High-Performance Highways
Pricing may be one solution to managing congestion on the transportation network, with implications for financing tomorrow's infrastructure.
Growing congestion on the transportation network poses a substantial challenge for the U.S. economy and for the quality of life of millions of Americans. Highway congestion in metropolitan areas has increased dramatically over the past two decades. According to the Texas Transportation Institute's The 2005 Urban Mobility Report, U.S. motorists spent 0.7 billion hours sitting in traffic in 1982; the number had jumped to 3.7 billion hours by 2003. During that period, the costs associated with congestion skyrocketed from $12.5 billion (in 2003 dollars) to $63.1 billion. Further, according to 2003 data, congestion wastes nearly 2.3 billion gallons of fuel, as motorists and truckers sit idling on congested roadways.
Congestion pricing — sometimes called value pricing — is one tool that may help relieve congestion in the short term and reduce the wastes associated with idling traffic. Congestion pricing charges motorists to use highways during peak periods. As a result, it encourages motorists to shift some rush hour travel to other transportation modes or to offpeak periods. Although commercial vehicles have less discretion to travel at different times of the day, the benefits from congestion pricing in terms of the delay and fuel cost savings far outweigh the extra monetary costs incurred for congestion charges.
By removing at least some of the noncommuting drivers from congested roadways during the rush hours, congestion pricing enables traffic to flow more efficiently. For example, daily traffic data for vehicles inbound in the morning on I-66 outside the Capital Beltway in the Washington, DC, metropolitan area show that vehicle flow drops from about 8,000 vehicles in the 1-hour period from 6 a.m. to 7 a.m. to less than 6,000 vehicles in the hour from 8 a.m. to 9 a.m.
Of particular interest to congestion experts at the Federal Highway Administration (FHWA) is a concept known as high-performance highways, which involves applying variable tolls on all lanes of existing tollways and toll-free facilities to manage traffic flow. Tolls would vary by level of demand, either on a fixed schedule by time of day (such as in half-hour blocks) or in real time (such as every few minutes, if necessary) and would be charged only on congested highway segments to manage traffic flow. Previously, the chief method for high-performance freeway operations was through ramp management. To provide travel alternatives, the high-performance highway concept also would involve promoting carpools and vanpools, offering park-and-ride facilities, and providing fast, frequent, and inexpensive express bus service on the tolled roadways. All vehicles, except authorized buses, would be charged a variable toll set high enough to guarantee that high demand will not cause traffic flow to break down.
Numerous economic, financial, operational, political, social, and technological considerations constrain how highway managers can price road use. And, ultimately, each State or locality will need to choose the pricing configurations deemed optimal to deal with these many considerations. "The high-performance highway is not the only, nor necessarily the best, approach to pricing roadways," says Regina McElroy, director of the Office of Transportation Management at FHWA. "However, it attempts to deal with multiple considerations on a comprehensive basis, illustrating ways to accommodate high levels of system performance and economic efficiency within a broad framework of public acceptance and political reality."
The High-Performance Concept
The high-performance highway approach is rooted in the fact that when traffic volume exceeds a certain threshold level, both vehicle speed and throughput drop precipitously. Data from the Washington State Department of Transportation's 2006 report Measuring Delay and Congestion: Annual Update show that maximum vehicle throughput on freeways in the State of Washington occurs at speeds of about 67 kilometers per hour, km/h (42 miles per hour, mi/h) to 82 km/h (51 mi/h).
"The number of vehicles that get through per hour frequently drops by as much as 35 percent when severe congestion sets in and speeds drop to crawl speeds of 20 to 30 mi/h [32 to 48 km/h]," says Pravin Varaiya, an expert in transportation operations research, optimization, and performance management. "Traffic flow is kept in this condition of collapse for several hours after the rush of commuters has stopped. This causes further unnecessary delay for offpeak motorists who travel after rush hours."
According to the high-performance highway concept, transportation operators charge variable tolls on tunnels, bridges, and existing toll roads and toll-free highways on all lanes but only on critical congested segments, not on the entire system. The variable toll dissuades some motorists from using limited-access highways (generally freeways) at critical bottleneck locations where traffic demand is high and where surges in demand could push the freeway over the threshold at which traffic flow collapses. Applying pricing to entire segments of roadways during congested periods in effect turns all lanes into premium service lanes. Not only are more motorists able to reach their destinations during the rush hours, but also they get there faster.
In Orange County, CA, for example, toll lanes built in the median of State Route 91 (S.R. 91) carry a much higher volume of vehicles per lane than the adjacent general purpose lanes during the hours with heaviest traffic on the eastbound direction. "With the use of congestion management pricing, travel on the 91 Express Lanes during peak period hours captures 40 percent of the traffic volumes with only 33 percent of the capacity," says Kirk Avila, general manager of the 91 Express Lanes for the Orange County Transportation Authority. Based on the success of tolling efforts such as the 91 Express Lanes, FHWA officials extrapolate that if all lanes were tolled, then the throughput would be greater across the entire network.
Implementing the Concept
On a high-performance highway, peak period commuters would have several traveling options:
- Pay a relatively low toll for the convenience of driving alone in free-flowing traffic on the high-performance highway.
- Join a carpool or vanpool and enjoy a fast trip on the high-performance highway for an even lower price by sharing the cost of the toll. (Free or discounted service for carpools or vanpools could be a policy option but would increase enforcement challenges.)
- Use newly expanded and more convenient services provided by express buses that run on the high-performance highway.
- Drive alone for free on the arterial street system.
Congestion pricing strategies involve "open road" tolling, which means collecting fares without using traditional stop-and-pay toll booths. All tolls are collected electronically at highway speeds, and transportation operators can choose from a variety of mechanisms to collect the tolls. Licensed drivers in the area covered by the priced network could be issued inexpensive electronic transponders (such as a sticker tag, which is a thin decal that is similar to a conventional battery-operated tag but requires less maintenance) free of charge, along with a transportation account. As with telephone service, the account holder receives a monthly bill. The bill may be paid automatically if the account is linked to a credit card provided by the account holder. Nonresidents could purchase the tags at retail outlets, such as convenience stores or gas stations, or from automated teller machines at welcome centers located at approaches to metropolitan areas.
Motorists without transponders could be video-tolled, meaning that cameras would take pictures of their license plates, and the vehicle owners would be charged for the toll plus a small administrative fee to cover the extra costs. On November 1, 2006, Florida's Turnpike Enterprise, in conjunction with the Tampa-Hillsborough Expressway Authority, launched a pay-by-plate system, the first video-toll account system in the United States. Customers who are occasional users of the Lee Roy Selmon Crosstown Expressway (between Tampa and Brandon, FL) and do not have a transponder can call a toll-free number to open an account.
"They pay a toll of $1.25," says Jim Ely, executive director of the Florida Turnpike Enterprise, "instead of the $1.00 paid by those with transponders, which is only a small premium charge in order to cover costs to process the license plate images."
To provide predictability of costs for motorists, highway managers could set the toll rates to vary by time of day on a fixed schedule. The schedule could be adjusted periodically, if necessary, to address changing demand. The operators of California's 91 Express Lanes adjust the toll rates on a quarterly basis.
To ensure that day-to-day variations in highway demand do not cause breakdowns in traffic flow, highway managers may need to manage freeway use in real time. Airlines offer free flight tickets or cash to induce some ticket holders to give up their seats when flights are overbooked. Highway managers can apply this concept to the freeway system by creating optional-use lanes with ramp meters at entrance ramps. If sufficient pavement does not already exist at the ramps, some minor pavement widening might be needed. As motorists approach an entrance ramp, they would choose between two lanes. Those who are in a hurry or choose to pay the posted toll simply enter the mainline traffic. Those who have more flexibility in their schedule could opt to wait in an optional-use lane. Traffic signals would control entry of traffic into the mainline freeway from the optional-use lanes. Those who choose to wait in those lanes would be compensated for their delays by credits to their transportation accounts. Highway managers could post the amount of credit offered and estimated delay times on variable message signs at the entrance ramps, so motorists arriving at the ramps could decide for themselves whether the compensation offered would be acceptable to them or whether they prefer to enter the freeway without delay.
The Transit Advantage
Another key component of the high-performance highway concept is the provision of travel alternatives, such as park-and-ride facilities and vanpool and transit services. It is important that premium transit services be available a few months before pricing becomes operational. Where free-flowing high-occupancy vehicle (HOV) lanes do not exist on the congested network, agencies could build or designate highway shoulders as bus lanes for use by transit and authorized vanpools and paratransit vehicles during congested periods. The lanes could operate in each direction on all freeways, in conjunction with the introduction of new transit and vanpool services, and could be discontinued after all freeway lanes are restored to free flow with pricing. Ensuring travel time advantages relative to driving alone is critical to the success of new transit and vanpool services.
The Minnesota Department of Transportation (Mn/DOT) has a 14-year history of designating bus shoulders on highways in the Minneapolis-St. Paul metropolitan area. Mn/DOT requires that the pavements for bus shoulders be at least 18 centimeters (7 inches) thick and at least 3 meters (10 feet) wide. "Where pavements are strong enough and wide enough, highway agencies could designate a bus shoulder using signage," says Jennifer Conover, the Team Transit project manager who oversees Mn/DOT's bus shoulder program. "Buses using the shoulder operate at slow speeds — 35 mi/h [56 km/h] or 15 mi/h [24 km/h] faster than the adjacent congested traffic. Restricting use of the bus shoulders to authorized vehicles with trained drivers traveling at slow speeds ensures that safety is not compromised while creating a transit advantage."
Benefits of High-Performance Highways
Because all lanes are priced, high-performance highways avoid the need for additional rights-of-way and pavement to construct barriers or buffer separations between priced lanes and toll-free, general purpose lanes. All lanes therefore become premium service lanes, accessible from existing interchanges and available for use by all vehicles. Without barriers separating the lanes, high-performance highways avoid the need for traffic to merge into and out of priced lanes from adjacent general purpose lanes. Such weaving movements are inconvenient for buses and motorists and reduce safety and highway capacity.
Further, because motorists are free to switch lanes, high-performance highways help maximize capacity. A slower moving vehicle in a separated single lane causes a gap to build up in front of it, reducing vehicle throughput per hour. In addition, when fewer adjacent lanes are available for use by all traffic, vehicle throughput per lane per hour is reduced, because faster drivers find it more difficult to switch lanes and overtake slower vehicles to occupy large gaps between vehicles.
Because new rights-of-way and expansion of highway infrastructure are not prerequisites for implementation, agencies could put an entire high-performance highway network in place in a metropolitan area in a relatively short time. The time-consuming environmental review processes generally associated with highway widening projects would be unnecessary and therefore would not delay implementation. Some new investment would be necessary for managing and operating the highway and arterial networks, initiating new express bus and vanpool services, and building new park-and-ride facilities. Those steps, however, would not require the extent of environmental review normally necessary for widening projects.
With high-performance highways, much more premium service capacity is available on multiple lanes. Therefore, relatively lower toll rates are sufficient to ensure that traffic volumes do not rise above available capacity, making the use of high-performance highways more affordable to a larger population of middle- and lower-income motorists.
Most important, with a high-performance highway, all lanes should remain congestion-free.
Addressing Public Acceptance
The notion of introducing high-performance highways may raise concerns among the driving public. Perhaps chief among them is that the scheme will not work, and motorists will pay more just to sit in the same traffic jams. To alleviate concerns about whether this approach would actually reduce congestion and benefit motorists, transportation officials could propose the pricing scheme as a short-term experiment, with permanent implementation subject to citizen approval by referendum. Stockholm, Sweden, offers an example, where a highly publicized trial run of congestion pricing lasted for 7 months from January through July 2006, with a public referendum on its continuation held that September. The measure passed with 53.1 percent voting in favor and 45.5 percent against.
To further alleviate doubts about the effectiveness of high-performance highways, the transportation operator could offer a money-back guarantee to toll-paying motorists who do not receive the promised levels of service (that is, travel speeds). If a crash or road construction slows travel, the delayed motorists would not be charged, or charges would be reduced to reflect the reduced level of service.
Another concern is the argument that the public already paid for the highways through taxes, so new tolls would amount to double taxation. One means to address this issue is to invite motorists to share in the net proceeds from the pricing scheme. Net toll revenues (after paying for maintenance and operation of the high-performance system) could be returned to motorists on a monthly basis in the form of credits to their transportation accounts.
Each motorist's share of the net revenue could be determined based on net revenue divided by the number of eligible motorists in the area. Registered drivers who have transportation accounts but rarely use heavily traveled roads — perhaps they take public transit or carpool — would be rewarded, while those who choose to use scarce and valuable highway capacity would have to pay more. This would change the current system used to pay for roads from something akin to an "all-you-can-eat" buffet system, under which everyone pays a fixed charge and can consume as much as he or she likes, to an a la carte system, where people make efficient consumption choices based on the price they are willing or can afford to pay. The credits could be used to pay for tolls, transit fares, or parking charges at park-and-ride facilities. Transportation agencies could issue credits on a rotating basis over the course of each month, thereby avoiding situations in which too many motorists attempt to use their free credits for tolls on the same day, thus swamping the network with extra traffic. At the motorist's option, unused credits could be redeemed at the end of the year toward payments for annual vehicle registration fees and taxes.
As an alternative to dispensing credits on a monthly basis, highway managers might opt to distribute net toll revenues annually to owners of registered vehicles in the metropolitan area, through rebates on vehicle taxes and registration fees generally charged annually.
In cases where toll rates are high enough to produce revenues that exceed the cost of providing additional transportation capacity, additional capacity may be necessary. In such cases, highway managers could use the surplus revenue to build new capacity so motorists who pay the high charges can see that their tolls are being used to address mobility issues.
A third possible concern regarding high-performance highways is the idea that low-income motorists, who have less flexibility in their work hours than many white-collar, higher income workers have, will be unable to afford new congestion tolls. Some might argue that it is unfair for the government to provide improved mobility only for the wealthy. However, the high-performance highway concept includes travel alternatives in the form of park-and-ride facilities and express buses, vanpools, and carpools. These alternatives provide an efficient and more cost-effective means of travel during rush hour for low-income commuters and others who may not have the flexibility to travel at less congested times or who simply do not want to pay the full cost of tolls.
For the high-performance highway to be successful, travel alternatives such as vanpools and new express bus systems must be in place before introducing congestion tolls, and routes and schedules must be beneficial to customers. Highway managers should offer fare-free transit promotions and vanpool trial periods to encourage use of these new services. These promotions would enable the public to become familiar with the new modal options and minimize public concerns about the viability of travel alternatives.
In addition, low-income commuters could receive discounts on tolls, transit fares, or parking charges at park-and-ride lots. Drivers could access the parking lots by using smart cards linked to their transportation accounts.
Finally, the public may be concerned that adjacent free roads will become more congested because motorists will divert from priced highways to the free routes to avoid the new congestion tolls. Experience has shown that when highway managers raise toll rates on existing tollways, some drivers divert to toll-free arterials or surface streets to avoid paying the higher rates. However, unlike conventional tollways, priced highways provide more travel options.
Several key differences reduce the likelihood of traffic diversion on parallel toll-free facilities:
First, variable tolls with a stepped rate schedule provide options for motorists to reduce or eliminate their costs for new tolls by shifting their time of travel. In the case of tollways with flat tolls all day, drivers cannot escape tolls or avail themselves of a lower toll rate simply by traveling at a different time.
Second, the introduction of variable tolls during congested periods would be accompanied by an expansion of transit capacity and the availability of carpool and vanpool options, so some solo drivers may shift to using transit or carpools rather than diverting to parallel toll-free roadways.
Third, when agencies introduce pricing on highways that used to be severely congested, some motorists who previously switched to parallel arterials may shift back to the free-flowing priced highways. And faster moving traffic on priced highways could result in an increase in vehicle throughput per hour, thus accommodating higher rush hour traffic volumes.
As long as parallel arterials remain toll-free, however, motorists who divert from the priced highways or shift from other less convenient times of travel may take the place of any traffic that shifts from the arterials to the priced highways. Thus, although total vehicle throughput per hour in the corridor may increase, arterial congestion is unlikely to improve during key congested periods. However, the duration of congestion (that is, the length of the congested period) may decrease.
On the other hand, highway managers could use toll revenues to pay for optimizing traffic signal controls on parallel arterials wherever they are not currently optimized. This approach could help improve traffic flow on the arterials.
Gaining Public Acceptance for a Broad Pricing Approach
|
Costs, Benefits, and Revenues
Researchers at FHWA recently used a modeling program called the Tool for Rush Hour User Charge Evaluation (TRUCE) to estimate the potential costs, benefits, and revenues of operating a high- performance highway. The planning tool is available on FHWA's Web site at www.ops.fhwa.dot.gov/tolling_pricing/value_pricing /tools/index.htm.
The researchers assessed three scenarios representing typical congestion levels on highway networks in major U.S metropolitan areas. They based the scenarios on a prototypical area (either a metropolitan area or a significant portion of a major metropolitan area) with approximately one million drivers and an existing 161-kilometer (100-mile) highway network comprising a total of 966 lane kilometers (600 lane miles). The study highways averaged six lanes (three lanes in each direction). The congestion-level scenarios were as follows:
A moderately congested freeway network, as defined in the study, would have average peak period speeds of 69 km/h (43 mi/h) and a total of 4 hours of congestion per day, that is, about 2 hours in the morning and 2 hours in the evening. The average speed represents a composite of higher traffic speeds on some segments of the network and much slower congested speeds on other segments. Assuming a free-flow freeway speed of 97 km/h (60 mi/h), this scenario represents a peak period travel time index (that is, ratio of average peak period travel time to free-flow travel time) of 1.4. Portland, OR, is a typical example of a metropolitan area with these characteristics.
A severely congested freeway network, would have average peak period speeds of 64 km/h (40 mi/h) and a total of 5.5 hours of congestion per day, or about 2.5 hours in the morning and 3 hours in the evening. This scenario represents a peak period travel time index of 1.5. The Washington, DC, metropolitan area is an example of a severely congested network.
An extremely congested freeway network, would feature average peak period speeds of 55 km/h (34 mi/h) and a total of 7 hours of congestion per day, or about 3 hours in the morning and 4 hours in the evening. This scenario represents a peak period travel time index of 1.75. Portions of highways in the Los Angeles, CA, metropolitan area qualify as extremely congested networks.
Using the TRUCE software, the FHWA researchers estimated the toll revenues, benefits, and costs associated with a multimodal pricing package that includes transit and park-and-ride services. They determined that benefit/cost ratios would range from 1.4 to 2.8, depending on the severity of existing levels of congestion, indicating that a multimodal pricing package would be economically efficient. "Because the TRUCE model uses conservative assumptions to project benefits, these estimates are on the low end," says Jack Wells, chief economist at the U.S. Department of Transportation (USDOT). "The model does not account for environmental and safety benefits, benefits to businesses and the economy, and increases in energy security."
Surplus revenue would be higher in more severely congested areas because of higher toll rates and longer congested periods during which tolls would be charged. Based on FHWA calculations, annual toll revenue surpluses could range from $40 million to $226 million. This surplus could provide an annual tax refund to area motorists of $40 to $226 per driver, or credits to their transportation accounts of $3 to $18 per month.
Initial Congestion Level |
|||
---|---|---|---|
Annual Benefits (Million $) |
Moderate |
Severe |
Extreme |
Highway Benefits |
$105.80 |
$184.51 |
$360.49 |
Transit Benefits |
$5.18 |
$9.02 |
$17.55 |
Multimodal Benefits |
$110.99 |
$193.53 |
$378.04 |
Annual Costs to Transportation Network(Million $) |
|||
Highway Costs |
$25.46 |
$33.34 |
$41.21 |
Transit Costs |
$52.50 |
$72.19 |
$91.88 |
Multimodal Costs |
$77.96 |
$105.53 |
$133.09 |
Multimodal Benefit/ Cost Ratio |
1.4 |
1.8 |
2.8 |
Annual Toll Revenues Versus Costs (Million $) |
|||
Toll Revenues |
$118.19 |
$196.35 |
$358.85 |
Multimodal Costs |
$77.96 |
$105.53 |
$133.09 |
Surplus |
$40.22 |
$90.82 |
$225.76 |
Urban Partnerships
High-performance highway networks represent one approach to curbing the growing burden of traffic congestion. Converting all lanes on existing freeways into premium service, variably tolled, free-flowing highways during peak periods only, coupled with fast, frequent, and inexpensive express bus service, could provide social benefits that far exceed multimodal investments and operating costs. When implemented in large metropolitan areas or portions thereof, possibly by public-private partnerships, these networks could generate toll revenues that would be sufficient to cover all management and operational costs, including the funding necessary to introduce new express bus and park-and-ride services.
The nation-city of Singapore is the only entity known to have implemented the high-performance highway concept on a network-wide basis. Singapore introduced peak period pricing in the downtown area during the morning rush hours in 1975. In spring 1998, the city shifted to a fully automated electronic charging system, with in-vehicle electronic devices allowing payment by smart card, and enforcement using cameras and license plate reading equipment. Singapore also introduced variable electronic charges on the expressway system, with charges set by time of day to ensure the free flow of traffic. The system is the first of its kind in the world and has reduced congestion in that city significantly. During the peak periods, more than 95 percent of expressways and major arterial roads are smooth-flowing and congestion-free, according to Singapore's Land Transport Authority.
In May 2006, USDOT announced its National Strategy to Reduce Congestion on America's Transportation Network (the Congestion Initiative), a comprehensive national program to reduce congestion on the Nation's roads, rails, runways, and waterways. One major component of the initiative is the Urban Partnership Agreement. USDOT is soliciting proposals by metropolitan areas to enter into partnership agreements with USDOT to demonstrate strategies with a track record of effectiveness in reducing traffic congestion. To support broader adoption of successful strategies, USDOT officials expect to use discretionary funding available under the Department's Intelligent Transportation Systems Operational Testing to Mitigate Congestion Program, its Value Pricing Pilot (VPP) program, and other grants. In addition, to the extent possible, USDOT will support its urban partners with regulatory flexibility and dedicated expertise and personnel.
FHWA is actively encouraging States and metropolitan areas to apply for grants under the urban partnership program to test and demonstrate congestion pricing concepts such as the high-performance highway. Metropolitan areas with severe congestion may submit applications for consideration for funding.
Patrick DeCorla-Souza, AICP, is program manager for urban partnerships at FHWA in Washington, DC. He has provided direction to FHWA's VPP program since 1999 and works with public and private sector partners nationwide to implement innovative road pricing strategies. DeCorla-Souza cochairs the Transportation Research Board's Congestion Pricing Committee and has master's degrees in transportation planning and civil engineering.
For more information on Urban Partnership Agreements, please visit www.fightgridlocknow.gov or contact Thomas McNamara at 202-366-4462, thomas.mcnamara@dot.gov or Patrick DeCorla-Souza at 202-366-4076, patrick.decorla-souza@dot.gov.
This article is the fifth in a Public Roads series on innovative financing. One of FHWA's priorities is encouraging the use of innovative financing.