The Role of Transit, Shared Modes, and Public Policy in the New Mobility Landscape (2021)

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.

89 In this chapter, the committee develops the mobility management frame- work for public transit agencies and other local and regional authorities proposed in Chapter 3. These entities could create and use the framework to capitalize on this new mobility landscape to improve the coverage, avail- ability, and reliability of personal transportation services while furthering societal goals such as equity and sustainability. The proposed framework identifies (1) the key capabilities needed to bring about this outcome, (2) strategies for obtaining these capabilities, and (3) the parties respon- sible for pursuing the strategies. Following a description of this envisioned framework, consideration is given to two specific capabilities needed by the customer—a ready way to know about all available and relevant service offerings and their attributes and a ready way to pay for them. While other capabilities are needed to provide consumers with quality multi-modal options, as discussed in Chapter 5, a focus on the needs of the traveler must be at the core of any mobility management initiative. MOBILITY MANAGEMENT FRAMEWORK A customer-centered, multi-modal system, with an emphasis on transit and shared mobility, would be built around providing travelers with choices about the modal attributes they value, such as convenience, versatility, com- fort, price, and travel time—while potentially also furthering social goals such as reduced emissions, safety, equity, and efficiency. A framework to support mobility management would identify needed capabilities, lay out strategies to create, or further, these capabilities, and indicate the entities 4 Mobility Management, Part 1: A Focus on the Traveler

90 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY responsible for implementing the strategies in fragmented metropolitan areas. Ideally, a well-managed system would ensure that individuals have mul- tiple travel options available to them that are convenient and appropriate for where and when they want to travel throughout a region. As described in Chapter 1, single-occupancy vehicle (SOV) use dominates most travel in the United States. However, without information about the availability and cost of other options (both individual and societal) and how to arrange and pay for them, many travelers are uncertain of options beyond their personal automobile and unaware of those that are competitive with SOV trips. A complete mobility management framework for achieving a customer- centered, multi-modal system requires the following capabilities: A. Easily accessible information for all customers about options for origin-to-destination (O-D) trips including modes, routes, travel time, cost, and emissions; B. Convenient and easily integrated payment for O-D trips by all customers; C. Provision of feasible and accessible multi-modal options to customers; D. Quality transit service and seamless transfers across modes; and E. Supportive development (land use) to enhance the appeal of transit and reduce trip distances. Each of these five goals can be broken down into capabilities that could be realized through a variety of implementation strategies, or “policy levers.” This chapter, which focuses on providing information to help con- sumers choose among multi-modal options, describes the first two of these capabilities and the means of achieving them. The next chapter focuses on the public role in supplying multi-modal options and shaping land use to enhance the appeal and cost-effectiveness of transit. Each section of both chapters includes a table that provides an outline for, and overview of, its content. The first column of each table identifies the capabilities being sought, the second column lists implementation strategies to achieve those capabilities, and the third column identifies the entities with responsibility for elements of the strategies. The text describes (1) the various policy and programmatic strategies that local jurisdictions can employ; (2) evidence of effectiveness when available; (3) the policy issues raised; and (4) the variety of agencies and jurisdictions that are, or would need to be, involved in implementing them.

A FOCUS ON THE TRAVELER 91 A. EASY, OPEN ACCESS TO O-D TRIP INFORMATION The first step toward a customer-centered, multi-modal system is giving individuals complete and readily accessible information about their options for making trips—modes, routes, time duration, and costs (both individual and social)—and allowing them to choose how to complete their journeys. Individual costs are those paid directly by the traveler and are standard considerations in travel choices. Social costs are those paid by society at large, and thus only indirectly by the traveler; not all travelers will con- sider these costs as highly as those they pay directly, but the provision of information about social costs such as greenhouse gas emissions may subtly encourage the traveler to consider the societal implications of their individual actions.1 In the aggregate, those individuals choosing the option with lower societal costs may have a noticeable impact on the optimization of the transportation system. As noted, the provision of information is but a first step in affecting travel choices; also important are public policies that enhance the availability and performance of shared modes and transit and charge drivers for the social costs of road use and parking, as described in the next chapter. A generation ago, navigating around metropolitan areas involved use of paper roadway maps and, in unfamiliar areas, consultations with friends and colleagues about the best routes, the presence of tolls, and where park- ing would be available and its cost, if any. Navigation in unfamiliar center cities could be simplified by hailing a taxi, whose driver relied on his or her mental map of the road network. In the case of transit, trip planning involved consulting poster maps of transit networks, pamphlets with route schedules, transit ticketing agents, and, again, consultation with friends and colleagues. For work trips made by transit riders with other options, the time costs associated with overcoming information barriers and solv- ing navigational problems could be easily justified by the necessity and regularity of such trips and subsequently be incrementally optimized with new information. For auto-oriented travelers making a non-routine trip, the time costs of overcoming knowledge about transit options could (and still can) be significant. The Internet, GPS, online routing algorithms, trip planners, real-time traffic information, and other applications available on smartphones and websites have greatly changed and simplified this process for auto trips (both by personal vehicle and ridehailing), micromobility, and microtransit. Electronic fare cards have made passenger flows and transfers on transit more efficient, and digital trip planners provide information about alternate 1 Hensher, D., et al. 2020. Understanding Mobility as a Service: Past, Present, and Future. Elsevier, pp. 41–47. https://www.elsevier.com/books/understanding-mobility-as-a-service-maas/ hensher/978-0-12-820044-5.

92 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY routes and transit modes. We are on the cusp, however, of having applica- tions that give metro-area residents complete information that simplifies choices and tradeoffs among routes, modes, and means of payment (dis- cussed in the next major section of this chapter). Problems and gaps remain; solving or overcoming them at the regional scale involves institutional issues more than technical ones. Providing additional information would help many travelers and influ- ence their travel choices. The most obvious information is the set of travel options available to go from origin to destination. However, to make the best choice among those options, a traveler needs more information than is currently available. The time for each option or set of options is among the most critical items for the traveler to know, as is the total cost of the trip. The traveler should have a reliable estimate of the total time required for each option that includes time to access the option, including getting to a car, walking to a transit stop and waiting for the bus or train, walking to pick up a micromobility option, waiting for a taxi or ridehailing vehicle, or any other “startup” time. It also includes time to access the destination on the other end of the trip, including walking from the parking, transit stop, or micromobility dock. Estimating durations of trips requires hav- ing data about real-time performance of each part of the transportation system, from roadway congestion levels to transit’s on-time performance. Cost would include expected parking, tolls, fares, fees, and other costs asso ciated with making an individual trip. Having improved access to this type of reliable information about time and cost of a trip may make some travelers more willing to alter their historical travel patterns. Some trip planning apps already provide some of the information. Multiple examples, including Google Maps and Apple Maps, provide in- formation at a local level, providing information about trip directions and duration by car, by foot, by bicycle, or by transit. However, the ability to combine multiple modes into one trip is not always clear and the informa- tion is not always in real time but instead is based on schedules. Other options for longer-distance travel include sites like Rome2Rio.2 Again, the information these sites provide generally includes the modal options and sometimes overall cost comparisons, but not detailed schedules, costs, or access time. They represent a first stage of mobility as a service (MaaS) evolution, providing a multi-modal route planner and price comparison service, but not including booking and payment capabilities.3 The current 2 See https://www.rome2rio.com. 3 Araghi, Y., et al. 2020. A Conceptual Framework for Drivers and Barriers of Mobility- as-a-Service. Proceedings of the 9th Transport Research Arena, April 27–30. https://www.isi. fraunhofer.de/content/dam/isi/dokumente/ccn/2020/tra2020_drivers_and_barriers_final-after- reviews-19-10-2019.pdf.

A FOCUS ON THE TRAVELER 93 options provide a model of how information provision might be handled, but true mobility management requires broader coverage of all modes and more complete and current information. A region can set policies that are likely to lead to greater information availability about modal options in the following ways: • Using and requiring standard data formats for information about all transportation services; • Requiring data from shared and micromobility providers during the permitting process; • Prioritizing acquisition of real-time data on the availability and status of all transportation modes, from shared mobility to local roads; • Developing requirements for mobility aggregation apps, including prediction of trip times, emissions, and any other factors that may influence the traveler; • Determining the most appropriate public–private split of responsi- bilities in developing a regional mobility management system; and • Establishing policies and procedures for protection of personally identifying information (PII). The sections that follow, summarized in Table 4-1, illustrate the capa- bilities, their implementation strategies, and the entities responsible for the legal and operational steps. A.1. Open Access to Real-Time Travel Information For a traveler to easily understand the full range of options, he or she would need readily available and comparable real-time information about their travel options. Achieving this goal depends on standardized data, re- quirements that private providers share data about the availability of their services, provision of system performance by transit agencies and road de- partments, and aggregation of data flows into applications that display the information for consumers. MaaS systems can provide such information in a comprehensive and unified fashion. Policies are needed to protect privacy. The following sections provide background for each of these elements and identify issues not yet resolved. Data Standardization Apps can aggregate and share information if the basic elements of the information are provided in standardized formats as described below, and if the jurisdictions that allow mobility providers to operate within their

94 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY TABLE 4-1 Easily Accessible Information About O-D Trip Options Capabilities Implementation Strategies Responsible Entities 1. Open access to real- time travel information at a sustainable cost Data standardization requirements Cities and other mobility service permitting agencies Data on private shared mode availability through permitting requirements and contracts City, county, metropolitan planning organization (MPO), or state agencies, as appropriate in individual regions Transmitting and sharing real-time vehicle location and roadway conditions Transit agencies, roadway operators, shared mobility providers Requirements for apps including prediction of trip times, emissions, etc., with models/algorithms Large cities, transit agencies, and possibly MPOs set these requirements currently Efficient, cost-effective models for MaaS Agencies have not yet begun addressing these issues in the United States but are in many nations, particularly in Europe Protection of personally identifying information (PII) States (open records laws), local agencies (for policies and strategies to protect PII), private providers, third parties (for processing and transmitting info between public and private entities stripped of PII) 2. Equitable information options for those without computers or smartphones Provision of travel data through phone, Internet, and other information resources City/county/MPO/state organizations boundaries require such data standardization for real-time information. Standardization is largely the case today for static information about transit routes, schedules, and modes and for trip information by shared mobility providers, but real-time performance of elements of the system is not as commonly provided. The following transportation data standards are in relatively common use at the time of this report. General transit feed specification For transit data, most agencies use the General Transit Feed Specification (GTFS) as a template for the structure of route and schedule data. By publishing their data in this form, transit agen- cies make them available to a wide array of software applications. GTFS data standards, which were developed through a collaboration between

A FOCUS ON THE TRAVELER 95 Google and Portland’s TriMet, are twofold: one portion of the data is static information such as route alignments, stop locations, and schedule data. A second component is real-time information about vehicle locations and arrival prediction times.4 GTFS was launched in 2005 in the city of Portland, Oregon, and has since expanded around the world, used by thou- sands of transit agencies and third-party apps providing both schedule and real-time travel information.5,6 Use of this standard is generally voluntary; agencies use it because it allows for easy integration of that agency’s data into existing third-party mapping and scheduling systems for customers.7 General bikeshare feed specification The North American Bikeshare Asso- ciation developed the General Bikeshare Feed Specification (GBFS) in 2015. It provides real-time information about docked and dockless bike and scooter systems, their docks, and the vehicles that are available, includ- ing battery charge levels for e-bikes and e-scooters. It does not, however, provide real-time information about vehicles that are in use. GBFS has undergone updates to match the features of newly available vehicles, such as the battery charge levels. Mobility data specification The Los Angeles Department of Transporta- tion developed the Mobility Data Specification (MDS) in 2018 and began requiring permitted micromobility operators within the city to provide information in this standard form. This specification is intended to stan- dardize the data passed between cities/jurisdictions and mobility operators so that the city could actively manage the public right of way. In 2019, the Open Mobility Foundation was established and took over management of this specification. MDS is primarily used for shared bikes and scooters as of August 2020.8 Unlike GBFS, it provides trip-level data for the vehicles, and cities generally require these data from mobility providers to be real time to within a matter of minutes.9 Some mobility providers and privacy advocates have questioned the legality of cities requiring such data, citing 4 See https://gtfs.org. 5 McHugh, B. 2013. Open Data and the Future of Civic Innovation. In Beyond Transpar­ ency. https://beyondtransparency.org/chapters/part-2/pioneering-open-data-standards-the-gtfs- story. 6 Zack, R. 2019. A Data Standard for New Mobility. ITE Journal, June 26–28. 7 McHugh. 2013. 8 Populus. 2020. A Practical Guide to Mobility Data Sharing and Cities. https://www. populus.ai/white-papers/mobility-data. 9 Alpert, D. 2020. Why a Battle Between Tech Visionaries, Privacy Advocates, Uber, and Transportation Officials Is About Much More Than Scooter Data. Greater Greater Washington, May 20. https://ggwash.org/view/77285/mobility-data-standard-scooters-bikes- autonomous-vehicles-uber-lyft-ddot-los-angeles.

96 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY the California Electronic Communications Privacy Act of 2015, and these issues remain unsettled as of the time of this report.10,11 In 2017, the National Association of City Transportation Officials (NACTO) organization suggested data-sharing principles for cities to use when integrating new mobility options.12 The recommended data for cities to require and collect are in three categories: data for transportation plan- ning (including speeds, pick-up and drop-off locations and times, vehicle occupancy, etc.), data for equity (unfulfilled ride requests, vehicle avail- ability by type, etc.), and data for safety (collision locations and severity, etc.). In 2019, in coordination with the International Municipal Lawyers Association, NACTO released another set of guidance on mobility data.13 This document lays out the challenges that both cities and private-sector players face in managing and sharing mobility data and suggests a series of best practices for doing so, along with best practices for data handling. Cities and other regional jurisdictions may look to these documents when developing their data standards for permitting and sharing and participate in future standards development efforts. These cities and jurisdictions can make their own choices about what data to require from mobility pro- viders and in what format, if any, to collect those data. However, by using the permitting process to require that mobility companies provide data in certain standard formats, they can pave the way for a set of standardized information that would form the basis for a MaaS app. Data for traffic flow are widely available currently and mapping applica- tions from Google Maps to Apple Maps to in-vehicle GPS systems all readily provide it. Some of these data come from sensors along roadways that road- way operators install at strategic fixed locations. Other data sources include the smartphones and in-vehicle GPS systems themselves as they move through- out the system. Both the fixed and mobile sources provide a vast amount of real-time data about the performance of the road network, allowing travelers to choose an optimal route and alerting transit agencies about potential issues for their services (primarily buses) that also use the road network. As described in the next section, the data that some cities are requir- ing from mobility providers are not necessarily the same as those required for mobility management. It may be important for a city to have data on 10 Hawkins, A.J. 2020a. The ACLU Is Suing Los Angeles Over Its Controversial Scooter Tracking System. The Verge, June 8. https://www.theverge.com/2020/6/8/21284490/aclu- ladot-mds-lawsuit-scooter-tracking-uber. 11 Alpert. 2020. 12 NACTO. 2017. City Data Sharing Principles: Integrating New Technologies into City Streets. https://nacto.org/wp-content/uploads/2017/01/NACTO-Policy-Data-Sharing- Principles.pdf. 13 NACTO. 2019. Managing Mobility Data. https://nacto.org/wp-content/uploads/2019/05/ NACTO_IMLA_Managing-Mobility-Data.pdf.

A FOCUS ON THE TRAVELER 97 vehicle trips to understand and predict traffic flows and to enforce regula- tions about vehicle use and public space. However, the type of mobility integration this report focuses on would require only information about where available vehicles are at the moment. As described in subsequent sections, some cities have required types of data from mobility providers that the mobility providers consider to be too broad and to be invasions of user privacy. Obtaining and providing only information about the locations of currently available vehicles can avoid the challenges that some cities and other local jurisdictions are facing in obtaining data about trips. These challenges are both legal and logistical, as cities attempt to manage the very large quantities of data being generated. Data Availability Private entities have begun providing real-time information in a variety of ways, such as collecting smartphone probe data of traffic flows, intel- ligent transportation system information about parking availability, and crowd-sourced information about sources and causes of congestion. Shared mobility providers have real-time information about the location, position, and availability of their vehicles and devices and, other than ridehailing companies, are usually required by cities to provide key data as a condition of permits to operate. Transit agencies generally have real-time information about how their rail networks are performing and, to a degree, how bus systems are performing.14 Agency contracts with private entities provide means of gathering information. The key issues in integrating such informa- tion for individuals’ use are the steps required to make it accessible through policies and having mechanisms for doing so. In recent years, public agencies seeking data from shared mobility providers have faced varying degrees of pushback. For example, in August 2019, Uber sued the City of Los Angeles over the city’s use of the MDS data specification, which applied to (then) Uber-owned Jump scooters and bikes. The company asserted that the city’s collection of the data would lead to an “unprecedented level of surveillance”15 as elaborated on further in the section below on protecting PII. In response to the company’s refusal to share the data, the city pulled Jump’s permit for scooters and bikes.16 Uber 14 Brakewood, C., and K. Watkins. 2018. A Literature Review of the Passenger Benefits of Real-Time Transit Information. Transport Reviews 39(3):327–356. 15 Hawkins, A.J. 2019. Uber Threatens to Sue Los Angeles, as the Fight Over Scooter Data Escalates. The Verge, October 29. https://www.theverge.com/2019/10/29/20938212/ uber-lawsuit-la-ladot-scooter-data-mds. 16 Nelson, L.J. 2019. L.A. Suspends Uber’s Permit to Rent Out Electric Scooter and Bikes. Los Angeles Times, November 4. https://www.latimes.com/california/story/2019-11-04/los- angeles-suspends-uber-jump-scooters-bikes-data-privacy.

98 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY and Lyft both have a contentious history with cities seeking various types of travel data,17 although both companies have taken steps toward increased partnership in recent years.18,19 Despite these increased partnerships, it is still not clear whether cities or app aggregators would have access to the real-time information about ridehailing availability that is needed for mobility management, or whether travelers would only be able to access that app in each ridehailing company’s own app. This area is contested and currently evolving. Real­Time System Information Transit passengers benefit from having real-time information about their transit options.20,21 These benefits include both real and perceived decreased wait times, reductions in overall travel time, increased satisfaction with the transit service, and an increased sense of personal security when waiting for and riding transit. All these benefits increase use of transit.22,23 Automatic vehicle location sensors, mounted in or on a transit vehicle, generally provide this information. The sensors transmit the vehicle’s loca- tion to a central database every few seconds, allowing the agency and a rider to monitor the location of any given vehicle. Not every transit agency has equipped all of their vehicles with such sensors, however, and even when the sensors are in place, they are not always transmitting accurate information, or at all. Some apps also use crowd-sourced data from third- party apps to track vehicle locations; in these cases, riders on the bus who have an app open or running in the background can easily (and sometimes unknowingly) transmit the vehicle’s current location to a central database.24 17 Zipper, D. 2019. Walled Gardens vs. Open Mobility: The Battle Begins. Bloomberg CityLab, October 8. https://www.bloomberg.com/news/articles/2019-10-08/what-s-behind- lyft-s-battle-over-bikeshare-access. 18 McGregor, J. 2017. Uber Launches New “Movement” Tool. Forbes, August 31. https:// www.forbes.com/sites/jaymcgregor/2017/08/31/uber-launches-new-movement-tool. 19 Edelstein, S. 2018. Ford, Uber, Lyft Join Urban Data-Sharing Project to Reduce Traffic and Pollution. The Drive, September 27. https://www.thedrive.com/tech/23874/ford-uber-lyft- join-urban-data-sharing-project-to-reduce-traffic-and-pollution. 20 Gooze, A., K. Watkins, and A. Borning. 2013. Benefits of Real-Time Transit Informa- tion and Impacts of Data Accuracy on Rider Experience. Transportation Research Record 2351(1):95–103. 21 Brakewood and Watkins. 2018. 22 Gooze, Watkins, and Borning. 2013. 23 Brakewood, C., G.S. Macfarlane, and K. Watkins. 2015. The Impact of Real-Time In- formation on Bus Ridership in New York City. Transportation Research Part C: Emerging Technologies 53:59–75. https://doi.org/10.1016/j.trc.2015.01.021. 24 Transit App. 2018. Transit Adds Crowdsourced Real­Time in 175 Cities. https://medium. com/transit-app/transit-adds-crowdsourced-real-time-in-175-cities-a90ec97685ec.

A FOCUS ON THE TRAVELER 99 For a traveler to have the greatest possible level of confidence in transit service, he or she would need to trust that vehicles will arrive as expected. The reality of vehicle arrival times is often not well represented by sched- ule data. However, not all mobility providers have real-time data available for all vehicles. A possible workaround for lack of real-time data could be models that predict a vehicle’s arrival time based on a combination of schedule, current road performance levels, and historical punctuality levels. A variety of model types exist; simple models use only historical data or, for those that are slightly more advanced, average travel times or speeds. More complex statistical models incorporate time-series, regression, or machine learning algorithms to improve their predictions of vehicle arrival times.25 Nonetheless, accurate real-time information is the best option. (Policies and strategies to improve road system performance and schedule reliability of transit are discussed in the next chapter.) Real-time information is also important for modes other than transit. Information about the current state of roadways and parking options is also highly relevant for travelers who have a personal vehicle as one of their travel options. Light traffic and parking that is widely available and inexpensive or free make a traveler more likely to drive,26,27 especially if they know that the information is current and accurate, whereas heavy congestion and limited or expensive parking will encourage travelers to consider other options that may be faster, cheaper, and less stressful. As discussed above in the data standards section, this information is widely available today through navigation apps that are both stand-alone and built into vehicles, and through an ever-evolving array of parking availability apps and websites. Providing this information alongside equivalent real- time information about transit, shared mobility, and any other available travel options would make a traveler’s options clear and comprehensive and has been shown to have a measurable, if generally small, effect on traveler behavior.28,29 25 Altinkaya, M., and M. Zontul. 2013. Urban Bus Arrival Time Prediction: A Review of Computational Models. International Journal of Recent Technology and Engineering 2(4):164–169. 26 Christiansen, P., et al. 2017. Parking Facilities and the Built Environment: Impacts on Travel Behavior. Transportation Research Part A: Policy and Practice 95:198–206. 27 McCahill, C., et al. 2016. Effects of Parking Provision on Automobile Use in Cities: Infer- ring Causality. Transportation Research Record 2543(1):159–165. 28 Pronello, C., J.P.R.V. Simao, and V. Rappazzo. 2017. The Effects of the Multimodal Real Time Information Systems on the Travel Behavior. Transportation Research Procedia 25:2677–2689. https://doi.org/10.1016/j.trpro.2017.05.172. 29 Khoo, H.L., and K.S. Asitha. 2016. User Requirements and Route Choice Response to Smart Phone Traffic Applications. Travel Behavior and Society 3:59–70. https://doi. org/10.1016/j.tbs.2015.08.004.

100 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY Real-time information such as this would require transit agencies, cities, roadway operators, and regional travel authorities to prioritize the acquisition of real-time data. Transit agencies would need to purchase, install, and oper- ate automatic vehicle location sensors. Cities and roadway operators would need to acquire, install, and maintain sensors that provide information about roadway operating conditions, parking availability, and more. In addition or alternatively, they could purchase such information from third parties. While many transit agencies, cities, and roadway operators have already established such sensor networks and are collecting the data, others are not and would need to prioritize this real-time data collection to enable mobility manage- ment. This, of course, requires additional funding, as discussed earlier. App Requirements Each travel aggregation app would need to provide a basic set of informa- tion to its users. Interfaces can vary widely, but a certain set of information would be needed for each alternative that a traveler might use. Policy makers could decide what set of information to include, such as estimated travel time, travel cost in dollars, carbon emissions generated, or other factors. Having a standard set of metrics on which a user could sort and compare their travel options helps to make the choice a more informed one. Cities, or possibly metropolitan planning organizations (MPOs), could establish requirements for these basic data, leaving app developers to design the rest of the interface as they wish. However, a “proliferation of mobility apps” required to understand one’s travel options may prove to be as burdensome for travelers as was the difficulty in acquiring information pre-smartphone.30 “App fatigue” is when many smartphone owners feel that they are overwhelmed by the number of apps required to accomplish a task.31,32 In the case of travel, this could include a few apps for direction and mapping, multiple transit agency apps for real-time information and/or payment, half a dozen scooter apps, a bikeshare app or two, ridehailing apps, and more. There is a growing inter- est among consumers in coordination among transportation services so that fewer apps can serve their needs. Unified, comprehensive MaaS platforms are designed to achieve this goal. 30 Hill, A. 2020. Most Europeans “Want Single MaaS App,” Says FIA. ITS International, June 4. https://www.itsinternational.com/its17/news/most-europeans-want-single-maas-app- says-fia. 31 Darrow, B. 2017. App Fatigue on the Job Is Becoming a Problem, Survey Says. Fortune, September 14. https://fortune.com/2017/09/14/business-pros-app-fatique. 32 Dillet, R. 2020. iOS 14 Gets Rid of the App Grid to Help You Find the App You’re Looking For. TechCrunch, July 9. https://techcrunch.com/2020/07/09/ios-14-gets-rid-of-the- app-grid-to-help-you-find-the-app-youre-looking-for.

A FOCUS ON THE TRAVELER 101 Efficient and Cost­Effective Organizational Models for MaaS33 As envisioned by its proponents, MaaS would provide travelers with (1) ready access to all desired information about service options and their qualities, (2) easy-to-use applications and ticketing for combining modes and services for making individual trips, and (3) convenient ways to pay. Information and communications technology (ICT) is key to making these service information and payment objectives possible, but questions arise about what entities are best positioned to implement the technologies and provide the information in the most efficient manner while also further- ing societal goals. One or more entities would need to be engaged both at the front end to provide the consumer interface and at the back end to aggregate service information and influence the quality of the service (e.g., ensuring service convenience, coordination, and coverage). Both the public and private sectors could serve in these roles, and indeed they do in differ- ent MaaS systems but not always in the same way.34 Examples of MaaS platforms from Europe and elsewhere suggest how the front- and back-end roles can differ and how they can potentially affect both the economic efficiency of the service and its effectiveness in meeting the individual demands of travelers and societal interests. The platforms that currently exist or that are being developed consist of a variety of orga- nizational models for providing both ends of the operation.35 It is typical for private companies to provide the customer-facing front end by developing and providing the app used by the traveler to plan, book, and pay for a trip.36 Private companies who are not the mobility service providers themselves have long competed with one another in providing trip planning platforms in individual markets and across multiple markets, and their competition has spurred continual innovation in consumer-facing trip planning applications. But while competition for users can bring about more innovative and consumer-friendly front-end apps by private front- end operators, there is no assurance that the offerings displayed will fully inform the consumer about all possible travel options. This outcome may 33 For a more thorough review of the issues associated with the role of aggregating and offering services see Hensher et al. 2020. 34 As noted by Hensher et al. 2020, despite several trials with MaaS in Europe and a grow- ing body of research on a variety of issues associated with MaaS, little attention has been paid to date to MaaS governance and the respective roles of the public and private sectors. For a broader discussion of these issues and knowledge gaps, see Hensher et al. 2020. 35 APTA (American Public Transport Association). 2019. Being Mobility­as­a­Service (MaaS) Ready: APTA International Study Mission Final Report. https://www.apta.com/ wp- content/uploads/MaaS_European_Study_Mission-Final-Report_10-2019.pdf. 36 Zipper, D. 2020. The Problem with “Mobility as a Service.” Bloomberg CityLab, August 5. https://www.bloomberg.com/news/articles/2020-08-05/the-struggle-to-make-mobility-as-a- service-make-money.

102 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY depend on the intensity of competition and the individual interests of the private companies. For instance, in the case where a private mobility service provider operates the front-end app, the provider may favor its own ser- vices. As an example, micromobility providers have withdrawn from MaaS platforms that include their competitors, and ridehail companies have ac- quired micromobility and microtransit providers to create “walled gardens” where consumer-facing apps only show their own service offerings.37 Competition among private companies, whether mobility providers or not, may not lead to front-end functionality that furthers the broader public interest—for instance, by directing more travel to energy-efficient modes and services. Some MaaS platforms, such as HannoverMobil, which is managed by the Hannover regional transportation association and public transit operator, and S’Hail, which is managed by Dubai’s public trans- port operator, rely on a government agency or public transit operator to provide the front-end operation.38,39 The public sector serving in this capacity could seek to further such societal interests. Here too, however, the government’s interest may be to favor its publicly owned and operated mobility services.40 Moreover, because the public sector does not face the same pressures as competing private companies to be cost conscious and technologically current, a possible drawback of this model is that the apps and their features will not keep pace with customer demands. Perhaps more than in the front-end operation—where private com- panies dominate, bolstered by their long-standing roles in providing trip planning software—it may be more suitable for the public sector to have a prominent role in the back-end MaaS operation, both for practical reasons and to further societal interests. The back end is the series of data stan- dards, access, and processing required to determine and optimize a user’s tripmaking options. The back-end operator serves as a broker or aggrega- tor of mobility services, creating combinations of service offerings from different providers and distributing them among customer-facing MaaS operators. Hensher describes the core benefit of these brokers as being the integration of actors, information, and services, that is, the transportation service, data, and information technology providers.41 37 Zipper. 2019. 38 Al Tayer, M. 2019. Exploring Dubai’s Plans for the Future of Mobility. Intelligent Trans­ port, June 26. https://www.intelligenttransport.com/transport-articles/82445/dubai-mobility- rapid-ncrease-use. 39 Röhrleef, M. 2017. Hanover’s “One Stop Mobility Shop.” Intelligent Transport, Oc- tober 17. https://www.intelligenttransport.com/transport-articles/69062/hanovers-one-stop- mobility-shop. 40 Hensher et al. 2020. 41 Hensher et al. 2020.

A FOCUS ON THE TRAVELER 103 One can find examples of MaaS back-end functions that are provided by public and private entities—sometimes by entities that do not provide mobility services themselves, including local and regional government agen- cies, and sometimes by private or public mobility service providers such as a transportation network company or public transit operator. The public transit operator in Dubai provides both the front end and back end of its MaaS platform, as does the Hannover public transportation operator.42,43 The best-known existing MaaS platform with a back end operated by a private company that does not provide a mobility service is Whim, whose back end is provided by MaaS Global. As of June 2020, Whim was operat- ing or soon to launch in Antwerp, Helsinki, Singapore, Turku (Finland), Vienna, and the West Midlands (United Kingdom).44 An example of a MaaS platform in which the back-end operator is a private company that provides mobility services is Uber, which opened its platform to include public transit operators.45 There are potential advantages and disadvantages to these public- and private-sector back-end models. By operating in many markets, private companies such as MaaS Global and Uber can reap economies of scale and scope, and may be compelled to be more innovative in competing for the business of front-end operators and market share. In this model, competitive pressure to attract customers would be expected to encourage innovations and discipline costs, but there would be no assurance that each operator will ensure that the full range of available transportation options is distributed to the front-end operator and ultimately the consumer. The private company may not be able to obtain service information from, or act as a broker for, some mobility service providers who want to maintain their brand and their data, or the back-end operators may not be interested in doing so in cases such as Uber where the back-end provider also operates a competing mobility service.46 In theory, a public entity that provides the back-end functions may be able to compel all mobility service providers to provide the needed service information and to adjust their services in ways that benefit more con- sumers and serve societal interest. However, not unlike a private operator, the public sector may also have interest in promoting its own publicly run transportation operations, which may or may not be conducive to further- ing societal interests or the interests of individual consumers.47 Here again, based on the experience of public entities not being able to keep pace with 42 Al Tayer. 2019. 43 Röhrleef. 2017. 44 Whim. 2020. https://whimapp.com. 45 Hensher et al. 2020. 46 APTA. 2019. 47 Hensher et al. 2020.

104 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY changing ICT capabilities when providing front-end transportation plan- ning apps, the ability of public entities to do so for the back-end ICT needs can be questioned. Noting some of these dynamics, Hensher et al. posit that government is better positioned to play a regulatory function, rather than to provide the MaaS front- and back-end operations themselves.48 They maintain that by acting in this capacity, the government can influence the efficiency and effective ness of MaaS operators at the margin—for instance, by setting some common service standards and by specifying the conditions and bar- riers (or lack thereof) for market entry, which presumably could include requirements for the standardization of data and its supply to both back- end operators and front-end operators for delivery to the consumer. There are many combinations of options for providing MaaS operation on the front and back ends and there is a range of public and private enti- ties having different characteristics (e.g., mobility providers, non-mobility providers) and operating in different contexts. As a result, it is not possible without more empirical evidence to point to the most desirable models in terms of their efficiency and effectiveness in meeting consumer interest and furthering societal goals. Protection of PII An important consideration in the design of a MaaS system is how to pro- tect PII. Recent privacy policies include the General Data Protection Regu- lation, implemented in 2018 and addressing personal data protection in the European Union and the European Economic Area, and the California Consumer Privacy Act, which enhanced privacy rights and consumer pro- tections for residents of California and became effective on January 1, 2020. Because sharing of data is the underlying basis for a MaaS system, providers and users of the data must allow a traveler to have control over the tracking and potential “profiling” of their trips and of their financial and social network information.49,50 The types of data covered by these policies include geolocation and personally identifying data that are often found in trip records of shared mobility providers. While both policies are theoretically limited to protecting those in their geographic limits, the wide- spread international availability of many shared mobility providers means that the providers will be subject to the policies’ requirements, potentially leading to different requirements in different locations. 48 Hensher et al. 2020. 49 Costantini, F. 2017. MaaS and GDPR: An Overview. arXiv e­prints, arXiv:1711.02950. 50 Cottrill, C.D. 2020. MaaS Surveillance: Privacy Considerations in Mobility as a Service. Transportation Research Part A: Policy and Practice 131:50–57.

A FOCUS ON THE TRAVELER 105 Nonetheless, collection and aggregation of trip data by public agencies risks the release of these data through open records laws, which could be a significant restriction to accomplishing the goal of seamless integrated mobility. Workarounds for this issue exist, one of which would be for public agencies to only require and accept aggregated information from private providers that would avoid the risk of exposing PII and/or use of private third parties to accept this information. Third parties would pro- vide public agencies with the analysis cities want for enforcing permit re- quirements on private providers and to understand where traveler demand and transportation supply are out of balance, and could do so without sharing detailed personal trip information. For the MaaS systems described in this chapter, distribution and display of trip-level data is not required. Travelers would need information about where the available options are in real time. As a result, there would be less risk of private providers sharing information about individual users and their travel history. Even so, private providers of mobility services might not wish to share data they consider proprietary about the real-time availability of their services through a third-party MaaS provider. A.2. Equitable Access to Information Much of the real-time information about and payment for travel is cur- rently provided through apps that are available on smartphones, which are in wide use but certainly not ubiquitous. Regional agencies can ensure that local residents without smartphones or bank accounts gain access to the information provided by smartphone apps through the same strategies used in the pilot projects of shared mobility–transit partnerships described in Chapter 2. These strategies include ensuring that retail outlets sell prepaid debit cards purchased with cash and providing subsidies for these and for low-income residents’ access to smartphones. Regional agencies can also encourage service providers who typically rely on smartphone app access to set up call centers and/or toll-free numbers that others can use for book- ing. Taxi companies have used call centers for decades, but the ridehailing providers are only beginning to do so in the United States.51 Information is often also accessible through computer websites, and computers are avail- able in public libraries, schools, and community centers to those lacking them. 51 Hawkins, A.J. 2020b. Uber Is Testing Out a New 1-800 Number for People Who Don’t Use Apps. The Verge. https://www.theverge.com/2020/2/13/21136090/uber-1800-phone- number-no- apps-smartphone.

106 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY Responsible Entities Metropolitan areas lack comprehensive strategies for creating readily avail- able and complete information about regional modal choices because they typically lack suitable institutions with this authority. Cities and counties can pass local laws or establish policy goals for their own jurisdictions, but, other than relying on state law or policy, they have no mechanism for ensuring consistency or collaboration across the multiple public road and transit authorities or private providers that operate within a region. For metropolitan areas that cross state lines, as several of the largest do, even reliance on state governors or legislators is an imperfect strategy. Despite these imperfect options, some research has concluded that government inter vention will be necessary to reach outcomes that are both efficient and equitable.52,53 That intervention can occur in a variety of ways; Smith iden- tifies four roles that the government can take, calling them MaaS Promoter, MaaS Partner, MaaS Enabler, and Laissez-Faire.54 Each has implications for the path of MaaS within that government’s jurisdiction, and a more com- plete discussion of these paths and implications is available there. MPOs could serve in many regions by establishing information goals and consistent policies, or proposing legislation, and doing so coopera- tively via the representatives of governmental authorities throughout their regions. Large central cities and urban counties can and do also act on their own and, in this instance, could thereby affect the decisions of other regional entities by their leadership and sheer scale. States could facilitate the process in a variety of ways: acquiring and sharing data (or funding such acquisition), persuading, or otherwise requiring that regional entities cooperate, and amending state laws on PII that both allow for the collection and sharing of aggregated information while also protecting privacy. In all these cases, a single region-wide entity that is tasked with establishing these goals and policies will be vital to achieving an integrated transportation system. Regions with multi-modal metropolitan transportation authorities or regional transit authorities could turn to such organizations to host a MaaS system. 52 Pangbourne, K., et al. 2020. Questioning Mobility as a Service: Unanticipated Impli- cations for Society and Governance. Transportation Research Part A: Policy and Practice 131:35–49. https://www.sciencedirect.com/science/article/pii/S0965856418309601. 53 Smith, G. 2020. Making Mobility­as­a­Service: Towards Governance Principles and Pathways. Ph.D. thesis, Chalmers University of Technology. https://research.chalmers.se/ publication/516812/file/516812_Fulltext.pdf. 54 Smith. 2020.

A FOCUS ON THE TRAVELER 107 B. CONVENIENT AND EASILY INTEGRATED PAYMENT FOR O-D TRIPS BY ALL CUSTOMERS Ease of payment is one of the many innovations of shared mobility pro- viders. In the not-so-distant past, travelers paid for tolls, parking, taxis, and transit mostly with cash. Toll agencies and parking providers have gradually transitioned over the past decade or more to credit cards and individual accounts. Taxis increasingly accept credit cards, although in some jurisdic- tions this can still be a clumsy and time-consuming transaction compared with a ridehailing trip. North American transit agencies have largely tran- sitioned to electronic fare cards and some offer payment by smartphones linked to credit cards or by smartphone app. Even so, individual travelers negotiating trips across a region confront multiple payment systems by different providers, thereby increasing information barriers and adding to transaction costs and delays. One way to make multi-modal travel easier is to have a simple and seamless way to pay for multi-modal trips. The apps developed by ride- hailing companies have made payment quick, convenient, and simple and set an example of what might be possible across modes. MaaS software developers accomplish this for multi-modal customers and subscribers to their systems. Ridehailing companies are moving in this direction by ac- quiring other shared mobility providers and, in some situations, by linking to transit agencies in their apps. Other key components, as described in Table 4-2, would include (1) requirements for robust policies to protect PII and (2) provision of payment options for those without bank accounts or smartphones. TABLE 4-2 Convenient Integrated Payment for O-D Trip Capabilities Implementation Strategies Responsible Entities 1. Simple, convenient payment Apps or smartcard systems Metropolitan planning organizations (MPOs), transit agencies, toll authorities, parking operators Credit card payment systems Transit agencies, toll authorities, parking operators 2. Protection of personal information Requirements for protection of personally identifying information Transit agencies, shared mobility providers 3. Options for those without bank accounts Prepaid debit cards, payment by cash, potential subsidies City/county/MPO/state organizations and possibly transit

108 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY B.1. Simple, Convenient Payment Multi-modal payment convergence—the ability to use the same pay- ment medium or technology to pay for services on multiple modes of transportation—is a focus of increasing research and testing.55 To a degree, the market is solving elements of this problem as ridehailing companies incorporate other shared mobility providers in their own apps. The trickier part is in integrating payment for O-D trips that span driving and parking, transit, and shared mobility. As mentioned in Chapter 2, some transit agen- cies and ridehailing companies already link to one another’s apps, but this is not the same as an integrated payment for a single trip. The process of payment convergence will be different for all transit agencies as each works to coordinate with the shared mobility providers. Many transit agencies’ current fare payment systems are card based, devel- oped in the 1990s or 2000s, and bespoke, based on “proprietary system architectures that do not conform to common interface protocols”56 and leading to costly and potentially disruptive transitions to a new system. As transit agencies take steps toward implementing some degree of conver- gence, some are leading their own development and others are contract- ing or coordinating with private vendors. In many cases, because of the significant capital investments required for complete change, agencies are beginning their convergence efforts with mobile ticketing apps instead of fare cards. The past 20 years have seen a great deal of change in fare payments. In 2000, the Washington Metropolitan Area Transit Authority (WMATA) in Washington, DC, was the first to launch a fare card system, which it called the SmarTrip card.57 In 2018, WMATA announced system changes that would make it possible to make SmarTrip payments through mobile devices by 2020. Other agencies followed WMATA’s introduction of the SmarTrip, each with their own system and branding. Mobile ticketing began in Germany and expanded to the United States in the mid- to late 2000s. Google Wallet and Apple Pay joined the array of options starting in 2011.58 Despite this increasing variety of payment options, many agen- cies are still working to implement smartcard fare payment systems, as the deployments can take several years to complete.59 55 Kok, J., and R. Liptak. 2020. Multimodal Fare Payment Integration: A Synthesis of Transit Practice. TCRP Synthesis 144. http://www.trb.org/Publications/Blurbs/180363.aspx. 56 Acumen Building Enterprise, Inc. 2006. TCRP Report 115: Smartcard Interoperability Issues for the Transit Industry. Transportation Research Board of the National Academies, Washington, DC, p. 26. 57 Kok and Liptak. 2020. 58 Miller, A. 2019. Presentation to the committee, October 29. 59 Kok and Liptak. 2020.

A FOCUS ON THE TRAVELER 109 In recent years, the rise of cloud-based processing has led to a trend away from card-centric systems to account-based systems. Card-based sys- tems require a traveler to use a specific card each time they pay a fare. Smartcard readers (fare boxes and fare gates) have a set of programmed fare rules built in and periodically synchronize with back offices to process the fares and update the user’s account. In account-based systems, the fare payment itself, not a specific fare card, identifies the user’s account. They allow multiple types of fare media (fare cards, smartphone apps, credit cards, student ID cards, etc.) to access the same customer account. This type of system sends the information from the fare payment media to the back office, which determines if the card is valid and what rules apply in calculating the fare. The ability of one consumer to use multiple fare media as they travel simplifies the mobility management process. As of 2019, however, although most transit agencies have a form of electronic payment available, only a minority of U.S. systems have account-based capabilities. Most agencies still have card-based systems, and a few have established hybrid card- and account-based systems.60 This does not preclude them from developing MaaS-like systems, however. Card-based systems tend to be older and less able to be easily updated but can still be used, especially in a single travel app (as opposed to data that are widely available to a variety of app developers).61 Mobility payment integration has been in ex- istence throughout Europe and Asia much longer than in the United States; however, the literature on payment integration across providers is limited and primarily generated from the private vendors. The experiences that transit agencies and shared mobility providers have had throughout both Europe and Asia is being incorporated into the products being offered by the private sector in the United States.62 Paying for shared mobility providers is one of the innovations that first set Uber, Lyft, and other ridehailing companies apart from taxi competition. When the user attaches a credit card to their account, all trips are auto- matically charged to that account, negating the need for either an in-person cash transaction or processing a credit card with the driver at the end of the individual trip. This does generally require the user to have a credit card, an issue that is discussed in other sections of this report. Because Uber and Lyft provide additional transportation options in their own apps, it has 60 Kok and Liptak. 2020. 61 Gooch, J. 2020. How a New Fare Payment Delivery Model Is Enabling MaaS. Mass Transit Magazine, February 5. https://www.masstransitmag.com/technology/fare-collection/ article/21124088/masabi-us-ltd-how-a-new-fare-payment-delivery-model-is-enabling-maas. 62 Bartinique, I., and J. Hassol. 2019. Mobility Payment Integration: State­of­the­Practice Scan. Federal Transit Administration Report No. 0143. https://www.transit.dot.gov/sites/fta. dot.gov/files/docs/research-innovation/134056/mobility-payment-integration-state-practice- scan-fta-report-no0143_1.pdf.

110 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY become possible to pay for Uber- or Lyft-branded scooter and bicycle rides with the same credit card already connected to the app. As of the writing of this report, a few jurisdictions including Denver, Las Vegas, and certain agencies throughout Ohio have partnered with Uber to allow the purchase of mobile transit fares through Uber’s app.63,64,65 While the costs of driving a personal vehicle are far more front loaded and less trip specific than in any other mode of transportation, there are still variable per-trip costs on many trips, including parking and tolls. Not every personal vehicle trip requires an out-of-pocket cost, but for a mobility management system to be successful, payment for trips that do require such a cost will need to be integrated as well. There are already some examples of multi-modal payment systems that include parking. For example, in the Washington, DC, metropolitan area, travelers can pay for parking at transit stations with their transit fare card (or with a separate credit card).66 Most toll roads also have an electronic pass system allowing contactless toll pay- ment across multiple toll road systems. The E-Z Pass provides electronic tolling for more than 30 facilities in the eastern and midwestern United States, from Maine to Florida to Chicago.67 The TxTag system allows for use on toll roads in several major cities in Texas, and, since 2017, it has also become compatible with the K-TAG system on the Kansas Turnpike and the PikePass system on the Oklahoma turnpikes.68 Integration through electronic toll passes is thus becoming more common across the United States and would be a necessary step to full mobility payment integration. As all of these systems come together to make payment easier for a traveler, one consideration is the entity that will serve as the merchant of record. The merchant of record is the entity that financial institutions authorize to make and also hold liable for credit and debit card purchases. The merchant of record’s name appears on the customer’s statement and so is the entity the customer associates with the purchase, even if another organization or provider provided the service or had the direct interaction 63 Hawkins, A.J. 2020c. Uber Customers in Las Vegas Will Be Able to Use the App to Ride the Public Bus. The Verge, January 6. https://www.theverge.com/2020/1/6/21048145/uber-las- vegas-rtc-bus-tickets-app-purchase-transit. 64 Bliss, L. 2019. The Uber-Transit Convergence Arrives in Denver. Bloomberg, May 2. https://www.bloomberg.com/news/articles/2019-05-02/denver-uber-users-can-buy-transit-fare- in-the-app. 65 Washington, J. 2020. New Uber App Allows Bus Riders to Buy Tickets in Lake, Medina, Portage, Summit Counties—but Not Cuyahoga. Cleveland.com, July 15. https://www. cleveland. com/community/2020/07/new-uber-app-allows-bus-riders-to-buy-tickets-in-lake-medina- portage- summit-counties-but-not-cuyahoga.html. 66 Parking. 2020. Washington Metropolitan Area Transit Authority. https://www.wmata. com/service/parking. 67 Members. 2020. E-ZPass Group. https://www.e-zpassiag.com/about-us/members. 68 Milestones. 2020. TxTag. https://www.txtag.org/en/about/milestones.shtml.

A FOCUS ON THE TRAVELER 111 with the consumer. If they serve as the merchant of record, transit agencies are responsible for ongoing compliance with Europay, Mastercard, and Visa specifications (specifications for credit cards and the payment termi- nals which accept them); card network specifications; and Payment Card Industry Data Security Standards (PCI DSS), among other requirements.69 In Hannover, Germany, which has been working toward an integrated mobility system since 2004, the transit system generates a joint invoice for all of the mobility services that a customer used in a given month. The invoice breaks out charges for transit, taxis, carsharing, etc., mostly at discounted prices because the customer booked them through the coordi- nating organization. The transit agency then charges the customer the total amount and sends each transportation service their share of the bill. This method requires some back-end work on the part of the transit agency but allows the customer to have a streamlined payment process.70 B.2. Protection of PII Transit agencies have long dealt with PII concerns as they have developed smartcards for customer payments. In addition to the payment data, the rider’s classification (e.g., senior, student, rider with disabilities, etc.) is also considered to be PII.71 Any transit agency that accepts payment via a credit or debit card must adhere to the PCI DSS and to the federal Office of Management and Budget’s PII guidelines that relate to accidental or pur- poseful disclosure of cardholder information.72 Agencies’ policies vary but they generally store PII only as needed for accounting functions and they have privacy and use policies in place to oversee its use. In addition, most states have implemented different laws and regulations about the amount of time that state and local government agencies may retain PII and other personal information.73 Although their protection of PII is not perfect, with regard to payment data, shared mobility providers have achieved broad consumer acceptance of use of credit card transactions by private entities. 69 U.S. Department of Transportation. 2016. ITS Standards Training. https://www.pcb.its. dot.gov/StandardsTraining/modtransit12/ppt/mt12ppt.pdf. 70 Röhrleef, M. 2019. Presentation to the committee, August 14. 71 Kok and Liptak. 2020. 72 U.S. Department of the Treasury. 2020. Treasury Financial Manual, at Chapter 7000, Credit and Debit Card Collection Transactions (T/L 675). http://tfm.fiscal.treasury.gov/v1/ p5/c700.html. 73 Thomas, L.W. 2017. Legal Issues Concerning Transit Agency Use of Electronic Customer Data. Transit Cooperative Research Program. Transportation Research Board Legal Research Digest 48. https://www.nap.edu/catalog/24730.

112 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY B.3. Payment by the Unbanked Title VI of the Civil Rights Act requires transit providers to provide ways for low-income and/or minority groups without bank accounts or smart- phones to pay for their services. As described previously in Chapter 2, transit agencies and other service providers would need to have policies and procedures for cash payments by those who do not have bank accounts and/or smartphones. Title VI requires that policies cannot favor open payment options in price or convenience, requiring that travelers without bank accounts have equally priced and equally convenient options for fare payment.74 If travelers without bank accounts cannot become MaaS users, the promise of system-wide efficiency is undermined.75 Responsible Entities One of the largest challenges to a seamless payment system is the prolifera- tion of entities to be paid. Multiple transit agencies and an evolving array of private mobility providers all may provide pieces of a multi-leg trip, and dividing the single customer payment among the different providers can prove thorny. A regional entity, whether existing or newly created, is key to overseeing and managing the payment of all vendors. Some regions in the United States are already moving toward uniform payment systems for parking, tolls, transit, and more. Some transit agen- cies already have smart fare cards that can be used for more than fare pay- ments; for example, the Bay Area’s Clipper Card, offered by the region’s Metropolitan Transportation Commission, works on trips across all transit agencies throughout the Bay Area and the Ford GoBike bikesharing system. WMATA’s SmarTrip card allows for payment of transit fares and parking at its garages. These systems can serve as starting points for an integrated mobility system, especially for larger metro areas with a dominant transit agency or fare card system in place. FINDINGS AND OBSERVATIONS A customer-centered, multi-modal system could improve efficiency, reduce emissions, and improve equity. This chapter defines major elements of a mobility management framework that would provide 74 U.S. Department of Transportation. 2016. 75 Pangbourne, K., M.N. Mladenović, D. Stead, and D. Milakis. 2020. Questioning Mobility as a Service: Unanticipated Implications for Society and Governance. Transportation Research Part A: Policy and Practice 131:35–49.

A FOCUS ON THE TRAVELER 113 A. Easy-to-access, real-time information on all modes for customers; and B. Convenient integrated options for arranging multi-modal trips and simplified ways of paying for them. Exercising these elements can serve the accessibility, efficiency, equity, safety, and sustainability goals identified in the committee’s Statement of Task. Many cities and transit agencies are addressing individual strategies included in this framework. The framework proposed in this report pro- vides a more complete, holistic approach in which multiple strategies would reinforce the impact of others. Within each of the mobility management categories, the chapter lays out a variety of implementation strategies to achieve these objectives. The findings that follow summarize the strategies and include observations about appropriate next steps. 4.1. Currently, there are barriers to having a fully integrated set of trans- portation services. Significantly, public agencies are unable to systemati- cally gather information on the availability and real-time performance of public and private shared mode options, particularly from the ridehailing companies that currently provide the majority of private shared mode trips. 4.2. For travelers to have consistent information, cities and regions will need to make clear what information aggregators are to provide. This would likely include all possible travel modes and combinations of modes, travel time, travel cost, and some measure of the negative externalities of trips by alternate modes. 4.3. Provision of information and integrated payment will require the co- operation and coordination of a variety of entities and raises the question of which regional entity would take the leading role. 4.3.1. Information currently comes from transit agencies, roadway operators, traffic data analytics companies, micromobility providers, parking managers, and ridehailing companies. All will need to provide their data in standardized form to a back-end processing mechanism so that it can be aggregated and provided to travelers. 4.3.2. Payment processing will require the coordination of transit agen- cies, ridehailing companies, shared mobility providers, and toll and parking operators. For a seamless mobility management experience, the customer will need to pay once, with the distribution of that payment happening in the back end. This degree of cooperation generally does not currently occur within the United States.

114 THE ROLE OF TRANSIT, SHARED MODES, AND PUBLIC POLICY 4.3.3. Given the lack of governance at the regional scale and the varia­ tion in laws and policies across the country, regions that commit to a regional mobility management strategy are likely to seek unique solu- tions to the question of which organization should take the leading role. Regions with metropolitan transportation authorities or regional transit agencies could task such organizations to take on the assign- ment of mobility manager. Those without either could possibly expand the roles and authorities of MPOs, perhaps with an MPO acting as a convener and other agencies following their own missions but in a coordinated fashion. 4.4. The data required for consumers to have a complete array of travel options and means of payment as described in this chapter should not require trip data that could risk disclosure of personal information about individual consumers. However, aggregation by a single app covering all public and private options may require private companies to disclose infor- mation about the availability of shared mode options that some companies may consider proprietary. If consumer-facing systems evolve into private providers’ walled gardens, this concern for companies would be resolved, but it could also result in individual systems that lack all options consumers might want to use. In principle, the disclosure of credit card transactions could reveal the trips of individuals, but that risk exists for trips made today. Policies are in place to minimize this risk, and the public appears to have much less reticence about their personal information being available to private companies than to government agencies. 4.5. A remaining open question is how to create an efficient, cost-effective regional information and payment system that is capable of keeping up with the rapid change in information and communications technologies. Current models for MaaS worldwide fall into three major categories: pub- lic only, private only, or mixed public and private. Public-only models, in which public agencies control both the data acquisition and analysis and the customer interface, may be the best at prioritizing the public good, but may limit competition and innovation that drive down costs for consumers. Private-only models, in which any number of private companies acquire their own sets of data on travel options and provide it to the customer, are likely to be more innovative and responsive to change than a public op- tion, but may result in walled gardens that limit customers’ choices. The mixed model represents a public agency setting regulations and acquiring the data and allowing any interested private companies to develop the cus- tomer interface using those data. This model might work best in the U.S. context given the proliferation of private companies willing and able to sell travel options to consumers and the public’s wariness about private data in

A FOCUS ON THE TRAVELER 115 public hands. However, for MaaS models with a substantial private role, there remain open questions about whether regional MaaS markets would have the needed level of competition to foster innovation and control costs. RECOMMENDATION The committee recognizes that there are uncertainties going forward for juris dictions and agencies attempting to implement the mobility manage- ment framework identified in this chapter. The committee, however, judges that the following recommendation would be an appropriate next step to take while evaluating and sharing the results of early experience. 4.1 Share information: To provide consumers with information about real- time service availability across all modes, cities and states should change their shared mode enabling legislation to require access to such informa- tion. With these data in hand, agencies and jurisdictions should collaborate to create publicly available platforms that integrate and share informa- tion from all sources about modal options and their cost, duration, and emissions.