- Atlanta looks at truck toll lanes
- Paying for better trips
- Financial community and toll roads
- Transportation system vulnerability
- Quotable Quote
Georgia's State Road & Tollway Authority (SRTA) is thinking outside the box. Given horrendous traffic congestion and a current political consensus not to add new freeways, how best can the metro area optimize the use of existing freeway rights of way? Is there a way to do this that could produce meaningful reductions in congestion, rather than just barely holding the line on it getting worse between now and 2030? In addition to studying the potential of HOT lanes, in a report released earlier this year, SRTA has recently released its new study of TOT (truck-only toll) lanes. It's a real eye-opener.
The idea SRTA gave to Parsons Brinckerhoff to examine was this: should some of the available right of way for expanding the freeway system's capacity be devoted to TOT lanes rather than HOV or HOT lanes? That's a good question to ask, given that Atlanta is the trucking crossroads of the Southeast, with I-75 and I-85, both of them major truck routes, intersecting in downtown Atlanta. Heavy trucks don't go through that mess; they are already constrained to go around downtown on the I-285 "perimeter," which as a result carries 23,000 trucks per day. So the study looked at three possible scenarios for incorporating TOT lanes into the system, estimated the time savings for trucks and cars, and estimated the costs and revenues, at a very general level.
Before telling you the results, I want to point out a very significant aspect of SRTA's study. Unlike some proposals for tolled truck lanes, they did not assume that trucks would be forced to use the toll lanes. Rather, TOT lanes would be an option, to be used by those truckers valuing the time savings and increased reliability more than the amount of the toll. With this as a ground rule, the Georgia Motor Trucking Association willingly cooperated with the study team, and has said positive things about the results. I've been saying for several years that if tolling is going to make headway in America , it's got to be done on the basis of value-added: offering highway customers a better deal in exchange for paying a toll. I'm pleased to see this idea being taken to heart in Georgia.
Overall, the results were very promising, so positive that SRTA executive director Doug Hooker has said that the truckers' positive reaction, combined with the importance of Atlanta as a trucking hub, provide "a compelling reason to pursue the concept of truck facilities in the region." The study modeled three scenarios for TOT lanes:
- Alternative 1 would add two TOT lanes in each direction on I-75 north and south of the perimeter and on the perimeter (I-285) itself; this would be in addition to planned HOV or HOT lanes on those freeways.
- Alternative 2 would be the same as Alt 1, but would allow light delivery trucks to use existing HOV lanes inside the perimeter, for a toll, during off-peak hours (10 AM to 3 PM).
- Alternative 3 would develop the TOT lanes of Alt 1 instead of planned HOV/HOT lanes on I-75, 85, and 285.
Overall, Alt 3 looks like the winner, in terms of most bang for the buck. Because it involves less new lane construction than the other two (because TOT lanes substitute for some of the planned HOT lanes), its estimated construction cost is only $578 million (vs. $909 million for Alts 1 and 2). It produces the greatest annual (year 2030) revenue, at $198 million (vs. $89 million and $101 million, respectively). As Peter Samuel points out in his article on the study (www.tollroadsnews.com), Alt 3 looks easily financeable, with an out-year revenue of $198 million a year on a capital investment of just $578 million.
That revenue number is based on the model's estimate that, thanks to time savings, the TOT lanes would attract 60% of the truck traffic away from the regular lanes. Alt 3, for example, would save a truck 70 minutes getting through/around Atlanta in the afternoon peak period. And that means benefits for car drivers, as well. Total weekday vehicle hours traveled would be 6.5% less, and the number of congested freeway lane-miles in 2030 would be cut from 29% in the base case to between 22% and 24% in the various TOT lane scenarios.
One idea not evaluated in the Atlanta TOT study, as Samuel notes, is the impact of allowing long doubles and triple-trailer rigs on the TOT lanes. These longer combination vehicles (LCVs) are not currently allowed on Georgia highways, but could dramatically increase the productivity of trucking by allowing a single driver to haul 50 to 100% more payload. Reason's 2003 assessment of the most promising long-haul corridors for such toll truckways identified I-75 from the Florida Turnpike through Atlanta to Tennessee and on to Detroit as one of the 10 best candidate routes in the nation (see www.reason.org/ps316.pdf). This would be a good topic for SRTA to address in a follow-up study.
Note: at press time, the TOT study was not yet posted on the SRTA website, but I'm told it will be there soon. Check on it at: www.georgiatolls.com.
When HOT lanes or Express Toll Lanes are being considered, one of the key questions is what rates will be charged. Elected officials, not quite grasping the need for market-clearing prices in order for these lanes to do their job, hope these rates will be low, while most transportation planners (especially if they hope to bond the toll revenues to pay for the new lanes) hope they will be as high as possible. In some recent feasibility studies, I've seen figures as low as 5 to 10 cents per mile, and seldom more than 30-40 cents/mi. Yet our best empirical evidence, from the two California projects with nearly a decade of experience actually doing variable pricing, suggests some people will pay a lot more.
The latest review of what we've learned from these projects is by David Brownstone and Kenneth Small of UC Irvine. "Valuing Time and Reliability: Assessing the Evidence from Road Pricing Demonstrations," appears in Transportation Research Part A: Policy and Practice, Vol. 39, Issue 4, May 2005 (available online at www.sciencedirect.com). It builds on Small's earlier work with Cliff Winston and Jia Yan that I reported on in Issue #17 of this newsletter last year.
Brownstone and Small review and compare a number of recent studies of the 91 Express Lanes and I-15 Express Lanes. All the studies involved detailed surveys of commuters in the respective corridors, both stated preference (SP) surveys where people are asked what they would pay under hypothetical conditions and revealed preference (RP) surveys, where they report what they actually paid when they made the decision to drive on one of the HOT lane facilities. The review finds the studies to be generally consistent in their findings.
There are many useful findings here, but I will focus on what seem to me to be the most important take-aways for planners and policy makers. First, the median value of time (VOT) found in most of these studies was in the $20-25 range (but as high as $40-45 in two of them). That's at least twice as high as people said they would pay, in SP surveys of the same general population. So don't rely on SP results to estimate toll levels for HOT lanes. Watch what people do, not what they say.
Second, there is a great deal of "heterogeneity" in the VOT figures. As measured by the inter-quartile range from the median, for example, 25% of the group whose median VOT was $25 would pay $38; in another study, 25% of those whose median VOT was $30 would pay $50. So on a HOT lane with limited capacity to sell (because most of it is being given away to HOV-2s), we would expect to be able to charge very high tolls to a small subset of the overall corridor commuter group.
The third important result is that for the small number of studies where VOR was measured, it was in the same ballpark as VOT. In other words, the total toll that people would be willing to pay is the sum of what they will pay for time savings plus what they will pay for a reliably on-time trip. One study found VOR to be equal to VOT, each at $20/hour. Even more fascinating, the two that broke out VOR separately for males and females found that women were willing to pay twice as much as men for a more reliable trip: males' VOR was $12 and $15/hr. in the two studies, whereas females' was $30 and $32/hr. And remember, these median VOR figures must be added to the median VOT figures to get a total rush-hour toll.
We certainly need more research on the value of reliability, and it will help when we have real data from HOT lanes in other parts of the country where congestion is not as horrendous as in Southern California. A few years from now, when new HOT lanes in Denver, Houston, and Minneapolis have been in operation for several years, there will be mountains of additional data for grad students to pore over. While I doubt if any of them will be charging 50 to 70 cents/mile during their busiest peak hours (as is now the case on the California projects), I'm sure they will be charging far more than the 10-15 cents/mi. now being used in some studies.
"U.S. Toll Road Privatization" as an investment opportunity is the subject of a new report from Merrill Lynch, dated July 29, 2005. Taking off from the much-discussed 99-year lease of the Chicago Skyway, the report reviews the handful of investor-built new toll roads of the past decade, notes Transurban's recent proposal to take over the troubled public/private Pocahontas Parkway, and discusses large state toll authorities such as the New Jersey Turnpike Authority and the New York State Thruway Authority as potential privatization candidates. Merrill Lynch points out that "The level of interest from large international infrastructure developers and operators in acquiring and operating existing U.S. toll roads is on the rise." It notes that the public-private partnership (PPP) or privatization structure, with long-term private ownership and operation of the toll road, "is gaining appeal as an alternative to traditional models of tax-exempt financing, ownership, and operation of the nation's roadways." And it goes on to identify characteristics of a good privatization target and other considerations that are important to getting a deal done. (www.ml.com)
Broader context on the overall toll road industry can be found in a recent report from Fitch Ratings, "The Global Toll Road Credit Landscape," dated May 16, 2005. It provides a very useful "world tour of road finances," with summaries of toll road developments in Asia, Australia , Europe, Latin America, and the USA. One of the most interesting conclusions is that " Europe is now poised for an explosion in new toll road projects, both in mature and emerging economies." Also useful is its overview of the Australian experience, in which nearly all the new urban expressways of the past 15 years have been developed as long-term toll road concessions. On the United States, Fitch is cautious, noting the trend toward more toll projects but pointing out what it sees as the political unpopularity of tolls and the relatively small fraction of highway mileage funded by tolls thus far. (www.fitchratings.com/corporate/reports/report_frame.cfm?rpt_id=239580)
The horrible London subway bombings last month led to lots of media and political discussions about the need to do more about transit security. Homeland Security Secretary Michael Chertoff was subjected to unwarranted criticism when he pointed out that providing airport-type screening at transit stations was wildly impractical. But in all the discussions about terrorism and mass transit, I didn't see anyone raise the question of the relative vulnerability of the urban highway system and rail transit. Yet this is clearly a point worthy of some thought.
There is a long-running debate in this country about where we should be making capital investments in urban transportation; some favor adding highway capacity while others would shift most of what are now highway funds (resulting from fuel taxes paid by motorists) into building rail transit. There are many points to debate on costs, benefits, environmental impacts, etc. on expressways versus rail lines, which I won't rehash here. But in the age in which we now live, one additional factor to consider is comparative vulnerability to terrorist attacks.
As we saw in London , a bomb in a subway halts service on that line, possibly for a long period of time. To be sure, the London system is something of a network, so riders have some degree of ability to route their trips around the out-of-service line. But even the most ambitious U.S. rail systems, like the Bay Area's BART and the Washington, DC Metro, are not networks. If a line goes out, that's it for its users, until it can be repaired and put back in service.
Now consider a freeway or expressway system. In Los Angeles , whose freeway system is much less of a dense network than many others, the Northridge earthquake knocked out a section of the Santa Monica Freeway. It was a huge inconvenience for the several months it took (round the clock) to rebuild it, but traffic flow was not seriously disrupted. That's because the freeway is part of a much denser network of streets and arterials that can take over from it. As with the Internet, traffic can easily route around a blockage.
Defenders of the auto/highway system have always made much of the idea of auto-mobility´┐Żthe ability of people to go where they want, when they want to. Less vulnerability to terrorist disruption is a thus-far under-appreciated aspect of auto-mobility.
"Now, I grant you, the idea of investor-owned highways might strike some of you as odd. I suspect that's because it's unfamiliar. But think about it. It is no more radical than investor-owned electric utilities or investor-owned telecommunications firms. All are vital elements of infrastructure that people use every day—just like roads and bridges and other transportation systems."
— Speaker William J. Howell, Virginia House of Delegates, at Fredericksburg Regional Chamber of Commerce Transportation Forum, June 14, 2005