two highest PTIs are 3.14 and 3.17, both on portions of I-75. In the afternoon, the worst of all are 3.45 and 3.90, both on portions of I-285.
The other new measure is the Buffer Time Index (BTI). It's a different way of communicating basically the same idea, defined as the difference between the 95th percentile travel time and the average travel time, expressed as a percentage. BTI has also increased since 2002, with the morning average going from 26.1% in 2002 to33.8% in 2006. With the afternoon's worse congestion, that BTI was already at 41.7% in 2002, increasing to 42.9% in 2006. Here again, the average mask some extremely congested segments. In the morning, there are BTIs of 94% and 76% on two segments of I-285, and in the afternoon a section of I-75 northbound reaches 142%.
One hopes these new measures will lead to stronger support for the kinds of market-based capacity enhancements recommended in Reason Foundation's Atlanta congestion-reduction study last year (www.reason.org/ps351.pdf).
Galvin Prize to End Congestion. Reason Foundation's Galvin Project to End Congestion, in partnership with the Institute of Transportation Engineers, is sponsoring prizes for new engineering solutions and technologies to facilitate the construction of new roadway capacity in urban areas. Up to $40,000 will be awarded, primarily for publishable research papers. Two-page expressions of interest should be submitted by Nov. 1, 2007 to firstname.lastname@example.org. Further information is available at www.reason.org/innovationsprize.
Tampa Elevated Express Toll Lanes Anniversary. The innovative reversible express toll lanes on Tampa's Crosstown Expressway celebrated their first anniversary in July. The project is getting rave reviews from motorists, who can zip into downtown Tampa from its eastern suburbs at 65 mph at rush hour. Traffic is running about 25% ahead of projections at over 16,000 vehicles per day. The elevated tollway is Florida's first without any toll booths, using transponders and license-plate reading instead. And thanks to its innovative design—using precast segments 60 feet wide—the entire thing is supported by central pillars just six feet wide, for which there was plenty of room in the Expressway's median.
California "VMT Bill" on Hold. Last issue I wrote about SB 375 on greenhouse gas reduction, which, among other things, would withhold transportation funding from cities that don't adopt high-density rules for all new housing. Shortly after the newsletter went out, the Sacramento Bee reported that sponsor Sen. Darrell Steinberg has removed or modified that portion of the bill and will wait till next year to try to get the revised measure through both houses of the legislature.
Excellent Report on Aging Infrastructure. Congressional Quarterly publishes something called CQ Researcher (www.cqresearcher.com). Its Sept. 28, 2007 issue includes a long, well-done report, "Aging Infrastructure" by Marcia Clemmit. It does more than document the problems with infrastructure that is not being properly maintained; it also shows how incentives resulting from current laws and policies lead to that outcome. I was one of a number of transportation researchers interviewed for this report, and I commend it to your attention.
Built-in Transponders in All New Cars? I've long expected that one of the early results of the ongoing industry/government research effort known as Vehicle Infrastructure Integration (VII) would be that new cars would start rolling off assembly lines with built-in electronic toll collection transponders, built to a national standard. That hasn't happened yet. But an article from The Economic Times of India (Sept. 25, 2007) crossed my screen the other day. It reported that the Ministry of Heavy Industry is considering a proposal to require radio frequency ID transponders to be built into all cars built in India. Motor vehicle growth is exploding in India (as in China), and in recent years the country has embraced toll roads and long-term concessions.
HOT Lanes as Private Toll Concessions. That's the title of a chapter I wrote for a book just out from Project Finance International (www.pfie.com) called Infrastructure Finance: The Road Ahead, edited by Rod Morrison. In the chapter I draw on a paper I presented at the Institute of Transportation Engineers technical conference in San Diego last spring. I contrast two very different models of HOT lanes, showing that key policy decisions make a huge difference in their financial viability and hence in their suitability for being developed as private toll concession projects.
Reason Foundation is a co-sponsor of several upcoming conferences, and my Reason colleagues and/or I are speaking at all of them. Here are the key details:
19th Annual ARTBA Public-Private Ventures, Nov. 1-2, Washington, DC
This is the premier conference on public-private partnerships in surface transportation.
Hilton Washington Hotel, www.artba.org/meetings_events/2007/PPV.
2007 Preserving the American Dream Conference, Nov. 10-12, San Jose, CA
This year's theme is "Recovering from Smart Growth."
Wyndham Hotel, http://americandreamcoalition.org/pad07.html.
IBTTA Transportation Finance Summit, Dec. 2-4, Washington, DC
Two intense days on tolling and finance trends. Previous ones have been excellent.
Fairmont Hotel, www.ibtta.org/Events/eventdetail.cfm?ItemNumber=2398.
North American Port & Intermodal Finance & Investment Summit, Dec. 3-5, Miami, FL
Includes port, rail, and roadway infrastructure.
Westin Collonade Hotel, Coral Gables, www.infocastinc.com/ports.html.
"I encourage you to examine the fundamental question of why states are looking to engage the private sector in the first place. . . . I am not sending my fellow citizens' hard-earned money to Washington, DC to have it redistributed, earmarked, and locked into programs that do little to relieve congestion. . . . In the absence of radical changes to the size, scope, and dependability of the federal government's support for the national transportation system, Texas and other states will have no choice but to seek new solutions to their transportation problems, including working with the private sector."
-- Gov. Rick Perry, response to May 10, 2007 letter to governors from Congressmen Oberstar and DeFazio that questioned the wisdom of public-private partnerships.
"Three broad factors inhibit efficient use of roads and highways . . . . First, many roads were not designed to meet current traffic volumes. . . . Second, the federal and state transportation revenue-raising structure, which collects the majority of revenues through motor fuel taxes and other user fees, does not provide incentives for drivers to take into account the external costs, such as increased travel time for other drivers, they impose . . . . Third, there is a limited focus in the current decision-making process on selecting projects that produce the highest net social benefits."
-- JayEtta Hecker, GAO, "Surface Transportation: Strategies Are Available for Making Existing Road Infrastructure Perform Better," GAO-07-920, July 2007.
"[W]e have an increasingly flawed investment model and a system performance crisis. . . . The underperformance in the highway sector is fundamental, not incremental. In other words, an increase in Federal taxes and spending would likely do little, if anything, without a more basic change in how we analyze competing spending options and manage existing systems more efficiently. . . . The degree to which one capital investment generates more returns than a competing investment is the most basic question asked in virtually every other capital intensive sector of the economy. Yet, when it comes to . . . highways and bridges, there is virtually no analysis of this question."
-- DOT Secretary Mary Peters, testimony before House Committee on Transportation and Infrastructure, Sept. 5, 2007.