System Trials to Demonstrate Mileage-Based Road Use Charges (2010)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

60 5. ADDITIONAL ISSUES TO EXAMINE IN THE TRIALS The preceding chapters in this report outlined the policy objectives that might be addressed by a system of VMT fees, identified and characterized technical mechanisms and design strategies for implementing and deploying VMT fees, and applied criteria from NSTIFC (2009) to discern promising options to examine in the context of expanded system trials. The report now considers a broader range of questions related to the scope and structure for the trials. First, in addition to specific technical approaches, what additional issues or uncertainties (e.g., cost, user acceptance, institutional structure) would be beneficial to explore in the trials? Second, how might the trials be organized and structured to resolve these issues and uncertainties in order to inform debate and prepare for the potential implementation of VMT fees? This chapter begins by cataloging the many issues and uncertainties related to VMT fees that could be helpful to address. Next, it considers which of these could be explored or resolved in the context of system trials. Finally, it discusses the process employed by the research team, including interviews and a one-day expert workshop, to solicit input from subject matter experts and stakeholders on how to design and organize the trials to address the critical issues and uncertainties. Chapter 6 outlines high-level strategic observations that emerged from the interviews and workshop, while Chapter 7 summarizes more detailed findings. 5.1. ISSUES AND UNCERTAINTIES TO RESOLVE Implementing and transitioning to a system of VMT fees would present a high degree of technical, institutional, and political complexity. While the trials conducted to date (see, for example, Kuhl 2007, ODOT 2010, PSRC 2008, and Whitty 2007) have explored various fee structures and metering and collection mechanisms, numerous issues have yet to be addressed. Some, such as the cost of certain implementation mechanisms implemented at scale, would require further investigation; others, such as the types of pricing structures that the system should be able to support, could be resolved through policy decisions. Remaining uncertainties can be organized in the following categories: system requirements, technical approaches, institutional arrangements, implementation and transition strategies, user acceptance, and system cost. Clearly many of these are interrelated, and there is some inevitable overlap among the categories. For example, the choice of technology might affect both user acceptance and system cost. Still, it is helpful to lay the issues out in a systematic fashion to the extent possible. 5.1.1. Broad System Requirements Lack of agreement among policymakers regarding required capabilities would hinder the ability to determine an appropriate design for a VMT-fee system. Key issues to resolve would include: • Metering capabilities. The intent to support a given set of pricing policies implies a required set of metering capabilities—for example, the ability to meter mileage by jurisdiction, by smaller area, by specific route, and/or by time—and this in turn might constrain the set of implementation mechanisms that could be considered.

61 • Applicable classes of vehicles. In terms of vehicle classes, one question to resolve is whether the system would apply to passenger vehicles, to trucks, or to both. Another is whether the system would only apply to new vehicles or if instead it would be necessary to retrofit older vehicles with metering equipment. • Geographic scope. It remains unclear whether a system of VMT fees would initially be implemented at the federal level, with states having the option to make use of the system if they wish, or would instead be implemented only in certain states to begin with. • Accuracy. The required level of accuracy for metering mileage (e.g., ensuring that mileage recorded along a particular route did in fact occur along that route) and assessing fees would need to be specified. • Privacy. Various degrees of privacy protection could be provided within the system, ranging from reasonable expectations of confidentiality to total anonymity. The question of which levels of privacy protection the system would need to support has yet to be determined. • Payment. Another requirement to consider is which forms of payment—e.g., credit, cash, or check—the system should facilitate, an important question to consider given that not all drivers have bank accounts or credit cards. • Interoperability. Interoperability standards could be employed to facilitate interactions among VMT-fee systems in different states or to enable multiple vendors to compete for the provision of metering devices and billing services. The question is whether interoperability would be viewed as a desirable goal or an absolute requirement for the system. • Complementary functions. Another issue to consider in specifying requirements for the system is whether the in-vehicle metering equipment would be designed solely to support VMT fees or instead might support additional driver services. 5.1.2. Technical Implementation Likewise there would be numerous options to consider in terms of technical implementation. Some of the uncertainties would relate to system design decisions, while others might be more dependent on cost considerations. Issues to be resolved would include: • Metering. Relevant questions include which technical configuration(s) should be employed to meter mileage, potentially including the time and location of travel, and whether drivers should have the opportunity to choose among different metering devices. • Collection. As with metering, decisions would need to be made about which mechanism(s) would be used to collect mileage fees and whether drivers would have the option of choosing among alternate payment channels. • Preventing evasion. There is still considerable uncertainty about which methods for preventing evasion would provide the best tradeoff between the cost and effectiveness of enforcement. • Privacy. As with metering and collection, questions include which specific options would be employed to protect privacy and whether drivers could choose among multiple options.

62 • Data security. The design of a VMT-fee system should be secure—that is, it should not be possible for malicious third parties to gain access to potentially sensitive data. Nor should it be possible to gain remote access to vehicle controls (e.g. braking or acceleration) via the on- board device (see Koscher et al. 2010). Data encryption and firewalls would obviously play an integral role in helping to secure the system, but specific details of the security structure would need to be resolved. • Interoperability standards. If the goal of interoperability were pursued, the specific content and form of the standards would need to be determined. 5.1.3. Public and Private Institutional Arrangements Issues surrounding the institutional structuring of VMT fees would face perhaps the greatest degree of uncertainty. Crucial decisions to address would include: • Provision of metering technology. An important question to consider is whether a single firm should be engaged to provide the metering technology or multiple firms should instead be allowed to compete. Both approaches have been employed in prior road-pricing programs. • Provision of collection services. Another issue to resolve would be the institutional roles in collecting and apportioning mileage fee revenue and managing accounts. Fees might be collected by the government, by a joint powers authority, by a not-for-profit entity, by a single firm or consortium, or by multiple firms in competition with one another. If fees were collected by the private sector, it would also be necessary to determine how to structure the interaction with federal, state, and local revenue agencies. • Default public metering/payment option. If multiple firms were allowed to compete for the provision of metering equipment and billing services, likely on the basis of cost and value- added services, another question to consider is whether the government should also provide a default metering and payment option—either as a minimum cost option for drivers not interested in additional in-vehicle functionality beyond the payment of VMT fees or as a hedge in case the private sector failed to deliver acceptable options. • Enforcement of federal VMT fees. The enforcement function would fall in the public domain, requiring the intervention of sworn law officers and the courts to prosecute those who sought to evade fees. For VMT fees implemented at the state level, state and local agencies would likely be engaged to help enforce the system. It is less, clear, however, how the enforcement of federal VMT fees would be structured, particularly in a state that had not chosen to levy its own VMT fees. • Vendor certification. Under the assumption that multiple firms competed for the provision of metering equipment and billing services, there would need to be a process for certifying that their equipment and services meet minimum quality standards. It is unclear who should be involved in managing the certifications process, or what form the process should take. • Development of interoperability standards. Should the decision be made to develop interoperability standards, it would be necessary to specify which organizations should be involved in managing or participating in the development and how the process should be structured.

63 • Support and maintenance for the “price map.” In an advanced system of VMT fees, one could imagine the application of different per-mile rates in different jurisdictions, in different local areas, on different routes, and at different times of day. If the system also handled such functions as the automated payment of existing tolls or parking fees, additional rate data would need to be stored. The resulting database of price information could be massive and might need to be updated on a regular basis (e.g., each time a city changed its parking rates or a toll road operator updated the rate schedule). Questions to be addressed include what entity, new or existing, should manage the price map, and through what processes the price map should be updated? • Pricing guidelines. In the case of a federally implemented system that also supports state and local VMT fees, it is possible that a state or local jurisdiction could levy exorbitant per- mile fees on stretches of highway that primarily serve non-local traffic. The question, then, is what types of constraints should be imposed on the type or level of fees that could be levied by states or local jurisdictions. This issue has also arisen in the context of traditional tolls. 5.1.4. Additional Implementation and Transition Issues Several additional issues related to implementing and transitioning to a full-scale VMT-fee system would also need to be addressed: • Phase-in timeframe. In planning the transition from fuel taxes to VMT fees, an important question is whether the system should be deployed all at once or phased in over time. Given the likely cost and complexity of the system, a gradual phase-in appears more likely, for passenger vehicles at least. For commercial vehicles, one could envision a transition that occurs for all trucks on the same date; the European weight-distance truck tolls have been implemented in this fashion. • Mandatory vs. voluntary adoption. Assuming that the system would be phased in over a period of multiple years, another issue to determine is whether the initial adoption process would be mandatory (e.g., with the purchase of new vehicles) or optional (i.e., providing an initial period of voluntary adoption for several years before instituting mandatory adoption at a later date). A related question is whether the adoption of VMT fees might be mandated for some classes of vehicles (e.g., electric-powered cars not subject to fuel taxes) and optional for others (e.g., conventional gas-powered vehicles)? • Augmenting vs. replacing fuel taxes. Another important transition issue is whether the system would be intended to augment or to replace fuel taxes. If the latter, and if it were envisioned that the system would be phased in over time, then it would be necessary to determine some mechanism for rebating fuel taxes to early adopters. Depending on the technical configuration of the system (specifically, whether the pay-at-the-pump collection mechanism were implemented), rebating fuel taxes could represent a significant challenge. • Charging foreign vehicles. In the context of a national VMT-fee system, another question to resolve would be how to charge foreign vehicles (e.g. a vehicle visiting the United States from Canada or Mexico) for road use. If VMT fees were implemented at the state level, a similar question would arise with respect to charging drivers from other states.

64 5.1.5. System Cost A system of VMT fees would almost certainly cost more to implement and administer than fuel taxes. The question is, just how much more? Key cost issues to resolve would include: • In-vehicle metering equipment costs. Questions include how much it would cost to produce and install various metering configurations at scale, and how the cost might vary depending on whether the equipment were installed by the auto manufacturer or supplied by a third- party vendor. • Additional capital costs. Another uncertainty is how much it would cost to acquire and deploy any additional infrastructure required to support the VMT-fee system (e.g., electronic readers at fuel stations to support fee collection or integrated cameras and readers located throughout the road network to support enforcement activities) within a state or across the nation. • Collection costs. The cost of assessing fees and collecting payment under the alternate collection mechanisms would be a critical question to address. • Enforcement costs. A better understanding of how much would it cost to prevent evasion under the alternative approaches, and to what extent might this be offset by violation fines, would also be valuable. • Cost sharing. If the system encompassed additional paid services—for example, PAYD insurance, the payment of parking fees, and user fees for other value-added functions—it is possible that the cost to the federal or state government of equipping vehicles and collecting VMT fees could be reduced. It is not certain, however, whether such savings would in fact materialize or how significant the savings might be. 5.1.6. User Acceptance The final category of questions to address, and likely the most crucial in terms of political feasibility, would involve user acceptance. Unresolved issues in this vein would include: • Support for VMT fees. At present there is little public support for transitioning to VMT fees, in large part because not many citizens understand the shortfalls of continued reliance on fuel taxes, or the potential benefits of VMT fees. An important question is whether it would be possible to garner broader support through concerted education and outreach efforts. Related to this question is the observation from many prior road pricing programs that, while voters are often initially reluctant to change, support tends to increase once a program has been implemented and its benefits become more tangible. • Support for alternate pricing structures. Another uncertainty is how the application of alternate fee structures—for example, reducing the fees for lower emissions vehicles or charging more to travel in congested areas during peak periods—might either enhance or undermine support for a system of VMT fees, taking into consideration relevant equity concerns. • Privacy concerns. Though there are several effective means for protecting privacy, many in the public continue to view VMT fees, and in particular the use of GPS-based metering

65 equipment, as a significant threat to their privacy. It remains to be determined whether such concerns could be overcome through increased education and understanding of the system. • Value-added features. Many proponents have argued that drivers would be more supportive of VMT fees if the required metering equipment could also be leveraged to provide a range of value-added features. It is unclear, however, whether this argument holds merit, or which among the many possible value-added features would be of greatest interest to drivers. 5.2. ROLE OF TRIALS IN RESOLVING ISSUES AND UNCERTAINTIES As discussed earlier, the preceding NCHRP 20-24(69) report outlined five sets of activities that could help resolve remaining uncertainties during the next authorization to inform policy debate and assist in planning and preparing for a potential transition to VMT fees—planning and policy guidance, analytic studies, technical R&D, expanded system trials, and education and outreach. The intent in this study is to focus mainly on just one of these areas: system trials. The next step in the methodology thus involves considering which of the issues outlined above might be resolved or at least partially illuminated through system trials. Table 5.1 summarizes the judgments of the authors—informed by prior trials, studies, and discussions with other experts—regarding the activities that could prove helpful in resolving remaining questions and uncertainties. For any given issue (shown in the left-most column), hollow dots indicate the activities (shown in columns two through six) likely to prove helpful or essential in resolving the question; black dots indicate cases in which it would be valuable to receive planning and policy guidance in advance of the trials (e.g., specifying a set of metering capabilities to be supported within the system would help guide the selection of metering mechanisms to examine in the trials).

66 Table 5.1. Resolving System Design and Implementation Uncertainties Category / Issue Means of Exploring or Resolving Planning & Policy Analytic Studies Technical R&D System Trials Education & Outreach System Requirements Metering capabilities ● ○ ○ Vehicle classes ● ○ Geographic scope ● ○ Accuracy ● ○ Privacy ● ○ ○ Payment ● ○ Interoperability ● ○ Complementary functions ● ○ Technical Implementation Metering ○ ○ ○ ○ ○ Collection ○ ○ ○ ○ ○ Enforcement ○ ○ ○ ○ ○ Privacy ○ ○ ○ ○ ○ Data security ○ ○ Interoperability standards ● ○ Institutional Arrangements Provision of metering tech. ○ ○ ○ Provision of billing services ○ ○ ○ Default public option ○ ○ ○ Federal fee enforcement ○ ○ ○ Vendor certification ○ ○ ○ Standards development ○ ○ ○ Price map maintenance ○ ○ ○ Pricing guidelines ○ Implementation and Transition Phase-in timeframe ○ ○ Mandatory vs. opt-in ● ○ Augment vs. replace fuel taxes ● ○ ○ ○ Charging foreign vehicles ○ ○ System Cost Metering equipment ○ ○ Supporting infrastructure ○ ○ Collection ○ ○ Enforcement ○ ○ Cost sharing ○ ○ User Acceptance VMT fees ○ ○ Alternate fee structures ○ ○ Privacy concerns ○ ○ Value-added features ○ ○

67 There are at least two important observations to make from the entries in Table 5.1. First, there are circles in the “system trials” column for almost all of the rows. This suggests that the trials, if intelligently structured, could prove enormously helpful in providing the necessary information and experience to inform policy debate and prepare for possible implementation. Second, there are also circles in the “planning and policy” column for many of the rows. This is because relatively few of the issues to be addressed would have unambiguously correct answers. In most cases there would be multiple possibilities, each with its relative strengths and limitations. In this context, trials could help clarify the tradeoffs, but the resolution would ultimately rely on the deliberation and judgment of policymakers and planners. 5.3. GATHERING PERSPECTIVES ON SCOPING AND STRUCTURING TRIALS With these two observations in mind, the next question is how to design a set of system trials to be as productive as possible. Our approach, as described earlier, was to outline a detailed set of relevant questions, solicit the ideas and perspectives of stakeholders and subject matter experts through a series of interviews, present preliminary findings and further explore the issues in a subsequent workshop, and then synthesize and interpret the findings. 5.3.1. Interview Questions The questions posed during the interviews were intended to probe two related issues. First, of the many issues and uncertainties that system trials might be designed to address, which would be the most crucial? Second, how might the trials be scoped, funded, overseen, managed, and conducted to gain sufficient insight for the most crucial questions? Appendix B provides the interview guide that was distributed to interview participants, including a brief introduction to the project along with a lengthy set of detailed questions. The content of the questions can be organized and summarized as follows: • Broad policy considerations. What potential policy goals should be considered in the trials? Which road user groups should be included? • Scale, geographic coverage, and duration. How large should the trials be, and how long should they last? How many states should participate in the trials? • Specific goals of the trials. What would be the most important technical, institutional, transitional, and user acceptance issues to examine in the trials? • Leadership roles and stakeholder participation. Who should organize and oversee the trial program, and how should stakeholder input into the design of trials be incorporated? • Organization and management of the trials. Should the trials be centrally organized and carefully coordinated to achieve particular outcomes, or should there be multiple independent trials to examine a broader range of potentially fruitful variations? Who should be eligible to conduct individual trials? • Funding allocation. Through what program (new or existing) should the trials be funded? Should the federal government fully fund the trials or look for matching funds? On what basis should funding decisions be made?

68 • State and local involvement. What would be the appropriate role for various state and local agencies in conducting the trials? Under what circumstances would enabling state legislation be required to participate in the trials? • Private sector involvement. How could the trials be structured to encourage multiple vendors to participate in the development and provision of competing metering and collection options? Would there be a role for auto insurers (pay-as-you-drive insurance), for auto manufacturers, or for other data-related service providers? • Trial participants. How might individual drivers be enticed to participate in the trials? Should particular user groups be targeted for inclusion in the trials? What educational and outreach activities should accompany participation in the trials? • Cost estimates. Based on prior VMT-fee system trials that have already been conducted or are underway, is it possible to estimate, at least roughly, the amount of funding that would be needed to support the trials? • Detailed structure. What standards (e.g., for communications protocols) might be developed to facilitate the participation of multiple vendors? How might the trials address such issues as the accuracy of metering equipment and billing statements, the effectiveness of enforcement mechanisms and auditing, issues that might arise when phasing in a new VMT-fee system, and options for collecting road use charges from foreign visitors traveling in their own vehicles? • Other. Would it be valuable to examine any additional issues in the trials? 5.3.2. Participation in the Interviews and Workshop Legislating, implementing, and operating a system of VMT fees would require the participation and support of elected officials and agencies at various levels of governance along with the private sector. Additionally, many stakeholder and advocacy groups would have a strong interest in the design and structuring of such a system. It was therefore important to include a broad range of entities and actors within the interviews and workshop. Specific agencies or groups that we sought to involve include: • U.S. Congressional staff • U.S. Department of Transportation (U.S. DOT) • Federal Highway Administration (FHWA) • Research and Innovative Technology Administration (RITA) • U.S. Department of the Treasury (Treasury) • Internal Revenue Service (IRS) • State legislative staff and the National Conference of State Legislatures (NCSL) • State departments of transportation (DOTs) and the American Association of State Highway and Transportation Officials (AASHTO)

69 • State motor vehicle administrations (MVAs) / departments of motor vehicles (DMVs) and the American Association of Motor Vehicle Administrators (AAMVA) • Metropolitan planning organizations (MPOs) and the Association of Metropolitan Planning Organizations (AMPO) • Technology providers, system integrators, and tolling system operators • Automobile insurance companies • Stakeholders (e.g., road user groups, environmental advocates) • Subject matter experts and policy thought leaders • Managers of related road pricing trials and programs Perhaps the only notable group not included in the interviews and workshop was automobile manufacturers, which might someday either choose or be required to install metering equipment with new vehicles. Yet most of the individuals with whom we spoke during the project viewed it as unlikely that such firms would voluntarily participate in the trials, for two reasons. First, they would likely be reluctant to become associated with new and potentially controversial forms of fees or taxation. Second, the technology for metering mileage is evolving so rapidly that auto manufacturers would be inclined to wait some time before choosing to provide standardized equipment in their vehicles. As indicated above, the initial set of interview questions covered a broad set of considerations in great detail, and relatively few individuals would be expected to have the background to offer meaningful answers to all of the questions. Most of the interviews therefore centered on a smaller subset of questions that aligned most closely with the participants’ areas of expertise. (Note that in describing responses to the questions in subsequent chapters, we often make statements such as “a majority of participants…” or “many respondents…” More precisely, what we are describing refers to the subset of interview participants who responded to a particular question.) Additionally, the interviews were conducted in a loosely structured, free-flowing manner; when a participant offered a particularly intriguing or insightful answer to a given question, we would often follow-up with additional queries and discussion before proceeding to the next topic in the prepared list of interview questions. The vast majority of individuals that were asked to participate in the interviews and workshop agreed to do so, suggesting that interest in this subject is particularly high. In total we conducted 55 interviews involving 69 individuals (several of the interviews included more than one person from the same organization). The workshop included 32 participants in addition to the research team and TRB staff. Of these, 12 were affiliated with the project panel, and many of the remaining 20 had also participated in the interviews. Appendix C provides a comprehensive list of the interview and workshop participants. The interviews and workshop proved extremely helpful in developing and refining thoughtful suggestions on how to scope and structure effective system trials. The next chapter summarizes high level observations from the interviews and workshop and interprets their implications, while the subsequent chapter presents the results for more detailed questions.