Lessons Learned From International Experience in Congestion Pricing

4.0 Conclusions and Implications

More than three decades of experience overseas with pricing programs and studies provides a rich body of knowledge about how travelers respond to price, the technology associated with implementing these concepts, the determinants of public acceptance, and other policy considerations.  Much like experience in the United States, some of the work done under these overseas pricing initiatives resulted in actual implementation.  Other work involved studies that did not result in implementation, but often provided valuable insights into potential user responses and policy issues needing attention. 

This section ties together the lessons learned from the projects and experience syntheses described in the preceding Sections 2 and 3 – experience focusing on areawide pricing that has yet to be tried in the United States. The discussion of lessons learned includes the effects of areawide pricing on: travel and traffic, costs and revenues, economy and business, equity and the environment. In addition, lessons are drawn relating to the successful adoption of areawide pricing and public acceptance requirements; and issues relating to the adequacy of technology for the administration and enforcement of pricing programs. Lessons are followed by implications of the lessons learned for U.S. officials and planners interested in pursuing areawide pricing programs.

4.1. Effects of Pricing

Mobility Impacts

Pricing Reduces Congestion Significantly and Sustains It Over Time

Without exception, areawide pricing strategies that have been implemented abroad have met their principal objective of reducing congestion and sustaining the relief of congestion over long periods. Areawide pricing in Singapore, London and Stockholm resulted in 10 to 30 percent or greater reduction in traffic in the priced zone and has sustained the reductions over time. Pricing has encouraged travelers to change their behavior by changing modes of travel, times, routes, or trip frequency. In the three areawide pricing programs described in Section 2, up to 50 percent of those foregoing car travel to the priced zone in the priced periods shifted to public transportation. In London and Stockholm, the greatest shift was to public transportation while in Singapore it was to 4+ carpools and to the shoulder time just before the start of pricing. The traffic reductions in priced zones in Singapore and London have been sustained permanently (over thirty years in Singapore and five years in London). The bypass routes in Singapore and London have seen modest but not overwhelming increase in traffic. The nature and magnitude of impacts on travelers and traffic has been found to depend on many factors including: price levels and schedules, coverage, pricing periods, the existing traffic levels and demographic characteristics.

Delays due to queues declined dramatically. Speeds increased significantly within the zone as well as outside along approach roads. A ten to 30% increase in speed has been realized. Buses in Singapore and London have particularly benefited from speed increases that now allow them to operate faster and enjoy operational productivity gains. Travel time reliability also has increased in the three cities for trips to the center. Again, this allows buses to offer more reliable service.

The success of areawide pricing in reducing congestion has raised questions regarding the many exemptions that have come to be seen as diluting the benefits. Singapore has eliminated many of the exemptions over time.

Cost and Revenue Impacts

Areawide Pricing Has Been an Important Source of Revenue For Transportation

Although the primary rationale for areawide pricing projects has been to improve mobility and efficiency, relatively large net revenues from pricing clearly have been of major interest to the sponsors. The significant revenues generated by pricing have been seen as an important source of benefits in all three projects reviewed in this report. Project revenues in London and Stockholm (as well as in toll cordon projects in Norwegian cities) have been used to cover operating and enforcement costs first and remaining revenues have funded improvements to bus and rail services. In London and Stockholm, the desire and ability to use pricing program revenues for public transportation was a major objective and “selling” point. In Singapore, while the revenues are not directly earmarked for public transportation, the availability of these funds probably has allowed the government to more easily pursue ambitious public transportation programs.

Areawide pricing projects are generating revenues far in excess of costs. In Singapore’s Area Licensing Program, revenues were more than 11 times the operating costs. The revenues from the central area cordon pricing component are nearly 14 times the operating costs.  If capital costs are included, the revenues are still 2.5 times the costs. These numbers imply very attractive financial “pay back” periods. For the London charging program, the revenues have been a little over twice the operating costs. Inclusion of capital costs brings this ratio down only marginally (to slightly over 1.8) because of the relatively low capital costs of the inexpensive technology adopted. In the Stockholm trial, the revenue to operating cost ratio was 4:1; and when capital cost were included, it was 2.5:1.

Economic Productivity and Business Impacts

Areawide Pricing Has Generated Positive B/C Ratios and Small Or Neutral But Evidently No Significant Negative Economic Impacts

Areawide congestion pricing applications likely have realized societal economic benefits in excess of costs (implying positive benefit to cost ratios and reasonable rates of return).

Singapore’s 1975 ALS program is estimated to have achieved a rate of return on investment of, at least, 15%, even without inclusion of realized savings other than the value of time savings and without the exclusion of the costs of expensive and poorly utilized PAR facilities. Some believe that the returns would have been higher also if the price had been somewhat lower. The economic returns also would have been much greater had variable instead of flat pricing had been used.

London scheme is estimated by TfL to have generated a B/C ratio of 1.4. Again, it is believed that a different charge and/or variable prices tied more closely to congestion levels likely would have achieved higher B/C ratio.

Regarding business impacts, in Singapore, surveys suggested that the ALS pricing did not, by itself, initiate changes in business conditions or location patterns. Overall, the business community responded positively to the ALS, probably believing that the combined package of actions by the government was necessary and beneficial in the long run.

TfL (2006, 2007) reports that the pricing in London seems to have had broadly neutral economic impacts. Annual surveys do suggest, however, that businesses in the charging zone have outperformed those outside and a majority of businesses continue to support the charging scheme, provided investments in public transportation is continued.

In Stockholm surveys, albeit over a very short time span of trial, no identifiable impacts on retail business or household purchasing power were identified.

CURACAO (2007) summarizes the implications of regional economic impacts by finding a “generally low level of measured impact.” While the result may be partly attributable to the unique economic vitality and strength of the cities in which pricing occurs, there is no evidence of economic damage. CURACAO calls for continued research on the impacts of road pricing on local and regional economies.

Environmental Impacts

Areawide Pricing Has Shown Some Positive Environmental Impacts and Energy Benefits

Better environment has been one of the primary objectives of the Stockholm areawide program. Environmental improvement has not been a stated major objective behind London and Singapore pricing programs. However, all three have made attempts at monitoring and measuring air quality implications of pricing recognizing that pricing would alter the operating speeds, number and timing of trips or even the mode on which trips are taken. These changes, in turn, could be expected to lead to reductions in fuel consumption and consequent reductions in vehicle emissions, resulting either from fewer vehicle trips being taken or smoother traffic flows under less congested conditions.

Evaluators in Singapore concluded that the tailpipe emissions most likely declined in the RZ because there was such a large reduction in automobile travel. Regarding smoke and haze, measurements showed declines, but they could not be unambiguously attributed to the ALS pricing. Subsequent surveys also revealed overall reductions in the RZ as a result of reduced and more dispersed resulting automobile travel patterns.

TfL in London has reported changes in air quality within and alongside the Inner Ring Road boundary of the zone. Levels of NOX fell by 13.4% between 2002 & 2003, CO2 by 15%, and particulates (PM10) by 7%. TfL (2007) states that between 2003 and 2006, NOX emissions fell by 17%, PM10 by 24% and CO2 by 3%, with some being attributed to the effects of reduced levels of traffic flowing better, with the majority being as a result of improved vehicle technology. In total, the rate of fall in CO2 has been 20%. The TfL report makes it clear, however, that only the initial reduction of emissions could be expected from the introduction of the charge. Further reductions are unlikely to be as a result of the charge.

Achievement of environmental objectives was supported by the observed reduction in the inner city of 10-14% in Carbon Dioxide (2-3% in the County), 7% in NOX and 9% in particulates. Furthermore, emissions declined in the “right” areas near population centers. there was no measurable change in noise impacts.

The results from all three areawide pricing projects indicate that areawide pricing may have some potential for reducing pollution. Except for CO2, other pollutants may be one-time reduced at the start of the program. On the other hand, indirect effect of public transportation expansion, made possible by the congestion charge revenues, has the potential to reduce all pollutants and sustain reductions over time.

Equity Impacts

Especially Income Equity Impacts Have Received General Analytic Attention, But Little Project Level Evaluation. Still, Available Findings Suggest Income Equity Is Manageable

Equity issues occupy a considerable literature as reviewed above, with focus on concepts of equity, modeling of impacts and pricing designs to address income equity issues. At the level of projects or proposed projects, Singapore has examined equity impacts; Edinburgh has grappled with equity and general fairness considerations and several tollrings in Norway have designed schemes with equity in mind but not done detailed equity evaluations after project implementation.

Research reviewed points out that equity issues are ever-present, but project evidence suggests they can be successfully dealt with. Any change, in the way charges are made for road use, will result in winners and losers. Some road users will value time savings and reliability more than the cost of the toll, others will not. Those who are “tolled off” the priced facility may shift to off-peak times or alternative routes, decide to carpool or switch to a different mode of travel, or simply make fewer trips. Changes in trip making behavior may also change patterns of commerce and affect businesses differently. Researchers also note the perception of fairness depends on how the revenues are used and what alternative policies are considered as ways of dealing with congestion.

Regarding specific cities reviewed for pricing activity, the perception that congestion pricing is “unfair” to low income drivers has not been a major concern in Singapore, London and Stockholm after implementation. Findings from Singapore are most in depth, though experience in the proposed Edinburgh program also is instructive.

The Singapore Area Licensing (ALS) project explicitly looked at equity implications by modeling, attitudinal surveys and before/after travel analysis. The results of modeling analysis based on before and after user survey data suggested that gainers outnumbered losers 52 to 48%. Attitudinal surveys carried out after the introduction of ALS pricing also provided indications regarding equity across various dimensions. Pedestrians, taxi riders and residents outside of RZ found the impact of ALS as neutral or negative while cyclists, bus passengers and residents within the RZ judged the ALS as favorable. Car drivers and passengers judged the ALS as mildly unfavorable. Overall, middle income travelers felt adversely affected by the ALS.

Travel evaluations and stakeholder surveys  in Singapore also examined impacts across income groups. The data on shift from cars to buses as a result of ALS in 1975 showed, somewhat counter intuitively, that the increases in transit were fairly uniform for low, medium and high income peak period travelers to the RZ.  The evaluators concluded that, overall, there were only small differences among income groups in modal response to the ALS. There was also no evidence that trip times increased or decreased more for any particular income group. Al in all, the evaluators did not find that low income travelers suffered more than high income ones due to the ALS pricing.

Addressing the fairness of pricing in Singapore, supporters claim that without pricing there was, and would be, a great imbalance between the travel conditions enjoyed by car drivers (a minority of travelers) compared to the majority who use alternative modes. ALS is perceived as far from unfair because it is said to have redressed a greater inequity and has allowed much greater increase in public transportation and road efficiency. 

In Edinburgh, issues of revenue distribution and transit improvements were vital to geographic equity considerations. Non-city residents viewed revenue distribution plans as unfair since they would pay the charge but not get any direct benefit. A key institutional issue appears to be neighboring authorities had no legal grounds to support public transport improvements which might have appealed to non-city residents. City residents would also have benefited disproportionately more from the public transport improvements. In short, Edinburgh shows geographic equity and improvement plans can make or break pricing plans. 

In light of Edinburgh findings, certain CURACAO (2007) conclusions appear especially relevant: “It therefore remains important to understand the scale of both vertical and horizontal inequities, and this will require a disaggregated analysis by person type, income level, journey type and specific person and journey characteristics.” And, “The principal solutions to equity problems lie in the design of the scheme itself, including location, time of day and level of charge; the use of exemptions and rebates; the application of complementary policies, particularly to provide alternatives; and the use of surplus revenues to provide direct or indirect support.”

4.2. Feasibility and Implementation of Pricing

A number of factors influence the prospects for successful implementation of pricing, including policy and institutional issues, public outreach and acceptance programs, and technological factors.

Public Policy and Institutional Issues

Areawide Pricing Projects Often Require New Policy and Institutional Arrangements

Major national level legislative initiatives were enacted before areawide pricing could be implemented in London and Stockholm. The experience shows that formal actions agreements may be needed for: power to impose and collect charges; use of selected technology to administer and enforce charges; to cite violators and collect fines; make modifications to the pricing scheme; and for the use of revenues from the charges. Experience also shows that policy and institutional arrangements and agreements have profound impact on public and political acceptability of areawide pricing proposals and operational success. Acceptability research shows stakeholder involvement and funding across government levels also are important.

Acceptability of Areawide Pricing

Successful Projects Depend on Effective Outreach and Sensitivity to Public Acceptability

Public opinion is perhaps the most critical determinant of the prospect for successful pricing project implementation. All of the projects overseas have paid considerable attention to measuring public attitudes and reaching out to the public, stakeholders, and elected officials so that they understand the new concepts in transportation that pricing represent.  As described in Section 2, projects typically are initiated after considerable public outreach and stakeholder involvement, and many have made changes in response to or anticipation of certain public attitudes and reactions. Outreach efforts as part of initial feasibility studies often find neutral or skeptical opinions, or outright resistance, but this is often followed by acceptance as projects get underway.  The outreach efforts, often using focus groups, surveys, and public forums, have resulted in a better understanding of public concerns that have been incorporated into project designs and public education materials. The support of a key stakeholder and/or a senior politician who was able to influence public opinion also was crucial for the successful implementation of several projects.

The evidence suggests that acceptability of areawide pricing schemes increase after implementation but business support for congestion charging may continue to be relatively mixed. Businesses in London are more supportive of the scheme than opposed to it. A majority of businesses continue to support the scheme, provided that there is continued investment in public transport.

Based on project experience and public opinion studies on pricing, certain key factors emerge as potential determinants of public acceptance of pricing as highlighted in the Section 3:

• The problem addressed resonates. Whatever the mix of problems addressed by pricing proposals, whether congestion, pollution or some combination, acceptability is enhanced where the problem is clear and severe to affected parties. Congestion may or may not be the most painful candidate problem for pricing; pollution may be more resonant. Pricing plans enhance implementation prospects when they home in on the most resonant problem or problems.
• Pricing Is Convincingly Effective: Acceptability studies suggest the public or decision makers may be skeptical about the effectiveness of pricing in reducing congestion or pollution. The implication is proposals will have better prospects where they can demonstrate effectiveness, perhaps by reference to like projects or through well evaluated test programs or both.

• Program Design Meets Program Concerns: Acceptability of pricing is enhanced where pricing program parameters are in line with public and decision maker concerns. Top concerns will vary by area, but planners increase the odds of acceptance by determining the concerns and structuring the program accordingly. Some top concerns may be about “free riders” and enforcement; others may be about complexity of technology; others about specific groups facing hardship or adverse boundary effects. Implementation prospects improve with full attention to specific concerns.

• Revenue Distribution Follows Preferences: Gearing revenues toward the most favored purposes is important to acceptability. Research shows revenues directed toward transit and/or road improvements may garner support in some locations, but may compete with other preferences elsewhere, including possible tax reductions.

• Fairness Is Broadly Addressed: Equity across income groups subject to pricing often leads equity discussions among analysts of road pricing. However, research shows acceptability does not vary greatly across income groups and equity defined more broadly may dominate and deserve more attention. Specifically, these fairness perspectives may be key: fairness of outcomes, i.e. assurance some are not evading the pricing scheme who should be paying; “procedural” fairness, i.e. people feeling full opportunity to participate in developing pricing plans; fairness to special groups, e.g. handicapped or emergency workers; use and spatial fairness  where occasional payers reap the same benefit from new roads and transit as frequent users; and ways to moderate different treatment of travelers within or to/from a cordon scheme.

• Government Planners Are Open, Responsive, Resourceful Solution Partners: Numerous findings suggest how government pricing planners are perceived may be as important to acceptance as the nature of their pricing proposal(s). It seems if government has at least some favorable image coping with bottlenecks, improving transit, improving traffic management, acceptability of pricing proposals is enhanced – and visa versa. Likewise important is sensitivity to governmental image as a taxing entity with already sufficient resources to deal with congestion. Transparency in pricing planning and decision making also will enhance acceptability, including the degree to which non-pricing options have been examined; and the extent of reference to pricing experience elsewhere. Finally, government as resourceful partner in the solution is important to acceptance, suggesting state and national governmental agreements and matching funds may be a necessary step.

• Pricing Schemes Operate Over Time: A consistent finding is acceptance tends to grow the longer pricing programs are in existence. The exact reason for growing acceptance are not well explored. It may have to do with experiencing demonstrating no harm to business, absence of feared queues at tollgates and the visible, proven link between revenues and transportation improvements. In any case, growing familiarity with successful operations seems to enhance acceptability over time.

Technology For Areawide Pricing

Effective, Reliable and Acceptable Pricing and Enforcement Technologies Are Key 

Technology is important the success of most pricing concepts, and the technology has been generally up to the task.  Different technologies for pricing and enforcement, both low and high end, generally are proving reliable and effective.

Singapore Area License Scheme (ALS) windshield license and manual enforcement system worked well. The Electronic Road Pricing (ERP) system, based on Direct Short Range Communication (DSRC) and In-vehicle Unit (IU) transponders with stored value “Smart Cards” technologies, is working seamlessly. It allows open road, multilane variable tolling at 120 KPH speeds and addresses privacy concerns.

London’s Automatic Number Plate Recognition (ANPR) charge collection and enforcement system has been effective, though TfL is proposing to move to an electronic system over the next few years in order to reduce administrative costs and allow variable pricing schedules. The DSRC and IU based system, supplemented by ANPR also worked well in the six-month trial.

Steps To Areawide Pricing

The areawide pricing programs abroad parallel several lessons stemming from U.S. pricing programs about how to conceptualize, plan, discuss, and carry out a successful pricing project:

Define the problem – Proposals for pricing solutions need to focus on the costs of severe congestion, including traffic delay, air quality problems, accidents, lost productivity, or other locally perceived problems. Proposals need to show how pricing will address these problems, and how pricing compares to alternative potential solutions. The revenue raising aspects of pricing programs are also important considerations and these need to be compared (in terms of revenue productivity, equity, etc.) to more traditional transportation financing approaches.

Take time to include all interests – Pricing is a significant departure from existing practices, and it may have far reaching impacts and necessitate realignment of existing institutional relationships. Several public and private interests and agencies are likely to have a stake in the workings and outcomes of pricing proposals. All these stakeholders and interests need to be involved in project development. This takes time but will make for a much more successful outcome.

Consider a full range of alternatives – Pricing should be viewed in the context of a range of strategies for addressing congestion and related problems. Alternative applications of pricing should be considered and an incremental strategy with continuous evaluation and potential broader applications should be contemplated.

Carefully attend to the estimation of impacts – The estimation of a variety of potential impacts of pricing is both difficult and essential. Impact estimation difficulties stem from the lack of experience with road pricing and limitations of forecasting tools. Fortunately, as findings summarized here indicate, impacts of pricing strategies are documented and can aid planners and local decision makers in estimating potential ranges of impacts.

Introduce pricing as a part of a package – Alternative travel mode enhancements and travel demand management programs should be considered along with pricing. Alternative uses of pricing revenues should be proposed and assessed.

Focus on customer relations – Public outreach and education have been a critical determinant of acceptance of pricing programs by public and decision makers. Focus groups, public opinion surveys and media campaigns have all contributed to project successes and should be attended to all through planning, implementation and evaluation of future pricing programs.

4.3 A Look to The Future

Interest in Areawide Pricing is Growing

Pricing projects overseas are breaking new ground and providing important lessons for those interested in exploring the use of pricing to address traffic congestion and transportation funding problems.  The implemented projects have shown that pricing can lead to more efficient use of existing roads.  Theorists have predicted this for decades, but the implemented projects have confirmed that people respond to price signals when making transportation decisions, just as they do in other aspects of their economic lives, and those responses can help diminish congestion and support alternatives to driving.  The projects have also been a test-bed for the technologies that enable pricing to be used, as well as a discussion forum for important issues related to economic and environmental impacts, equity and public perceptions.

Whereas areawide pricing was once something talked about by a small corner of academia, it is now front-page news all over the World, and is openly discussed by transportation professionals, interest groups, and elected officials.  Areawide pricing has come to be viewed as an innovative way of coping with recurring congestion problems and as a source of funding and an effective complement to existing transportation improvement programs.

Revenue-generating Effects of Pricing Provide Important Stimulus

Although the primary rationale for areawide pricing is that it improves mobility, the revenue-generating effects of pricing have also been an equally important reason for interest overseas in this approach to reducing congestion. In contrast to the pricing projects in U.S. that have focused on single facilities and generated moderate revenues, areawide pricing programs overseas are more than self-supporting and generate large revenues far in excess of costs. Overall, areawide pricing shows considerable promise in a time where transportation agencies are struggling to find new and robust approaches for financing transportation programs. Moreover, it has been shown that, with proper attention to acceptability issues specific to project areas, pricing can be a fair and equitable part of a road user charge program.

Much Remains to be Learned

Much has been learned about the promise and potential of areawide pricing over the last several years, yet much more remains to be learned.  Long-run impacts on land use, auto ownership, business and productivity need to be monitored more carefully over time. Continued progress in implementing acceptable, effective and ultimately informative pricing programs has required careful planning, coalition building, public education and participation, and sufficient time and resources for the development of well designed and locally acceptable project plans. Several models for effective and workable ways to proceed are shown in this report.

Areawide pricing holds the promise of reducing congestion, enhancing mobility and economic productivity, reducing environmental and energy costs, and providing new sources of funding for transportation investments.  Yet, despite this potential, the concept of congestion pricing remains controversial in many potential applications.  It involves a new approach to dealing with congestion problems and charging for road use.  However, the overseas programs demonstrate favorable outcomes are possible with careful and inclusive planning and outreach. 

The overseas experience provides a valuable guide to planners in exploring the feasibility of future pricing applications and identifying projects for implementation. A particularly important consideration in the U.S. cities considering areawide is the use of revenues generated by pricing to address and mitigate plausible equity impacts. At the same time, operating pricing programs abroad have demonstrated broad definition and attention to fairness versus simple income equity, a wise focus given the extensive findings summarized here on the link between fairness and acceptability.

Overseas experience suggests the need for focusing more on the potential environmental and energy benefits of pricing. Overseas analysts have made limited but important preliminary findings about air quality impacts, and more work can be expected. While the air quality and energy conservation benefits of small-scale U.S. pricing projects implemented to date may have modest effects on overall regional environmental quality, areawide projects beginning to receive attention in the U.S. will do well to pay attention to results and methods from overseas, as areawide systems may have more significant environmental and health benefits worthy of tracking.