Executive Summary
This report presents the results of the United States Department of Transportation (U.S. DOT) Federal Highway Administration (FHWA) study of methods for before-after measurement of the travel and environmental impacts resulting from congestion pricing projects. Congestion pricing encompasses a variety of strategies which feature roadway facility charges to reduce traffic congestion, such as high occupancy toll (HOT) lanes which allow vehicles with fewer occupants to pay a charge to access lanes available to vehicles with more occupants at no charge or at a reduced charge. Other congestion pricing strategies include zone-based pricing where vehicles are charged to enter or drive within a specific geographic area and full roadway pricing in which a toll is imposed upon a previously un-tolled roadway.
This report presents a summary and analysis of current practices as well as a set of recommended practices for conducting before-after evaluations of congestion pricing projects. This study focuses on the environmental impact areas most commonly considered in the literature: air quality, noise, and environmental justice—sometimes termed “equity”—which considers how impacts distribute across different types of people, especially low income and minority groups. Since environmental impacts are driven by the broader travel impacts of congestion pricing projects, this study also investigated state-of-the-practice and assessed gaps in travel evaluation methodologies, including traffic, transit and traveler behavior.
Published literature providing detailed environmental evaluation methodology information is not plentiful and there are very few critical assessments of the state-of-the-art, limitations and best practices. This study is intended to address that gap and provide recommendations that will inform U.S. DOT congestion pricing evaluations such as the approximately $1B Urban Partnership Agreement and Congestion Reduction Demonstration (UPA/CRD) program and other future projects.
A sample of eight projects from among the more than 70 projects that were identified worldwide was selected for analysis. The sample included only before-after evaluations, a variety of project/study types, and several high-visibility, frequently cited projects which have not been investigated from an environmental evaluation methodology perspective. The eight study projects are:
- Oregon Mileage Fee Concept and Road User Fee Pilot Program
- Puget Sound Traffic Choices Study
- Commute Atlanta Mileage-Based Value Pricing Demonstration
- Minnesota Interstate 394 MnPASS HOT Lanes
- San Diego Interstate 15 HOT Lanes
- The Stockholm Trial
- Central London Congestion Charging
- Singapore Area Pricing.
The findings presented in this report are based on a review of the published literature associated with these eight study projects as well as some of the relatively scarce congestion pricing synthesis reports. These include reports from the Federal Highway Administration Value Pricing Pilot Program and the European Commission’s “Coordination of Urban Road User Charging Organisational Issues” (CURACAO) reports.
State-of-the practice and knowledge gaps and limitations are presented separately below for travel and environmental impacts.
- The most common impact areas considered include traffic (either describing the usage of a roadway, such as traffic volumes, or the performance of a roadway, such as average speeds), transit (either describing usage, such as ridership, or performance, such as schedule adherence), and traveler behavior, e.g., route, mode, and time of travel.
- The most common traffic impact performance measures, used in nearly every study project, are: traffic volumes, vehicle miles traveled (VMT) and average speeds. Travel time is also frequently considered.
- Most data for traffic performance measures are objective data, that is, data collected in the field using a wide variety of mostly automated techniques. One important area, vehicle occupancy (key to mode share and other person trip considerations), is often collected manually through visual observation. Probe vehicles, whether driven by evaluators or general public volunteers, are becoming increasingly common for traffic data collection.
- There is less variation and fewer performance measures in the area of transit impacts. The most common measure, included in most evaluations that examine anything other than traffic impacts, is transit ridership, which is often collected automatically using on-board sensors. Bus travel times, schedule adherence, and rider perceptions are less common.
- Consideration of traveler behavior impacts is common, but it is not included in all evaluations. Most evaluations that consider traveler behavior impacts use similar measures, including time of travel, route, mode, origin and destination, and collect the data using traveler surveys.
- Traveler behavior data are usually collected through panel (or “longitudinal”) surveys in which the same people participate in the before and after surveys. More robust evaluations survey all adult travelers within a household; other surveys include only one traveler from a household. More robust evaluations survey both general traveler behavior (i.e., focusing on “typical” travel) and collect detailed trip information for one to three specific days using travel diaries. Across evaluations in general, travel diaries are not common. Coupling travel diaries with instruments in the respondents’ vehicles to record actual mileage and other data is rare.
Table ES-1 identifies those areas where the current understanding of the travel impacts of congestion pricing projects is stronger as well as the areas where there are gaps and limitations.
Table ES-1. Travel Impact Knowledge Gaps
Better Understood Impacts
(Knowledge) |
Less Understood Impacts
(Knowledge Gaps) |
Short-term impacts (from a few months up to a year after deployment) |
Long term impacts |
Localized impacts |
Regional impacts |
Cumulative impacts (projects plus exogenous factors) |
Project-attributable impacts |
Individual travel behavior changes |
Household travel behavior changes |
Vehicle trips |
Person trips |
Average speeds |
Vehicle speed fluctuations (driving cycle) |
Average performance |
Variability in performance (reliability) |
Transit ridership changes |
Transit crowding implications |
Most of the gaps and limitations have implications for environmental impact evaluation. Long term impacts refers to the understanding of land use changes such as changes in home or work locations that will only fully manifest over time periods far beyond typical project evaluation timeframes. In the case of regional impacts, it is not that most evaluations are failing to consider expected regional impacts but rather that there has not been enough analysis to understand even whether such impacts are likely. The failure to fully understand the influence of exogenous factors is a pervasive problem. Few evaluations understand those influences well enough to quantitatively adjust observed impacts to reflect only the pricing project. Of significance to air quality is the fact that there is a poor understanding of how congestion pricing projects impact traffic flow in ways that significantly impact vehicle driving cycles.
- Most evaluations do consider environmental impacts; most commonly air quality, noise and environmental justice.
- There is little variation in air quality or noise impact evaluation methodologies among evaluations.
- Air quality analyses consider project-related vehicle emissions and/or ambient pollution concentrations; the former are always calculated and the latter are always measured using roadside monitors. Calculation of emissions is more common than monitoring, which normally does not allow differentiation of project-attributable changes.
- There is little variation in methods used to calculate vehicle emissions. Emission rates (or “factors”) expressing emissions of various pollutants in grams per mile at various average speeds are derived using models and applied to roadway link-specific, observed VMT at various speeds (usually observed speeds) to determine emissions. Total emissions are determined by summing all of the study roadway links.
- In the U.S., vehicle emission rates have been developed using the Environmental Protection Agency (EPA) MOBILE model or, if the project is in California, using the similar California Air Resources Board EMission FACtors (EMFAC) model. These models require region-specific inputs on temperature, vehicle fleets, fuel types and vehicle inspection and maintenance programs. These models provide little ability to examine impacts of vehicle driving cycle (proportion of travel under acceleration, deceleration, cruise and idle) impacts.
- Like air quality impacts, noise impacts are examined by calculating noise levels—often with the FHWA Traffic Noise Model—or by roadside monitoring of ambient noise levels. No examples of project-attributable, significant noise impacts were found in the literature. Most evaluations show no perceptible changes in noise levels.
Table ES-2 shows those areas where the current understanding of the air quality impacts of congestion pricing projects is stronger as well as the areas where there are gaps and limitations. Some of the gaps, such as uncertainty regarding project-attributable changes in VMT and speeds and lack of driving cycle changes—flow directly from limitations in traffic impact evaluation. Others, such as under-consideration of hourly variations in VMT and speeds are less about a lack of traffic data and more about the choice of air quality impact methodology.
Table ES-2. Air Quality Knowledge Gaps
Better Understood Components of
Congestion Pricing Vehicle Emission Changes |
Less Understood Components of
Congestion Pricing Vehicle Emission Changes |
Cumulative impacts (VMT, speed) |
Project-attributable impacts (VMT, speed) |
Localized impacts |
Regional impacts |
Average daily impacts |
Hourly variation (VMT, speed) |
|
Driving cycle changes (traffic flow change) |
|
Vehicle mix |
It is unlikely that most congestion pricing projects will demonstrate the magnitude of traffic volume or speed changes necessary to produce perceptible changes in noise levels. The smallest noise level change perceptible to most people—about 3dBA—requires a doubling or halving of traffic volume.
Gaps and limitations regarding environmental justice are somewhat less clear, but generally include the following:
- The need for more results overall to support development of more standardized approaches and to solidify knowledge.
- The need to more fully explore geographic location and other “horizontal equity” issues (issues unrelated to income levels and ability to pay) and focus less exclusively on the “vertical equity” issue of the varying incomes and abilities to pay of different stakeholder groups.
- Greater investigation of long-term impacts.
- More research into how the uses of pricing revenues impact perceived and actual equity.
Key air quality evaluation recommendations include the following:
- Calculate vehicle emissions (ambient monitoring can also be performed if resources permit but the priority should be on emission calculations).
- In those unusual cases where a congestion pricing project is likely to significantly increase localized traffic congestion, especially near sensitive land uses such as nursing homes or schools, consider using a dispersion model such as the EPA CAL3QHCR carbon monoxide model to estimate project-attributable pollutant levels.
- Always use a “project-level” analysis that considers project VMT and speed impacts on individual roadway links.
- Select pollutants for analysis based on local air quality attainment status and issues. Common pollutants of interest include the criteria pollutants carbon monoxide, nitrogen oxides (ozone precursors), volatile organic compounds, and particulate matter; the greenhouse gas-related pollutants carbon dioxide and methane; and the mobile source air toxic benzene.
- The geographic area of analysis should include as many of the roadway links as possible that are expected to be significantly impacted (e.g., a ±5 mile per hour change in average speed) as possible. Include at least all freeways and major arterials within the priced zone and the parallel routes that may experience significant traffic diversion.
- Collect at least a couple of months of VMT and speed data and a full year’s worth if possible. Longer data collection time frames allow seasonal variation to be controlled and exploration of changes as travelers “settle into” their responses to the pricing project.
- To the extent possible, utilize VMT and average speed data that reflect only project-attributable changes by controlling for exogenous factors. Methods for determining whether and how much exogenous factors have impacted observed traffic data include:
- Comparisons to control roadways/corridors/areas
- Statistical modeling that can remove or control for the effect of exogenous factors by including such variables in multivariate equations
- Utilization of household survey data (travel diary data being ideal) to understand the causes behind reported changes in travel behavior
- Tracking fuel prices and employment levels
- Elimination of traffic data from times and locations within the study area characterized by severe weather, significant traffic incidents, and significant roadway construction
- Collecting before and after tracking data for the same month(s) of the year
- Examination of historic traffic trends.
- Calculate total daily emissions by summing calculated hourly emissions based on hourly VMT and average speed data; as opposed to calculating daily emissions based on 24-hour VMT and average speeds.
- Test for possible driving cycle changes and if present, collect observed driving cycle data using “floating car” test vehicle procedures (in which vehicles are equipped with Global Positioning Systems and drive train sensor) and use the new EPA MOVES model (Motor Vehicle Emission Simulator) to calculate emissions, as MOBILE and EMFAC provide little to no consideration of driving cycle changes.
- Understand and carefully consider the inputs and default variables and methods utilized in the emission rate (e.g., EMFAC or MOBILE) or emissions model (e.g., MOVES) utilized as they can have significant impacts on calculated emissions. Reflect all local, user-specified inputs as accurately as possible.
- Test for before-after changes in vehicle mix (the proportion of different vehicle types) and if present, vary VMT-by-vehicle type breakdown accordingly when calculating emissions.
Congestion pricing project evaluations have not shown significant, project-attributable noise impacts and since few congestion pricing projects are likely to produce the dramatic increases in traffic volumes necessary to produce perceptible changes in noise levels, noise analysis is not recommended as a standard component of project evaluation. The recommended framework therefore focuses on air quality and environmental justice.
Key environmental justice recommendations include the following:
- Include the common tools and techniques which provide a solid foundation for understanding environmental justice impacts:
- Regional geographic information systems to map the locations of low-income and minority populations within the likely impact area.
- Attitudinal surveys, interviews and focus groups of the general public, corridor travelers and specific types of residents and travelers to gather attitude and perception as well as general travel behavior data.
- Travel diary surveys to gather detailed, specific travel behavior data of various groups of interest.
- Consider broadly how the congestion pricing project impacts different types of people across a wide range of dimensions, not just income and minority status. Other important factors include access to transit, access to private vehicle, and residential and work locations.
- Integrate the collection of demographic data as fully as possible into the overall evaluation data collection plan. Such data can and should be collected in any and all surveys, but also consider how other data of environmental justice importance can be collected in other ways, such as origin-destination information via license plate recognition technology.
- Explicitly consider the transportation environmental justice implications of how congestion pricing project revenues are reinvested.
