- Are we headed for a traffic apocalypse?
- A closer look at hyperloop feasibility
- Electric trucks’ infrastructure needs
- Concerned scientists vs. ride-hailing
- How many crashes will AVs eliminate?
- More states testing per-mile charges
- Upcoming Transportation Events
- News Notes
- Quotable Quotes
As car travel trended upward in May, we began to see scary headlines, quoting various transportation experts predicting that once the coronavirus pandemic is over, we may see much worse congestion on urban roadways due to many people changing how they commute. Perhaps the most notable paper attempting to quantify this is “Impacts on COVID-19 Mode Shift on Road Traffic,” by Ph.D. candidate Yue Hu and professors from Cornell and Vanderbilt. They built a model and crunched numbers for 100 metro areas to show what might happen due to changes in travel choices, such as:
- Shifting from transit to driving alone (SOV—single occupancy vehicle)
- Shifting from carpools to SOV
- Increasing work at home (telecommuting).
The paper uses a widely accepted traffic model (the BPR model) and predicts very large increases in congestion and travel time for most metro areas. In addition, the authors have provided a handy online calculator that enables you to select any of the hundred metro areas and use sliders to set different levels of possible shifts in the above commuting behaviors.
For context, here are the latest journey-to-work data for 2018, which represent the status-quo ante:
|Drive alone (SOV):||76.3%||, holding steady in recent years|
|Carpool:||9.0%||, in a long-term decline over past 20 years|
|Transit:||4.9%||, declining over past 5 years|
|Telecommute:||5.3%||, in a gradual uptrend over past decade|
|Other (bike, walk, etc.)||4.5%||, stable over past decade.|
These are national averages, and apart from a handful of metro areas that still have a significant fraction of jobs in the historic central business district and lots of transit (New York, Chicago, San Francisco, and a few others), most metro areas have smaller transit and carpool shares than the national averages (which include the handful of transit-intensive metro areas).
When you play with the online calculator, you will be surprised to see that fairly large changes in the adjustable variables don’t seem to have much impact. For example, I selected Houston, a very high-auto, low-transit metro area, and selected various combinations, as follows:
|Case 1||Case 2||Case 3||Case 4|
|Change in SOV||10%||10%||5%||10%|
|Change in carpool||10%||10%||5%||10%|
|Change in telecommute||0%||10%||10%||20%|
|Increase in travel time (min.)||30.6||28.8||28.7||27.5|
As you can see, the model appears to have a built-in bias toward increased congestion. I think this is because the researchers confused percent increase and percentage point increases. Both telecommuting and transit are small fractions of commuting, so what is important is not their percentage change, but the percentage point changes. Many telecommuting experts suggest that it is likely to at least double, from 5.3% of the commute to 10.6%—a 100% increase. But the sliders on the model allow telecommuting to increase only up to 24%, not 100% or potentially more, given that one-third of all jobs could be performed remotely full-time. And most full-time telecommuters are (and will be) college-educated professionals, as opposed to service workers or skilled tradespeople. Those professionals overwhelmingly commute either by car (SOV) or rail transit (where available), so mostly by SOV. Thus, a mere doubling of telecommuting’s mode share in a metro area would, ceteris paribus, shift up to five percentage points from SOV to telecommute—i.e., from the current SOV average of 76.3% to 71.3%.
Needless to say, the freed-up space on freeways would attract some traffic back into those lanes. But how much of that would be from transit?
If 10% of transit riders switched to SOV, that would add only 0.5 percentage points to the SOV mode share. And if 10% of carpoolers switched to SOV, that would add 0.9 percentage points to SOV’s mode share. It’s hard to see massive increases in congestion from that.
I think the modelers of this paper owe readers an explanation of why they grossly under-estimated the likely large impact of increased telecommuting on traffic congestion. It looks to me that predictions of a traffic apocalypse are quite unlikely.
Hyperloop—passenger and cargo pods whizzing between cities in near-vacuum tubes at more than 600 miles per hour—is an exciting transportation concept. But as a degreed engineer with a lot of background in economics, I have two basic questions:
Is it hyperloop technically feasible today and would it be cost-effective (e.g., commercially-viable, or at least having benefits greater than costs)?
The answer to the first question is, potentially, yes. We know how to make tubes with very low air pressure, and magnetic levitation and propulsion have been demonstrated (though not commercially), so something resembling hyperloop is possible in principle. Whether it is ready to be commercialized is a different question.
Recently, a company called Lux Research produced a research report called “Analyzing the Technical Barriers to Realizing the Hyperloop.” Their team analyzed global patent filings of hyperloop startup companies worldwide. Their findings identify two key technical challenges: to minimize leakage of air into the tube (which requires ongoing energy to run the pumps to maintain the near-vacuum) and to come up with a cost-effective tube and support structure design. The study reviews some of the trade-offs involved, on which the startups are presumably working.
The report also points out that none of the companies has actually demonstrated the kinds of high speeds they claim their systems will operate at—often stated as 600 mph, but sometimes as much as 750 mph. That’s because none of them have built and operated a tube anywhere near long enough to get a pod up to such speeds. Among their conclusions is: “A test track long enough to allow pods to reach maximum speeds and able to test tube pressures down to tens of Pa would allow for validation of the concept and refining the design of key subsystems.”
Because of the likely capital and operating costs of hyperloop, Lux Research estimates that it will operate at “a cost premium compared to high-speed rail.” And they opine that “The issue of high costs is the most likely reason hyperloop will fail.”
With this background in mind, I read a “2020 Hyperloop Feasibility Study” produced for the Mid-Ohio Regional Planning Commission. They contracted with AECOM to analyze corridor alternatives and the economic benefits of a hyperloop from Chicago to Pittsburgh (about 460 road miles). The corridor analysis required right of way that could handle 600 mph service, which ended up using some existing rail right of way (ROW) and some highway ROW. No cost estimates were provided for ROW acquisition—or for the infrastructure itself. That should have been a clue to what followed, the “economic benefits analysis.”
Using established U.S. Department of Transportation (DOT) benefit/cost analysis methodology, AECOM’s 38-year assessment (8 years construction followed by 30 years of operation) estimated $19.1 billion in benefits, using a 3% discount rate. Of course they also carried out a non-standard benefits analysis (assumptions and discount rate not stated) which claimed to find “$300 billion in wider economic benefits.”
Since no details are provided on the source of the $300 billion figure, I reviewed the explanation of the analysis conducted via DOT guidelines. Here are the categories:
|Travel time and operating cost savings||
|State of good repair||
So the first problem is that the numbers don’t total the claimed $19.1 billion.
Second, the travel time and operating cost savings “for” the commercial truck industry are actually the result of its calculated loss of mode share to freight carried on hyperloop instead of trucks ($336 million of the travel time and operating cost savings); I don’t think the trucking industry will consider those economic benefits.
The cargo diverted to hyperloop is stated as time-sensitive freight such as electronics, pharmaceuticals, and meat/seafood. The study’s graph shows such freight (carried in little passenger-size pods, not standard 40-foot shipping containers) as rising from 40 million tons in 2030 to over 80 million tons in 2059—but hedges this by stating it as “freight capacity,” not actual projected freight traffic.
The vast majority of the travel time/operating cost savings comes from trips diverted from driving and air travel, based on the assumption of much faster travel time and “the affordability of hyperloop.” That is directly contrary to the Lux Research estimate that hyperloop will likely operate at a cost premium to high-speed rail.
But the most telling problem with the estimate of the benefits is the $11.1 billion “state of good repair.” This is defined as “the residual value of the infrastructure” after 30 years, at a 3% discount rate. Using standard interest tables and assuming depreciation at 3% per year, my calculation shows the estimated initial cost of the infrastructure must be $26.9 billion. That cost is much greater than the alleged benefits, whether $18 or $19 billion.
From reading both studies, the clear message is: Hyperloop is not ready for prime time.
By the way, the Lux researchers agree. They think the earliest date for any hyperloop system to be ready to begin operations is 2040.
In May the California Air Resources Board unveiled its Advanced Clean Truck Rule, under which a growing percentage of new truck sales, starting in 2024, must be electric powered. The fractions ramp up over time, until in 2035, 55% of new light trucks, 75% of new heavy trucks, and 40% of tractor-trailers must be all-electric. This was music to the ears of electric-truck companies Tesla and Nikola, and seemed acceptable to the legacy truck producers, all of which have electric prototypes in operation. CARB estimates that by 2035 there will be 300,000 electric trucks on the road in California (about 15% of the total fleet).
The unanswered question for long-haul over-the-road electric trucks is: Where will they recharge (and how long will it take)?
Two recent studies have attempted to answer the first of those questions. The National Center for Sustainable Transportation at the University of California—Davis released a 52-page report in April suggesting that highway rest areas are an ideal location for vehicle charging. The study, which looked into charging for both personal vehicles and commercial trucks, focused on rest areas “because they are situated conveniently at locations alongside highways that facilitate long-distance or intercity travel.” Its modeling focused on personal EVs and sought to model the increased electricity demand based on projected EV penetration by 2050.
The other study, released in June, was more narrowly focused on trucks, and used the multi-state (CA, OR, WA) I-5 corridor. The “West Coast Clean Transit Corridor Initiative Study” was carried out on behalf of electric utilities in the three states. It mapped out locations at 50-mile intervals the length of north-south I-5, with 27 total charging sites. The study also included charging locations along I-8, I-10, I-80, I-210, and I-710 in California, I-84 in Oregon, and I-90 in Washington.
Since the second study focuses on trucks and almost entirely on Interstates, the potential policy collision is between the desire to provide convenient charging sites “along” major Interstate truck routes and the 1956 federal ban on any commercial services at “Interstate Rest Areas.” That ban is mentioned on page 45 of the UC Davis study but is misinterpreted as a mere regulation. Alas, it is embedded in the statute that created the federal program to fund the creation of the Interstate highway system. Hence, it can only be changed by Congress, not by FHWA.
Of course, if the electricity were to be given away, rather than being sold to motorists and truckers, that would not be a “commercial” transaction. But I cannot imagine that the nine utilities which funded the I-5 study plan to give away their electricity. Apart from a change in the law (which is long overdue), the only loophole I can think of is for state DOTs to allow companies to develop commercial service plazas not at existing “Interstate rest areas” but at other parcels directly adjacent to the Interstates in question. Whether that would pass muster is a question for legal researchers, not a mere engineer like me.
By Baruch Feigenbaum
Earlier this year the Union of Concerned Scientists (UCS) released a study detailing how ride-hailing is contributing to a climate crisis. But instead of focusing on climate and emissions, the study included a wide variety of concerns ranging from increased traffic congestion to a lack of regulation. Lacking a clear focus, the paper recommends a bunch of heavy-handed government policies that have proven largely ineffective in the past, instead of recommending a simple solution that is more likely to be effective.
The study provides a good overview of the recent history of ride-hailing, which has grown rapidly since 2010. By 2018, Uber had provided more than 10 billion trips while Lyft provided 1 billion. Ride-hailing trips outnumber taxi trips — in New York City the ratio is two ride-hailing trips for every taxi trip. In San Francisco, the ratio is 12 to one. The study claims that ride-hailing trips account for 2-to-13% of vehicle trips in downtown areas, although the 13% number sounds too high to me.
The UCS piece starts with the premise that travelers who switch to ride-hailing emit 69% more greenhouse gas emissions than they did when using their previous mode. This occurs for two reasons. First, a deadheading vehicle travels more miles (the miles without passengers) and is, therefore, more polluting than a trip in a personal car. The UCS study estimates that single-person ride-hailing generates 47% greater emissions than a private vehicle. Second, ride-hailing is increasing—not replacing—car trips. The study indicates that without ride-hailing, 24% of non-pooled riders would have instead taken transit, walked, and cycled.
However, some of the study’s recommendations won’t reduce emissions. One is for the ride-hailing industry to promote better connections to transit for environmental reasons. Yet the assumption that transit is always more environmentally-friendly is wrong. In 93 of the largest 100 metro areas, cars emit less carbon dioxide per passenger mile than transit buses. And since buses handle a far higher share of travelers than rail, walking and cycling, this difference is significant.
Even when factually correct, the emissions discussion does not consider tradeoffs. Reducing emissions is an important goal. But so is increasing access to economic activities. One advantage of the automobile is it significantly reduces the travel time needed to reach a job. Employees can reach far more jobs (and employers far more workers) by automobile than transit, walking, or cycling. In fact, transit trips take twice as long on average as automobile trips, yet are shorter in miles.
Most workers don’t routinely commute by ride-hailing vehicles. But those vehicles are a good option when those workers are running late or the transit vehicle breaks down, or it is raining (and they don’t want to walk or cycle in the rain).
Ride-hailing services have other economic and safety benefits. Many customers take them to nightclubs and restaurants, which may not be accessible by transit. The alternative is either customers driving home drunk, which is a major safety problem, or customers not going at all, which would create an economic hardship for the industry.
Another of the study’s concerns is traffic congestion. The study suggests that ride-hailing is increasing congestion, citing San Francisco travel speeds declining by 2 miles per hour and Manhattan travel speeds declining by 3 mph in a six-year period. Yet that could be a result of increased travel due to a growing economy between 2012 and 2019. If ride-hailing was the key factor, wouldn’t traffic speeds have declined more in San Francisco, which has a higher share of ride-hailing trips than New York City?
The UCS paper’s third concern is that the ride-hailing industry is a bad actor, simply because it is less regulated than taxis. Yet taxi regulations have failed to reduce emissions, improve safety, or lead to innovation. Those same taxi regulations protected taxis from competition for decades and allowed them for years to ignore customer-friendly features such as credit card readers. The report acknowledges that taxis have the same negative effects on emissions and traffic congestion. If the goal is reducing emissions, how are taxis better?
The most bizarre claim is that, while transportation choices are good, the addition of ride-hailing as a choice is bad. So, following that logic, subsidized choices provided by the government such as transit, cycling, and walking are good but unsubsidized choices, such as ride-hailing are bad. Why? Is it because ride-hailing increases emissions and congestion? Most transit vehicles also increase emissions and congestion. Who makes the arbitrary decision that one mode is better than another and based on what data or logic?
If the Union of Concerned Scientists wanted to develop a cohesive future vision, it could examine linking shared, electric vehicles with automated vehicles. Alain Kornhauser at Princeton University and Dan Sperling at UC Davis have studied these connections. While I am skeptical of some of the environmental claims of the shared-electric-automated future, the concept is a cohesive, logical vision. However, that would require endorsing both automation and ride-hailing services, neither of which this UCS study recommends.
If the UCS paper’s goal was to reduce emissions and encourage ride pooling or transit, it could argue for congestion pricing. Cities could vary road prices based on the level of congestion, encouraging customers to use other modes or to travel at off-peak times rather than rush hours. Cities could charge single-occupant vehicles more and electric vehicles less. Not only are drivers responsive to pricing but implementing this one policy is easier than implementing UCS’s 7 separate regulatory provisions. However, if the real goal is to spend taxpayer funds to control ride-hailing through government regulation (in the guise of environmental and congestion concerns), this laundry list of mostly ineffective policies would do the job.
By Marc Scribner
The Insurance Institute for Highway Safety (IIHS) recently published a study (May 2020) finding that automated vehicles (AVs) would prevent just 33.1% of crashes. Compared to the approximately 94% of crashes that involve driver error, this estimated safety benefit of AVs was surprisingly low and predictably led to headlines such as, “Self-Driving Cars May Not Be The Game Changer For Safety We Think” Fortunately, the IIHS study is not nearly as pessimistic as it appears when a key assumption is adjusted to better reflect reality, raising potentially preventable crashes by automated driving systems to around 74%. The IIHS study does an admirable job analyzing National Highway Traffic Safety Administration’s (NHTSA) National Motor Vehicle Crash Causation Survey (NMVCCS) data. It closely examines driver-error crashes contained in the NMVCCS database, assigns them to five categories, and estimates the weighted percent for each driver-related crash factor:
- Sensing/perceiving (i.e., not recognizing hazards), 22.6%;
- Predicting (i.e., misjudging behavior of other vehicles), 17.0%;
- Planning/deciding (i.e., poor decision-making regarding traffic law adherence and defensive driving), 41.3%;
- Execution/performance (i.e., inappropriate vehicle control), 23.2%; and
- Incapacitation (i.e., alcohol-impaired or otherwise incapacitated driver), 10.5%.
So far, so good. But the IIHS study then assumes AVs will only prevent “sensing/perceiving” and “incapacitation” crashes, or 33.1% of total crashes. However, the study is clear that this scenario applies only if AVs somehow allow their occupants to prioritize “rider preference” over safety.
In other words, the assumption here is that occupants would have the ability to direct the AV to violate traffic laws so as to speed, tailgate, drive aggressively, or make other expressly illegal maneuvers. As the study notes, “When rider preferences and safety conflict, however, AVs must be programmed to prioritize the latter.”
The good news is that automated driving systems are being programmed to follow the law and the study even highlights one developer’s explicit declaration that this will be the case. There will be no “rider preference” switch in the back of a self-driving taxi that allows passengers to violate traffic laws at their whim. Such a design would present significant—perhaps insurmountable—challenges to being permitted to enter service in the United States in the first place. And if it somehow managed to escape scrutiny from regulators, it would expose AV manufacturers to immense product liability and send them into the open arms of trial lawyers across the country.
Simply put, if a hypothetical irrational engineer employed by an AV developer designed such a rider preference selection device, corporate compliance, federal and state regulators, and the plaintiffs’ bar would quickly make sure such a dangerous technology never made it into the hands of the public.
Given that AVs remain under development, we cannot reliably estimate how much better they will handle “predicting” (17%) and “execution/performance” (23.2%) crashes relative to conventional manually driven vehicles. But we can safely assume that AVs will be programmed to follow traffic laws. Adding these “planning/deciding” crashes (41.3%) to the IIHS study’s stated 33.1% of AV-preventable crashes from “sensing/perceiving” and “incapacitation” errors raises the total crash reduction potential of AVs to 74.4% without changing the study’s general methodology.
Overall, the IIHS study is praise-worthy in its general analysis of driver-error crash factors and the potential for AV crash mitigation, but it deserves criticism for its inaccurate key assumption and how it presented and summarized its findings. Transportation and general news outlets that published articles based on the study’s misleading abstract and accompanying press release without reading the full nine-page study also deserve some criticism—If they had analyzed the full IIHS study, their headlines should have been closer to, “Insurance group finds self-driving cars could prevent the large majority of crashes.”
On July 9, the Federal Highway Administration announced new grants to five state DOTs for projects that will continue testing various ways of collecting per-mile charges that could eventually replace per-gallon fuel taxes.
Oregon, which is furthest along (by having an ongoing, voluntary program in which motorists are actually paying per-mile charges instead of fuel taxes) received a grant to help fund a project to assess the feasibility of using telematics data from a connected-vehicle system to collect per-mile charges. It won a second grant on behalf of an 11-state consortium to investigate per-mile charges and blockchain.
Utah DOT received two grants dealing with ways of integrating per-mile charges with express lane tolls. Utah is the only state besides Oregon that has collected real (as opposed to simulated) money in its mileage-charge pilot projects.
Washington won the largest grant for the next phase of its seven-year road usage charge program. The new project will focus specifically on hybrid and electric vehicles, plus the state’s own vehicle fleet. In a January 2020 report on its ongoing program, the Washington State Transportation Commission estimated that its transition from per-gallon taxes to per-mile charges would take place over a 10 to 25-year period.
The I-95 Corridor Coalition (which has just changed its name to The Eastern Transportation Coalition) received a new grant, via Delaware DOT, to address barriers to per-mile charges identified in its several previous pilot projects. The Coalition members include 17 states and the District of Columbia.
Finally, Wyoming DOT received a grant for a truck mileage-charge project. Trucks have been involved in several of the previous pilot projects, including California and the I-95 Corridor Coalition.
TRB/AUVSI Automated Vehicles Conference, Regulatory Breakout Session, Online, July 27-30. (Baruch Feigenbaum speaking).
Domestic and International Aviation Recovery, TRF-DC, WTS-DC, and YPT-DC, hosted by Reason Foundation, Online August 5, 2020.
Michigan Approves Interstate Tolling Study. Michigan Gov. Gretchen Whitmer recently signed SB 517, sponsored by Sen. John Bizon. It calls for a two-part consultant study of (1) the feasibility of using tolls to pay for modernization and maintenance of major highways in Michigan and (2) a strategic implementation plan, laying out which corridors are good candidates and the possible timing of the phase-in of tolling. The study would be comparable to the pair of studies carried out recently in Indiana. My colleague Baruch Feigenbaum has provided some context on the implications of the planned studies.
State Gas Tax Diversions Identified in New Study. About half the states are diverting portions of their “highway user tax” revenue to non-highway purposes, according to a new study by the Reason Foundation. The largest diverters of state gas tax money to non-road uses are New York (37.5%), Rhode Island (37.1%), New Jersey (33.9%), Michigan (33.9%), and Maryland (32.5%). The diversion of gas tax money away from highways may help explain why some of these same states receive among the lowest scores in Reason’s Annual Highway Report rankings of state highway systems by value and cost-effectiveness (New Jersey ranks last, Rhode Island at 48th, and New York at 45th.) Two of the country’s largest states divert gas tax revenue—Texas diverts 24% of fuel taxes but managed to rank 23rd in the last highway conditions report, and Florida diverts 13.6% of state gas taxes and was 40th in the last Annual Highway Report rankings
How Florida Could Transition from Per-Gallon to Per-Mile. The James Madison Institute, based in Tallahassee, released my policy brief explaining the upcoming decline in state gas-tax revenue and the need to shift, over time, to charging (electronically) per mile. The brief suggests beginning the transition with the state’s limited-access highways (freeways and Interstates), taking advantage of the statewide SunPass electronic tolling system.
Florida and Georgia Join E-ZPass. The Florida Turnpike Enterprise and Georgia’s State Road & Tollway Authority have announced their joining of the multi-state E-ZPass interoperable electronic tolling system. Florida’s statewide SunPass and Georgia’s Peach Pass transponders will now be usable in any of the 18 states in the E-ZPass system, ranging from Maine to Illinois. In June, the Central Florida Expressway Authority (which was already interoperable with E-ZPass) announced that it will offer its customers a new transponder that is usable nationwide, called Uni. It is made by TransCore and is similar to the nationally interoperable transponder offered by Bestpass to trucking fleets nationwide.
HOV-5 Proposed for Los Angeles Express Toll Lanes. As part of its planning for the 2028 Olympics, LA Metro has proposed that its (by then larger) network of express toll lanes require five or more occupants for free passage. Buses, vanpools, and five-member carpools will pay nothing, while carpools with two to four occupants will pay discounted tolls; single-occupant vehicles will pay the full amount of the variable toll. One of the planned express lane corridors is through the highly congested I-405 in the Sepulveda Pass between the LA basin and the San Fernando Valley.
Virgin Trains Making Progress on Victorville to Las Vegas Line. Having acquired the rights to the defunct XpressWest project, Virgin Trains has made progress in financing the $4.9 billion project. It has been approved by U.S. DOT to issue $1 billion in tax-exempt Private Activity Bonds and has received California’s approval for $600 million in tax-exempt bonds, with another $200 million pending in Nevada. Unlike its Miami to Orlando route in Florida, which is diesel-powered, the Victorville-Vegas line will be all-electric. On June 30, Caltrans announced an agreement for Virgin Trains to use the right of way in the median of I-15. The company has also signed a memorandum of understanding with San Bernardino County to study extending the line from Victorville to Rancho Cucamonga, where it could connect passengers to a Metrolink commuter rail station.
Miami-Dade Expressway Wins a Round. In its battle to overturn a 2019 law that called for its elimination, the Miami-Dade Expressway Authority filed suit to have the law thrown out as a violation of the County’s home rule authority. A lower court last year agreed with MDX’s position, but the state appealed. On June 25, Florida’s First District Court of Appeal denied the Legislature a writ that would dismiss MDX’s lawsuit. So the agency’s battle against the Legislature and (amazingly) Florida DOT will continue.
Toll Bypass Tunnel Opens in Australia. The new 9 kilometer M8 twin tunnel project in Sydney opened to traffic early this month. It is part of the A$16.8 billion WestConnex road network which has been under construction for a number of years. The tunnel parallels the congested M5 East expressway (built in 1992) and is expected to cut 30 minutes off the peak-period trip between downtown Sydney and the southwestern suburbs. The variable toll for the new tunnel will be up to A$6.95 for personal vehicles. Both tunnels have the capacity to add a third lane in the future.
Interchange Bottlenecks Being Tackled in Florida and Georgia. Reconstruction of three major interchanges promises future relief from congestion in Atlanta, Orlando, and Tampa. In Atlanta, a $400 million project has been approved by Georgia DOT for the I-285/I-20 interchange, which is the 24th most-congested truck bottleneck on the American Transportation Research Institute’s list of the top 100 truck bottlenecks nationwide. (Three other I-285 interchanges rank #2, #5, and #7 on that list.) In Orlando a major personal-vehicle bottleneck is the I-4/SR 408 interchange, which has been redesigned and is being rebuilt at $960 million as part of the $2.4 billion I-4 Ultimate project, to be finished next year. And in Tampa, Florida DOT has approved the $1.4 billion reconstruction of the congested I-275/SR 60 interchange; still to be planned is Tampa’s major truck bottleneck, I-275/I-4 (#91 on the ATRI list).
More Conversions to All-Electronic Tolling in Prospect. Many toll roads eliminated cash tolling during the lockdown phase of the COVID-19 pandemic but considered this a temporary measure. But in early June, the Pennsylvania Turnpike announced plans to lay off its 500 toll collectors, to make all-electronic tolling permanent, as a growing number of other toll roads did prior to the pandemic. Ten days later, some state legislators asked Pennsylvania Gov. Tom Wolf to overturn the Turnpike’s decision, so the outcome remains in doubt. On the West Coast, the San Francisco Chronicle reported that permanent all-electronic tolling is under serious consideration for the seven state-owned toll bridges in the San Francisco Bay Area, but is unlikely to be implemented until lower-cost billing procedures are worked out for the FasTrak tolling system.
Two New Express Toll Lane Projects Open in Denver. Last month saw the addition of two new express lanes in the Denver metro area. One project adds ETLs to the stretch of I-25, between 120th Ave. and E-470. The other ETLs are on C-470, across 12.5 miles of the southern portion of the Denver ring road. The latter ETLs will begin variable tolling after an initial period for motorists to get used to driving on them. Consistent with the existing I-25 express lanes policy, only HOV-3 vehicles (and buses and motorcycles) can use the I-25 lanes without charge. But on C-470 (which has been financed based on toll revenues), all personal vehicles will be required to pay.
New Paper Quantifies Value of Urgency. In a working paper released by the National Bureau of Economic Research in April, Antonio Bento, Kevin Roth, and Andrew Waxman used detailed transaction data from the I-10 express toll lanes in Los Angeles to study the prices paid by motorists choosing to enter and leave the lanes during peak periods. They defined a decision variable “value of urgency” to explain those decisions, and estimated that VOU accounted for 87% of the average toll paid by such users. The paper is NBER Working Paper 26956.
“[Managed Lanes] in congested areas can be extremely effective at delivering increased capacity, superior driving time and reliability, and revenue generation. However, these benefits come with a price that is inextricably linked to the free market. The downside is that toll rates may be very high, and the upside is that the users have an option. Operators that are unwilling or unable to allow the market to dictate ML pricing should be prepared to forfeit the benefits that MLs were designed to deliver.” —Scot Monroe, “Managed Lanes: A Framework for Prudent Pricing,” Fitch Ratings, October 1, 2018
“The toll technology revolution that began several decades ago with the introduction of electronic toll collection has positioned the tolling industry to be at the forefront of the impending revolution on how infrastructure is funded into the future on a much broader scale, as a means to deliver critical infrastructure projects, and even continue to expand its use as a means of congestion management, but I do believe the industry is standing at a precipice. There is tremendous opportunity right now to be the leader in the discussion on finding a sustainable solution to fund ongoing infrastructure needs, both at the federal and state levels. But the industry must be open to change, to expanding its vision, to partnering with new industries and emerging technology leaders. If not, someone else will ultimately fill the void, and today’s industry will be relegated to the sidelines. What’s the saying? Lead, follow, or get out of the way.” —J.J. Eden (North Carolina Turnpike), “Tolling Can Lead Transportation Into the Future,” Traffic Technology Today, January 22, 2020