What is the Traffic Choices Study?

In 2002, the Puget Sound Regional Council (PSRC) received a grant from the Federal Highway Administration to conduct a pilot project to see how travelers change their travel behavior (number, mode, route, and time of vehicle trips) in response to time-of-day variable charges for road use (variable or congestion-based tolling). The project, called the Traffic Choices Study, placed Global Positioning System (GPS) tolling meters in the vehicles of about 275 volunteer households. The project observed driving patterns before and after experimental tolls were charged for driving on any of the major freeways and arterials in the Seattle metropolitan area.

Why did PSRC undertake this study?

The primary aims of the study were to:

• learn about how drivers respond to the congestion-based tolling of roadways
• better understand issues of policy related to the implementation of road tolling
• test an integrated system of technical solutions to the problem of tolling a large network of roads without deploying substantial physical hardware on the roadside
• familiarize the public and policy makers with road network tolling
• generate price response data for use in other modeling and analysis
• develop an understanding of technological applications and standards
• better define a set of policy issues to be addressed in actual program design

The study has met these goals and is the most comprehensive study of demand response to network tolls in existence.

How did the study work?

The Traffic Choices Study recruited a statistically significant sample of volunteers and, after establishing their baseline “before-tolling” driving routine, began charging them for access to selected roadway facilities during particular time periods in the day. In other words, they had to pay road tolls. The study monitored driving behavior of participants for an average of approximately 18 months per household.

Participants did not lose money. They were given a travel budget (or endowment account) from which tolls were deducted. If their driving patterns remained unchanged over the experiment, they would “spend” their account balance by the time the experiment concluded. If they changed their driving patterns to reduce the amount of driving on toll roads, they would keep the difference. This method held participants financially harmless, yet offered them the incentive of keeping their leftover budget if they changed their driving patterns. In this way, the study introduced real price incentives of a toll system, and measured whether and how much participants responded to those incentives.

How were the experimental toll rates set?

The toll rates were set to permit researchers to observe driver behavior in response to paying tolls to use the roads at various times of the day. The toll rates were an approximation of tolls that would result in traffic flows that minimize the costs of travel time aggregated across all the users of the road. The experimental toll rates were only an approximation of these “efficient” tolls and should not be considered appropriate as actual policy. The toll structure can be seen on the project toll map.

Might the toll rates used in the study rates be applied to actual roads?

The experiment was an approximation of efficient pricing policy. The toll rates themselves were designed to provide an opportunity to observe the response of drivers in a controlled experiment. The experimental tolls are probably unrealistically high during some times of travel and in some locations, and too low in others. And, if tolls are to be set in the real world, there are many other factors that policymakers would take into consideration besides economic efficiency.

The study report mentions $28 billion in net benefits over a 30-year period. What does this mean?

Economic analysis often tries to estimate the benefits of a policy above and beyond the costs of implementing that policy. In this case we have estimated the benefits of variable tolling after considering the costs of implementing such as system. The benefits are primarily in the form of time savings to motorists that result from less congested roadways. These benefits accrue over time and future benefits are discounted to current or present dollars to be consistent with the estimated costs of implementing the toll system.

Is anyone actually proposing this kind of toll system for our region?

No. The Traffic Choices Study is a research effort to better understand driver behavior. The study did evaluate the costs and benefits of such a system applied to our region, but no proposal for regional tolling is being advanced. In the future, if fuel-based taxes become unworkable due to changes in how we power vehicles, a replacement for gas tax funding may need to be found. This approach to charging road users is but one idea for how to support funding for transportation investments.

Under the laws of the state of Washington, tolls cannot be placed on state facilities unless authorized by the State Legislature. Toll rates are set by the Washington State Transportation Commission.

What is variable road tolling?

Variable pricing, based on peak periods of use, is a common form of pricing in other industries. It is used when capacity is fixed in the short-run, and demand fluctuates significantly between the peak and off-peak use periods. Before cell phones, phone companies used peak-period pricing to encourage consumers to shift their use of the fixed capacity of the phone system to off-peak hours (e.g., by charging lower rates evenings and weekends). Some energy utilities use peak pricing. So do theaters. Economists recommend congestion pricing of roads for the same reason private firms use peak-period pricing: to use limited resources more efficiently.

Since road congestion is what happens during peak periods, economists tend to use the terms “variable tolling,” “congestion pricing,” “congestion-based tolling” and “peak-period pricing” synonymously.

Where can I find out more about tolling in this region?

Until recently, tolls have only been part of our state’s past, not our future. That changed when the new Tacoma Narrows Bridge opened to traffic on July 16, 2007. Tolls are being used to finance the new bridge. And a single High Occupancy Toll lane runs in each direction of SR 167 for nine miles between Renton and Auburn. The highway’s two general purpose lanes in each direction remain toll free and open to all.

The Washington State Department of Transportation has more information about tolling in Washington state. The Washington State Transportation Commission is the state’s tolling authority.

Are other regions considering variable road tolling?

Numerous regions throughout the U.S. have either already implemented variable tolling or are evaluating proposals to do so. The Federal Highway Administration’s Value Pricing Program has extensive information about what other regions are doing.

In Europe, “Road Pricing” is part of European Union transportation policy. Heavy vehicles are charged this way in Germany, Switzerland and Austria. Area charges have been implemented in London and Stockholm. And the Dutch government has plans to implement a national road pricing system for all users by 2016.

What did PSRC learn from this experiment?

The core technology for satellite-based (and whole road network) toll systems is mature and reliable. However, not all aspects of a road network tolling system have been fully demonstrated yet.

Study participants made small-scale adjustments in travel that, in aggregate, would have a major effect on transportation system performance. Done right, network tolling could provide broad benefit, including lower vehicle emissions, fewer accidents, travel time savings, improved roadway performance reliability, and lower vehicle operating costs. For motorists to be better off, however, the revenues from road tolling must be used to provide additional benefits to users of the transportation system.

Questions of fairness and maintaining privacy must be reasonably addressed before the public would accept such an approach to transportation finance. Some experience and familiarity with road tolling makes people more open to the concept, but all programs are unique and will succeed or fail on their own merits.

What will PSRC do with the results?

PSRC is publishing detailed reports from the study and will share findings with a broad audience of researchers and professionals. We will also assist other researchers in broadening our findings to help answer a range of tolling-related policy questions that are on people’s minds.

What other policy issues were raised during the course of the study?

Privacy and fairness, in particular, are issues that received quite a bit of attention.

What about privacy?

Volunteers recruited for the test were asked up front for permission to let PSRC save information concerning the routes they traveled to conduct price and time sensitivity analysis. Participation was completely voluntary.

In the future, however, privacy must be maintained before variable-pricing systems can even be considered. There are a number of ways that privacy concerns can be addressed, but all of them will require additional development and scrutiny.

Privacy is a very important issue that received some special attention during the research project; the intent was that everyone should understand the issue better by the end of the study.

Is time of day variable road tolling fair?

It all depends upon how variable pricing is used, and what happens with the revenues that would be collected. A fee to use roadways assesses costs according to those who are responsible for the costs. In this sense the pay-to-use concept is fair. However, variable pricing, like many utility-type taxes, does not always reflect an individual’s ability to pay.

To the extent that revenues are used to improve transit and roadway capacity, users may share those benefits. In theory, those with more flexibility in their schedules would be less affected than drivers with little flexibility in their work or school schedules. Once again, the key is improving travel choices. In Orange County, California, where drivers can chose to pay to access HOT lanes, there was significant usage of the toll facility by individuals at all income levels.

Conversely, if variable pricing cut peak period road use by 10-20% – as evidenced elsewhere – many of those no longer traveling cannot simply travel later or stay at home. A good percentage will still need to get to work – by transit, carpool, or other means. If alternatives are not available, commuters will return to their previous travel patterns and pricing will become a significant hardship.

These are some of the important aspects of fairness that were considered in the research study.