Anyone who has driven an urban highway during rush hour knows the frustration of an endless line of back-to-back bumpers. Congestion pricing offers economic incentives to improve traffic flow—but agencies need to know how congestion pricing impacts commuting patterns, based on where drivers live and how much money they make.

In 2007 and 2008, the U.S. Department of Transportation (U.S. DOT) awarded more than $1 billion to help six urban areas pursue aggressive congestion reduction programs. The Federal Highway Administration (FHWA) in turn tapped Volpe to conduct before-and-after household travel surveys in two of those cities, Atlanta and Seattle, to capture a dynamic picture of how congestion pricing shapes household travel choices.

The Challenge

While congestion pricing strategies are becoming increasingly common in major metropolitan areas across the U.S., there is still little data on how road pricing affects traveler behavior.

How do households respond when congestion pricing is applied on their roadways? Do they pay the toll? Or, do they seek alternative routes or travel modes to bypass the higher prices? And, does congestion pricing affect the number of miles that households drive? Previous research also has not addressed socioeconomic impacts of congestion pricing. Are lower-income households disproportionately affected by tolling? Do lower-income households pay a disproportionate share of tolling?

The Solution

Volpe recruited travelers for this study using license plate recognition technology during peak hours on the 520 and I-90 bridges in Seattle and along a 16-mile corridor of Interstate 85 (I-85) and Buford Highway in Atlanta. Researchers invited all adult members of each sampled household to participate in the panel survey. Volpe researchers recruited a pool of participants representative of the population that uses these roads during the most congested hours.

The final sample included more than 3,500 adults in Seattle and more than 3,000 adults in Atlanta. Participants completed surveys and travel diaries six months before and after congestion pricing was put in place. They were asked questions about their travel habits, attitudes, values, and household characteristics. The methodology was based on well-established survey techniques for following a sample over time, and the questionnaires were built on Volpe’s decades of travel survey experience, giving researchers and officials high-quality data and narratives to refine future congestion pricing systems.

The Impact

The following selected findings from Volpe’s work provide a comprehensive understanding of the real-world impacts of congestion pricing that inform FHWA and urban areas that are considering congestion pricing.

Seattle

The Washington State Legislature levied a congestion pricing toll on the 520 bridge between Bellevue and Seattle across Lake Washington to help pay for a replacement bridge. Tolls are higher during peak hours, and tolls are collected by transponder or by mail.

Surveys showed the following results:

• Recorded trip segments fell 43 percent on the 520 bridge.
• Travel reductions tended to fall into discretionary categories, such as shopping and dining.
• Nearly 25 percent of regular 520 bridge travelers switched to Interstate 90 (I-90), a parallel, free roadway four miles south of the 520 bridge. These travelers were primarily motivated by avoiding the toll.
• Telecommuting was unaffected by the tolling program.
• Eight percent of 520 bridge travelers fell into the lowest-income group measured (less than 300 percent of the federal poverty level) before the toll. After the toll, this group substantially reduced their cross-lake travel—particularly for discretionary trips—and were more likely to have switched to the toll-free I-90 route.
• Public transit users were generally neutral toward the congestion pricing plan.
• Respondents reduced their overall vehicle-miles traveled by 15 percent.

Atlanta

Transportation planners in Atlanta turned an existing high-occupancy vehicle (HOV)  lane along a 16-mile stretch of I-85 into a high occupancy toll lane (express lane), and increased the occupancy requirement from 2+ to 3+. Before tolling, two or more people could use HOV lane for free, whereas after tolling the express lane was free for cars with three or more people and tolled for cars with fewer than three people. The express lane toll goes up as congestion increases, and all users are required to have a transponder. All other lanes are not tolled.

Surveys showed the following results:

• Trips on the I-85 corridor fell 18 percent, though there was an increase in the number of trips reported in the express lanes.
• More than 80 percent of drivers using the express lane were solo travelers; 9 percent were drivers with three or more people.
• Driver satisfaction increased among express lane users, many of whom had been making their I-85 trips in the general purpose lanes prior to tolling.  More than 40 percent of express lane users reported being satisfied or very satisfied with their trips on I-85 after tolling had commenced, compared to about 25% prior to tolling.  By contrast, HOV lane users became less satisfied with their I-85 trips, as many of them shifted from the HOV lane to the general purpose lanes when tolling was introduced.
• Nearly 50 percent of households making more than $150,000 owned a transponder, compared to 20 percent of households making less than $50,000.
• There was a significant decrease in the belief that highway tolls are unfair to people with limited incomes, from 74 percent to 57 percent.
• Prior to tolling, two-thirds of respondents agreed that they would use a toll route if the tolls were reasonable and would save them time.  Following tolling, however, disagreement rose sharply, resulting in evenly divided opinions on this question (41% agreed and 41% disagreed).

In Seattle, Volpe’s research demonstrated the multifaceted impacts of the State Route 520 (SR-520) variable tolling program on travel choices. Overall travel in the corridor declined, but the shares for transit and vanpools increased and drivers who chose to pay the toll experienced greater satisfaction with their trips. Average driver satisfaction with peak-hour speeds on SR-520 rose from 3.4 before tolling to 5.2 after tolling (on a 7-point scale).

In Atlanta, Volpe’s research has increased our understanding of the impacts of congestion pricing on travelers, providing officials with insights on how to better and more equitably manage travel demand along the I-85 corridor. Results show that pricing has a significant impact on route choice and affects the timing of trips, that attitudes on pricing change with direct experience, and that pricing has differential impacts on corridor users.

One of the more important takeaways that Volpe’s research provided FHWA and the cities of Seattle and Atlanta is that the more that agencies provide the public with ongoing, clear communication that explains the need for congestion pricing, the more their congestion pricing programs will be well-received.