APPENDIX A

Literature Review

Appendix A presents the literature review conducted to explore how historically underserved groups interact with transformational transportation technologies. Each section includes information and insight, where available, regarding the future of automated mobility. Although the modes in question are not currently automated on a large scale, trends suggest each will adopt some level of automation as that technology improves.

The review is organized into sections as follows:

• Historical Transportation Inequities and Barriers: This section discusses the history of transportation inequities and barriers faced by marginalized and underserved groups to promote a better understanding of the impacts of new technologies.
• Use of New Mobility Options by Underserved Groups: This section examines the use of select transformational technologies with a focus on underserved communities. As an emerging technology, this section also explores the impact of automated vehicles (AVs) and shared automated vehicles (SAVs) on the transportation system.
• Factors Preventing Access and Use: This section highlights spatial, temporal, economic, physiological, and social barriers associated with transformational transportation technologies and new mobility.
• Inclusion-Focused Policies and Goals for New Mobility: Much of the work to improve inclusion in new mobility stems from the policy. This section explores some of the policy goals that have been implemented at different levels of government before outlining specific policies that either improve or limit access and some of the performance measures that can be used to gauge progress toward higher levels of inclusion.
• Infrastructure for New Mobility Options: Often in response to inclusion-focused policies, infrastructure can improve inclusion in new mobility [ranging from better internet networks in rural areas to accessible vehicles in transportation network companies (TNC) service]. This section outlines findings related to new mobility infrastructure.

Historical Transportation Inequities and Barriers

In the United States, transportation access and ease are inherently imbalanced and inequitable due to various contributing factors, including the country’s foundational crimes (e.g., colonization and slavery) and public decision-making influenced by market-based outcomes. As a country, the United States was established on values that emphasized separation and inequity—ranging from the exploitation of enslaved people and land theft for territorial expansion to governmental messaging arguing certain immigrants are criminals or less worthy of citizenship and that women are overly emotional objects with limited personal agency. These conditions have resulted in a tradition of “negative and dehumanizing stereotypes about women and people of color . . .

in the news media and in popular culture” that persists to this day (Osta and Vasquez, 2019). Such conditions limit the representation of communities (e.g., indigenous people and people of color; people with low incomes; immigrants; people with disabilities; women; LGBTQIA+ individuals) that have been traditionally and strategically disadvantaged in decision-making processes, fuel implicit biases, and provide the foundation for structural inequity in all aspects of American life (Marshall-NeSmith et al. 2019; McKittrick 2011; Powell et al. 2019; Sheller 2013, 2018).

Arising from the control required to maintain an economy that functioned from slave labor and a society that required some to suffer so others could prosper, American urban planning, a key factor in access to transportation services, has a “racist history” (Jauregui 2020). This history includes redlining and other government programs to control where Black people live and policies and practices that strategically underdevelop and disinvest in ways that have disproportionately and sometimes specifically targeted communities of color (Jauregui, 2020; Osta and Vasquez 2019; Sheller 2013). The urban renewal practices of the 1950s and 1960s built highways through communities of color and low-income neighborhoods and created precedence that informs ongoing policies and practices and has “led to the destruction of thriving neighborhoods, eviction of minorities, and negative health effects” (Sanchez et al. 2003).

In America, access to transportation—both traditional and emerging modes—is frequently predicated on one’s representation among decision-making stakeholders, influenced by implicit biases, and controlled by the structural inequities that form the foundations of governance and society in the United States. Table A-1 presents definitions of these terms.

Barriers that arise because of services being planned and designed without meaningful input from everyone in the service area preclude certain people from easily or fully using the transportation network. Outlining the then-current transportation landscape in the United States, Sanchez et al. (2003) described persistent challenges and barriers to participation and access that result from historical decisions. For example, many areas of the country rely on personal vehicles due to historical emphasis on highways over transit. As a result of these types of imbalanced investments and the associated residential segregation and land-use disparities (e.g., low-skilled jobs located far away from low-cost housing), underserved populations often experience restricted access to opportunity (both economic and social), education, and health. During decision-making processes, additional barriers are introduced due to limited information availability or language barriers; those users with the time and resources to understand the options and the methods of influencing decisions are often the ones who benefit most from the outcomes (Sanchez et al. 2003). In other words, “Inequity experienced by some people is accompanied by unfair privilege for others who are not burdened by the same disadvantage and who benefit from a relative position of greater power than oppressed communities” (Equiticity 2019).

Table A-1. Definitions—Representation, implicit bias, and structural inequity.

Lugo (2018) explained that the history of race and class division in America has become habitual and self-reinforcing—this “human infrastructure of racism” established a habitual process of separation and division (with enforcement via “reprimands big and little sent down the line to those who deviated”) that supports today’s subconscious or implicitly biased decision-making and continued segregation in our communities. The imbalance in decision-making around transportation means that “access to a car, the ability to safely ride a bike, or the use of public transportation are all impacted by systems of power and inequality” (Baltus 2019). Fullilove (2017) described the American experience as “an ecology of inequality” in which our governance structures rely on the inequality that was first leveraged to establish and facilitate the slave trade and slavery in America. Despite hard-earned civil rights legislation, centuries of racism and resistance have resulted in a “condition of being black in the Americas that is predicated on struggle” (McKittrick 2011). Hamilton and Strickland (2020) explained that this extreme level of inequality is tied to “the use of strategic racism to consolidate economic and political power for the few at the expense of everyone else.” Such inequality results in uneven accumulation of network capital, which Elliott and Urry described as a combination of capacities to be mobile, including appropriate documents, money, and qualifications; access to networks at a distance; physical capacities for movement; location-free information and contact points; access to communication devices and secure meeting places; access to vehicles and infrastructures; and time and other resources for coordination (as cited in Sheller 2018). The result is that those within the elite classes “accumulate network capital, while relegating others to situations of slow, encumbered, or vulnerable mobility” (Sheller 2018).

Even public transit, often promoted as a solution to transportation inequity, is “always pursued for distinct purposes, and for the benefit of some urban dwellers and to the detriment of others” (Enright 2019). According to Enright (2019), this fact highlights the challenges of representation, trying to determine “who belongs in the city, who is allowed to participate fully in urban activities, who decides how space will be planned and produced, and who gains from urban transformation,” factors that are “a complex bundle of often invisible political relations—of for example, poverty, racism, ecology, and citizenship—that condense power dynamics.”

Use of New Mobility Options by Underserved Groups

An aspirational benefit of new mobility services is that they may provide transportation for those who are underserved and disadvantaged due to not being able to drive as a result of disability, age, or lack of access to a car (Zmud and Reed 2019). For example, new mobility services can provide better access to employment. In low-density cities with sparse public transport services, a private car can be a critical factor in finding (and maintaining) paid employment. A study by Junken (2015) examined public transit data from 43 U.S. metropolitan regions (Levinson 2013; Owen and Levinson 2014) to compare the accessibility of work by car compared to transit. For Los Angeles, 92% of jobs required a public transit commute of greater than 1 hour because of multiple transfers, whereas only 7% of jobs required a car commute of greater than 1 hour; other U.S. urban regions have a similar ratio. New mobility services may also provide new opportunities for employment, such as becoming a TNC driver. In terms of healthcare, improved mobility can enable citizens to attend medical appointments more readily, as noted in TCRP Research Report 202 on dialysis transportation (Edrington et al. 2018). More specifically, people might be less likely to miss school/work due to more severe conditions, and it might be possible to treat persons with chronic conditions as an outpatient rather than via inpatient care, thus greatly reducing the overall cost of treatment. Improved mobility might also mitigate the negative consequences of food deserts, which contribute to social and spatial disparities in health outcomes (Beaulac et al. 2009). Although new mobility options have the potential to increase access to opportunities for underserved populations, they may also jeopardize access

by weakening the viability of existing options, such as public transit (Zmud 2018). Therefore, deploying accessible new mobility fleets in conjunction with the expansion of public transit and active transit will help safeguard low-cost and accessible transportation options.

Private transportation options, such as mobility on demand (MOD) or employer shuttles, can expand transportation options for all, “sometimes as an amenity and sometimes as a necessity” (Valenzuela et al. 2005). According to Feigon et al. (2018), private transportation options such as MOD may be the only viable options in certain areas or for certain trips because a public option does not exist. Provision of private options offers underserved communities the chance to be better connected to their regions but also presents potential barriers to access associated with the nature of private business (e.g., cost, span of service, technology or banking requirements, or preferential service provision). Cervero (2017) acknowledged the lack of “inclusive service mandate[s]” for private transportation providers but highlighted the fact that services like jitneys (route-based, flexibly scheduled, low-cost private bus or car services) have “improved access for immigrant and geographically isolated communities for decades.”

Although the user base of new mobility services is still growing, at present, these services serve a small fraction of the U.S. population. For example, as of fall 2018, 36% of Americans had used TNC apps (Jiang 2019), but only 19% of rural residents, 24% of people aged 50+, and 24% of persons in households with incomes less than $30,000 had used such services. TNC users are overwhelmingly urban, young, and affluent adults. Even as bikesharing membership continues to grow, carsharing membership numbers appear to have flattened. Both services are still mostly confined to dense urban areas and represent a small percentage of regional travel (Zmud 2018). Despite potentially providing disadvantaged communities with additional service offerings and the fact that many new service pilots were launched specifically to broaden access to shared systems, these new mobility services have failed to gain traction among older, low-income, and non-White users (Shaheen et al. 2018). Shaheen et al. (2017) discovered that shared mobility users do not reflect their communities well. According to the authors, in addition to the limited representation of people of color among shared mobility users, typical users share the following characteristics:

• Are young.
• Have a high income.
• Have a high education level.
• Are tech-savvy.
• Have access to the full banking system.

Bikesharing and E-Scooter Sharing

Bikeshare systems have the potential to provide additional access to underserved populations because they offer low-cost services and are available around the clock; however, research shows that outcomes do not match expectations. In 2012, statistics about users of bikeshare systems in Minneapolis; Montreal; Denver; and Washington, DC, indicated that “bike share users in these systems are White (79%), highly educated (85% holding a bachelor’s degree or higher) and middle-to-upper income (72% earn an annual income of $50,000 or above)” (Goodman and Handy 2015). Similarly, Ursaki and Aultman-Hall (2015) determined that bikesharing was less accessible to disadvantaged groups. Special Report 319: Between Public and Private Mobility: Examining the Rise of Technology-Enabled Transportation Services showed that bikesharing attracted users who were “disproportionately” moderate- to high-income earners (Committee for Review of Innovative Urban Mobility Services 2016). A 2016 report noted that, in four of eight bikeshare systems in the United States, White users with college education and higher incomes were overrepresented among users (Ursaki and Aultman-Hall 2015)—findings supported

by a survey of bikeshare users conducted by Gavin et al. (2015), which found that respondents were majority White, young, male, higher income, and better educated. In 2015, San Francisco bikeshare members had the following characteristics: 70% were male, 87% held a college degree, 75% were White, 80% had an annual salary of $75,000 or above, and almost 30% reported an annual salary of over $200,000 (Shaheen et al. 2017). According to Populus (2018), in a study of bikeshare use, women had not embraced traditional bikeshare options (docked systems) at the same rate as men (12% of women had used these services compared to 21% of men); however, in early analyses of micromobility options (focused on electric scooters), 3.2% of women had tried the service compared to 4.4% of men. This finding may indicate that micromobility options have a narrower gender gap among users than docked bikeshare systems (Populus 2018). Rayle et al. (2016) noted that there may be a correlation between smaller household sizes and increased demand for micromobility. A study of bikeshare equity in Santiago, Chile, found that 20% of trips were made by users under 15 or over 65 years old (Tiznado-Aitken et al. 2021).

When assessing a reduced-fare program for residents of New York City’s Housing Authority properties, Greenberg (2018) found that fewer than 2% of annual bikeshare members were housing authority residents (despite the fact that the program was developed specifically for this community). One exception to these research findings comes from Baltimore, Maryland, in which Chavis et al. (2018) discovered that bikeshare users in that city have lower incomes and less education and are more likely to be people of color, Hispanic, or female. In a survey of people with disabilities in San Francisco, only 7% reported using bikeshare services due to barriers to inaccessible vehicles and safety concerns (Ruvolo 2021). Dill and McNeil (2021) found that even when bikeshare vehicles are nearby, populations that include people of color, low-income individuals, women, and older adults use the mode less and are less likely to be members of the service.

A case study analyzing bikeshare data in Chicago found that although annual subscription rates were lower in low-income communities, trips from these neighborhoods on bikeshare were longer trip distances, subsequently causing the average trip expenditure to be higher for these users (Qian and Jaller 2020) and possibly indicating a mismatch between land uses. Research on usage of shared mobility pilots in Los Angeles found that each micromobility pilot with programs designed to include low-income riders had different levels of success in attracting these users; more successful programs had more dynamic outreach and input in communities with user groups. Overall, 6% of the total trips taken between the three programs were completed by low-income users (McKinney 2020).

Carsharing

Carsharing presents the opportunity for users to travel according to their schedule, use their preferred route, make multiple stops, travel with family and friends, and carry baggage easily—features that may be most sought after by individuals who do not otherwise have access to automobiles. Despite the potential for high demand among underserved populations, research on carshare programs shows a similar pattern to bikeshare programs—carshare users are predominately high-income, highly educated, young, and White. Furthermore, some researchers have found a correlation between income level and propensity to use carshare, showing that lower income is predictive of lower carshare use (Committee for Review of Innovative Urban Mobility Services 2016; Dias et al. 2017); sometimes, this result can be due to the lack of available services in low-income neighborhoods (Martin et al. 2020). Dias et al. (2017) also determined that the effect of low income on the use of carshare (and ridesourcing) is more pronounced when a family has children. In separate studies of New York City and Oakland, respectively, Shellooe (2013) and Brown et al. (2017) learned that carsharing locations were more prevalent in areas with higher incomes and, in the case of New York City, higher education levels. Despite

multiple studies showing that carsharing is more common among people with higher incomes, Dill et al. (2014) found that Portland’s carshare system was most used by adults 35 and older with lower incomes. Mitra (2021) analyzed data from the 2012 California Household Travel Survey and found that lower-income households are less likely to utilize carsharing (compared to other households). Focus groups in East Oakland found that many participants (particularly those who were Spanish-speaking) had not previously heard about available carsharing services in the area and were interested in using them for one-way errand trips (Pan and Shaheen 2021).

Thumm (2017) noted that carsharing is inherently restricted to those users who possess or can legally acquire a driver’s license—thereby limiting this option to users who are old enough to drive, are able to accept the liability associated with carsharing, and have access to credit; are able to obtain a social security number; and do not have a criminal record that precludes them from having a license. Scooter-share programs can also require users to hold a driver’s license as part of their rules, which can disproportionally affect users of different ethnic backgrounds based on unequal application of the law; for example, racially discriminatory policing on African Americans can result in suspension of driver licenses and limit access to shared mobility options as a side effect (Patterson 2020).

Ridesourcing and Microtransit

Research on ridesourcing users shows more diversity among users than what is seen in research on bikesharing and carsharing; however, much of the user base consists of younger, wealthier, and higher-educated people. A key difference among the users of ridesourcing and other new mobility options is that people of color are better represented. According to a survey conducted by Morning Consult in 2015, “Minorities may use ridesourcing in higher proportions, with 25% of Caucasian respondents reporting having used ridesourcing apps compared to 49% of Hispanics and 41% of African Americans” (Shaheen et al. 2017). The authors noted that the high level of use by people of color could be tied to greater availability of ridesourcing in urban areas and lower automobile ownership rates among this group (Shaheen et al. 2017). Looking at the rate of ridesourcing adoption by different communities, Clewlow and Mishra (2017) found that the highest rate of adoption was among Black users, followed by Asian and Hispanic users, respectively, with White users having the lowest rate of adoption.

Despite the increased participation in ridesourcing by people of color (compared to bikesharing and carsharing) shown by some research findings, the other user characteristics—wealth, age, and education—are unrepresentative of the communities where service is offered. According to Shaheen et al. (2017), a 2016 poll conducted by the Pew Institute found that ridesourcing users had college educations and household income levels above the national median, were under 45 years old, and lived in urban and suburban areas. Rayle et al. (2016) documented findings like the Pew results when assessing ridesourcing users in San Francisco. Though the survey did not ask respondents about their race or ethnicity, Rayle et al. concluded the following: ridesourcing users were generally younger, owned fewer vehicles, and were better educated than the general public; this group often traveled with others; and the group of respondents did not include adequate representation from households making less than $30,000 per year. Focus groups with residents in East Oakland found that TNCs were used mainly for social or recreational trips as a backup mobility option when their personal vehicle broke down; TNCs were not used more frequently because of the cost of trips and rides being canceled by drivers (Pan and Shaheen 2021).

Of note, microtransit is a form of ridesourcing; however, most literature does not differentiate these services from traditional ridesourcing options available from TNCs. An evaluation analysis of a microtransit project in East Gainesville, Florida, noted that agency-owned microtransit

services (wherein the public agency owns the vehicle fleet) can help ensure better accessibility for historically underserved populations through increased provision of wheelchair-accessible vehicles (WAVs) and service-zoning in underserved areas (Mohebbi et al. 2021). Microtransit pilots in Los Angeles and Washington State, which were partially aimed at providing increased transit accessibility to historically underserved populations, revealed instead that characteristics of riders were not significantly different from standard transit users (Lewis and Puentes 2021).

AV Technologies

Shared versus Personal AVs

Recent literature and policies largely assume a shared or Mobility as a Service (MaaS) model for AVs rather than a private use model. SAVs and associated shared model policies can have more equitable outcomes by decreasing the cost of operating, maintaining, and using SAVs, which benefits lower-income populations as well as rural populations who may have higher transportation cost burdens.

Developers currently testing and deploying shared AVs, such as the Lyft–GM partnership as well as Uber in Pittsburgh, Tempe, and San Francisco, provide opportunities for consumers to test AV services to relieve concerns about the new technology (Lewis et al. 2017). This opportunity can provide more equitable outcomes by providing experience to the user—and feedback to the developer—for populations who are distrustful of AV technology or technology in general.

AV market penetration is dependent on the overall costs associated with AV design and deployment. If SAVs can decrease this cost, they will be more competitive and produce more equitable transportation options compared to personal or asynchronous AVs (Ongel et al. 2019). However, SAV convenience could also increase demand for vehicles and result in greater vehicle miles traveled (VMT). To mitigate these effects, early policy intervention is necessary (Paddeu et al. 2020).

On Demand versus Microtransit AVs/SAVs

Although much of the current literature highlights the predicted risks and benefits of on-demand AVs, microtransit AVs present additional travel opportunities to connect underserved populations. Autonomous microtransit (AMT) vehicles vary in size and service, are driverless, and can be used to supplement larger transportation systems. For example, the company EcoPRT successfully piloted an AMT program that transports two passengers at slow speeds (10–20 mph) on shared-use paths at a low cost. The AMT program was used at North Carolina State University in conjunction with the local bus system to increase access and reduce parking [North Carolina Department of Transportation (DOT) 2019].

A study by Ongel et al. (2019) analyzed the cost of electric AMT vehicles compared to the cost of traditional single-occupancy vehicles (SOVs) and public buses. The study found that although the initial cost of electric AMT vehicles was higher than traditional SOVs, the total cost of ownership (including operation and maintenance) was reduced by 75% compared to internal-combustion-powered SOVs and buses. The reduction in cost could be attributed to the electrification of the AMT vehicle’s powertrain (including the engine, transmission, and driveshaft). Furthermore, the authors noted that van and mini-bus-sized AMT vehicles with 10–30 passengers can be used for fixed or on-demand service and operate alongside traditional public transit systems (Ongel et al. 2019). To be used in fixed-route public transportation services in the United States, the vehicles have to be ADA-accessible. ADA-accessible minibuses can serve transportation-disadvantaged populations, particularly low-income and rural populations, at a lower price for riders due to the shared nature of the service and the lower operational costs.

Potential Adoption of AV Technologies

It is difficult to measure consumer perceptions of AVs/SAVs since the majority of consumers lack sufficient information on AV/SAV technology. As a result, most studies focus on early adopters of other new vehicle technologies (Berliner et al. 2019). A 2019 study of early plug-in electric vehicle adopters in 36 U.S. states by Hardman et al. (2019) indicated that early private AV adopters will likely be wealthier and have positive perceptions of safety, comfort, technology, and price when it comes to AVs, along with considerable knowledge about the technology.

Hardman et al. (2019) also identified a socioeconomic study cluster grouping called the “laggards,” the members of which tend to be distrustful and resistant to technology and have negative perceptions of safety and cost. It could be inferred that low-income populations who experience AV cost barriers, older adults who are distrustful of AV technology, and older adults and populations with disabilities who are concerned about AV safety might all fall within this category. Kassens-Noor et al. (2020) found from survey efforts in Michigan that respondents who were older adults, identified as female, or had mobility disabilities were less likely to be willing to ride AVs.

An additional study by Berliner et al. (2019) focusing on early electric vehicle adopters in California found that younger, wealthier men were the participants most interested in purchasing a private AV and were therefore likely early adopters of AVs. The study participants had an average household income of $185,000, and 48% held master’s, doctorate, or professional degrees. Also, irrespective of demographics, participants who perceived AVs to be safer than non-AVs were more likely to purchase a fully automated vehicle. From these findings, it can be inferred that lower-income populations and older adults are not as likely to purchase a fully automated vehicle, which will diminish travel opportunities and widen the travel equity gap. However, greater integration of transportation-disadvantaged populations, such as low-income people, people with disabilities, and older adults, into early adoption studies will allow researchers to better understand the barriers to access facing these populations as well as provide an opportunity for these groups to learn more about AV/SAV deployment and potential increases in accessibility as a result.

An onboard intercept survey of bus riders in Michigan found that safety was a main concern for riders with disabilities not willing to ride in AV buses; these perceived safety concerns included risk of accidents, mechanical failures, and hacking or malfunction of the vehicle computer systems. A few respondents also indicated the need for human assistance either for their disability needs or in case an emergency occurs as reasons not to ride in an AV (Kassens-Noor et al. 2021).

A study by AARP on older adults and new mobility noted that shared mobility companies and developers of AVs have focused on the needs of older adults that are similar to the overall use case population, including with respect to mobility, cognitive and physical capabilities, income, and technology fluency. Although older adults have begun using ridehailing services as an alternative mobility option to driving themselves, not considering the specific needs of other types of individuals may likely result in some populations being excluded from future new mobility developments at the onset. Exclusion of specific needs from the development of AVs will continue to leave older adults reliant on unreliable mobility options even while they represent a demand market for AVs (Fraade-Blanar et al. 2021). Wu et al. (2021) stated that policymakers must require technology companies to include historically underserved populations during the transitional period of AV development, adjust technology development to be free of biases, and test AVs with different users from different socioeconomic backgrounds.

Stantec and Applied Research Associates (2020) highlighted the potential of AVs in both urban and rural areas to enhance mobility and connectivity for transportation-disadvantaged

groups, including older adults, youth under 16 years old, people with disabilities, and people who cannot (or choose not to) drive. This benefit was echoed by Paddeu et al. (2020), who noted that SAVs can improve independence and decrease isolation for older adults, people with disabilities, and people with acute health conditions. AVs/SAVs have the additional benefit of providing first- and last-mile (FLM) connectivity, particularly for people with ambulatory disabilities and older adults who may be otherwise mobility-limited.

A study by Faber and van Lierop (2020) of older adults in the Netherlands found that participants were interested in using SAVs daily to improve mobility and accessibility. Older adults preferred the increased flexibility of on-demand booking and using SAVs for FLM services to alternate modes. Additionally, participants cited the benefits of traveling with friends for increased socialization; access to essential services, shopping, and social/leisure activities; and increased independence and connection with the community.

Factors Preventing Access and Use

Barriers to access and use are associated with various, often overlapping, factors. Peterson et al. (2019) outlined the diverse and interrelated nature of the barriers experienced by underserved communities (and the required solutions), noting that

. . . low-income and transportation-disadvantaged populations face several barriers, and equitable solutions are equally multifaceted, ranging from where a service is located, when it operates and its travel time, to affordability and financial access, physical access, and any number of social and cultural influences.

Freund et al. (2020) created a framework that describes the factors for older adults’ use of rideshare services, which considers biological and social traits, physical traits, special needs, and personal behaviors or preferences. The framework is organized as a hierarchy for each broader category of needs, beginning with individual needs and then working up through interpersonal, organizational, community, and public policy/marketplace needs.

Fraade-Blanar et al. (2021) researched issues in new mobility options for older adults and developed a framework to organize factors impacting use of these services. The framework for older adult mobility factors is organized by individual, organizational, and societal levels, each of the included factors for using mobility services. The framework is organized as follows:

• Individual Level.
  • Demand-Side Factors.
    • Smartphone Access.
    • Smartphone Use.
    • Access to Online Payment.
    • Affordability of Services.
    • Barriers of Trust and Attitude Toward Technology.
    • Physical Barriers.
    • Language Barriers.
    • Cognitive Barriers.
  • Safety Factors (Individual).
    • Non-Crash-Related Health Factors.
    • Personal Security Concerns.
• Organizational Level.
  • Safety Factors (Organizational).
    • Crash Prevention.
    • Vehicle Occupant or Pedestrian Safety.

drivers in these neighborhoods, which effectively increased the cost for users wanting to take trips (Pandey and Caliskan 2021). This challenge of driver/vehicle availability in underserved areas can exacerbate barriers to transportation for low-income populations without access to a vehicle in shelters or temporary assistance facilities that are located away from the most desirable service areas (Robinson et al. 2021).

Because the service relies on fixed stations, traditional docked bikeshare systems are often associated with geographic availability challenges (Populus 2018). A study by Hosford and Winters (2018) noted that in Canada, “advantaged areas have better access to bicycle share infrastructure in Vancouver, Toronto, Ottawa, Gatineau, and Montréal,” while “disadvantaged areas have better access in Hamilton.” Similarly, Kodransky and Lewenstein (2014) determined that fewer stations in lower-income neighborhoods limit access to shared mobility for underserved communities. This pattern of development is enforced by business strategies that focus on locating stations in “attractive, multi-use neighborhoods and commercial corridors with vibrant economies and public spaces, areas where decades of social and financial pressure have minimized the presence of LIM [low-income and/or minority] residents” (Goodman and Handy 2015). Some bikeshare systems acknowledge the need to place stations in areas where they can be used by underserved communities, but these efforts are frequently hampered by limited funding (Ursaki and Aultman-Hall 2015), thus forcing the systems to target areas that have higher densities and higher incomes to garner higher rates of revenue generation from user fees. Whalen (2022) analyzed bikeshare docking stations’ relationship to gentrification using data from Chicago. The study found that low-income neighborhoods located closer to stations in 2014 experienced an increase in median household incomes 5 years later. Although other factors might be at play, this finding raises concerns about the potential for bikeshare stations and infrastructure to inadvertently contribute to the displacement of the very populations they aim to serve.

The spatial availability challenges associated with new mobility are influenced by public resources and planning decisions. McNeil et al. (2019) learned that smaller bikeshare systems have experienced limitations associated with staff availability and funding that reduced the size of the system and its resources. Emphasizing the impact of funding on bikeshare systems, one expert interviewed by the authors divulged that placing the bikeshare service in areas that can generate revenue is the only way some jurisdictions (e.g., cities) are willing to implement a system.

Carsharing services designed specifically to facilitate access to bus systems and bridge FLM gaps are often limited or completely unavailable in lower-income neighborhoods (Thumm 2017). Additionally, low-income areas within service markets tend to have difficulty getting trip requests fulfilled by drivers or having micromobility vehicles staged nearby. Concentrations of dockless micromobility in already highly dense areas can create social consequences through an effect of “splintering urbanism” by widening the gap between citizens with lots of transportation options and citizens without in historically underserved areas (Chen et al. 2020).

AVs/SAVs may introduce specific challenges depending on the operational environment. Although this technology can offer city dwellers increased travel accessibility, if left unregulated, private AV companies can potentially focus efforts on, and profit from, wealthier urban service areas, thereby increasing mobility disparities that have already been observed with current shared mobility modes (Wu et al. 2021). Stantec and Applied Research Associates (2020) noted the resulting possibility that AVs could reduce mobility options and services for disadvantaged populations, particularly low-income urban populations and low-density rural populations.

An additional concern cited by Emory et al. (2022) is the indirect impact on access that AVs can have on rural and low-income populations. Because low-income populations have low rates of car ownership and urban low-income populations are often transit-dependent,

a fear exists that AV deployment will decrease affordable and accessible travel options, such as public transit. Additionally, rural populations already have low access to multimodal transportation, and people without vehicles or the ability to drive are dependent on others for mobility, further limiting transportation options for these populations. Currently, the limited supply of shared mobility options away from the urban core can result in longer trip durations and therefore more expensive travel costs for people living in rural areas (Martin et al. 2020).

To prevent the minimization of low-cost and accessible transportation options for low-income and rural populations, scholars have suggested that AV/SAV deployment should occur simultaneously with the expansion of public transit and active transportation options in both urban and rural settings (Emory et al. 2022).

Temporal Barriers

Time is a component of travel decisions and convenience regardless of mode. Traditional transportation options often involve certain types of time-based compromise (e.g., wait times, pre-planning and scheduling, congestion-based delays, or lack of service options) that many new mobility options seek to overcome (Shaheen et al. 2017). However, while new mobility options are temporally flexible, real-world experience shows that time may still be a barrier for some users of new mobility.

Ridesourcing, often described as a service that is available on demand 24 hours a day, reacts to market forces (e.g., densities of people and jobs) in a way that creates wait times that reflect rush-hour commutes or other high-demand events (e.g., concerts) and might reduce service availability for people traveling outside of these peaks due to the drivers’ ability to focus on high-demand and higher-paying peak periods (Shaheen et al. 2018).

According to Zalewski et al. (2019), the lack of non-SOV transportation options for late-shift workers results in a disproportionate number of late-shift workers experiencing transportation cost burdens (approximately 30% of late-shift workers’ pre-tax earnings are spent on car ownership). Expanding on the potential for increased negative impacts associated with a lack of affordable late-shift transportation options, the authors noted that the late shift is “only growing in importance . . . sectors, such as healthcare, food services and hospitality/leisure, are expected to grow faster than overall employment over the next five to 10 years.” Goodman and Handy (2015) found similar restrictions on late-shift transportation among bikeshare systems, noting that low-income and minority users are burdened by the fact that some systems limit the hours of operation to traditional 9-to-5 work shifts and/or use bicycles that prevent users from bringing their children or carrying baggage (e.g., groceries).

Traditionally, public transportation services have been designed to serve morning and evening commute trips; as such, the needs of other system users (e.g., off-peak commutes, mid-day errands, travel with children) have been neglected (Halais 2020). New mobility services are frequently considered as options to augment or replace transitional transit services; however, if these new options are developed under the same planning assumptions as traditional services, they may not provide benefits for all users of the transportation system. Halais (2020) outlined some of the current challenges imposed by existing service design that may continue in new mobility options if planners do not confront them:

• “Most transit systems aren’t designed for women, who tend to run errands and care for children. . . . Public transportation is sexist.”
• “Women around the world do more care and domestic work than men and their resulting mobility habits are hobbled by most transport systems. The demands of running errands and caring for children and other family members mean repeatedly getting on and off the bus, meaning paying more fare. . . . Transport systems are ill-suited to [women’s] needs.”

factors (compatibility and system quality) impacted participants’ level of trust. For older adults, riders with disabilities, and non-English speaking riders, trust in technology and the perceived safety of AVs will be influential factors in AV acceptance and should be considered in AV design and educational outreach.

Faber and van Lierop’s (2020) study of older adult perceptions of AVs/SAVs found that older adults are particularly concerned about potential AV breakdowns and how AVs will anticipate complex traffic scenarios. The authors argued that trust in AV technology can be fostered with an AV ambassador—a rider/community member/pilot participant who encourages the use of AVs/SAVs by educating others about AV safety and assuaging perceptions of feeling unsafe. Persons with visual disabilities may also have issues with identifying their matched AV from information in the smartphone app. For example, current TNCs often use license plate numbers, vehicle makes and model names, and images to help identify the vehicle; these users can sometimes overcome this barrier by contacting the driver, but the issues will persist with AVs if not addressed. TNC users with visual disabilities can find communicating with the driver to be the most challenging aspect of the travel mode (Brewer and Ellison 2020).

Stantec and Applied Research Associates (2020) cited the issue of trust in AV/SAV technology and lack of awareness of AV services as a potential barrier for non-English speaking populations. To counteract this barrier, educational efforts, as well as in-app AV information, should be included in multiple languages for non-English speaking riders, and alternatives to apps may be considered if riders have a lack of trust or understanding of technology in general.

Service Costs

New mobility services may require additional investment from users compared to traditional options because they (a) require the users to invest in new technology to access the service, and (b) do not receive the same levels of public subsidy as traditional services. Shared mobility may also provide a low-cost solution for bridging existing transportation gaps by providing FLM connection to transit or offering a stand-alone service option in areas without existing public transit. Shared modes can be deployed in underserved areas in less time at a lower cost than traditional transportation projects due to the speed of private-sector investment and operational setup (Shaheen et al. 2017). However, some shared mobility systems are owned by firms that rely on venture capital funds, which increases pressure for the service to be profitable above other considerations and inflexibility with making adjustments to pricing structures (Dill and McNeil 2021; Yaffe 2020). Additionally, when a service withdraws from a market (due to profitability or other concerns), local residents may become confused about the mobility options available for use (Dill and McNeil 2021).

Families with children experience challenges related to overall travel costs when using new mobility options. Dias et al. (2017) determined that the trips and trip-chains required by families with children are so complex (requiring multiple stops and/or longer distances) that use of TNCs can become cost-prohibitive. Affordability of accessible transportation options on TNCs is a key concern for persons with disabilities who wish to use the service (Ruvolo 2021). During the COVID-19 pandemic, TNCs were used by some populations dependent on public transit who were faced with reduced service schedules, which created financial hardship for low-income populations as well as some persons with disabilities and older adults taking more trips on a higher-cost transportation option (Brown and Williams 2021).

Bikeshare users also cite cost as a barrier to using the system. The bikeshare costs associated with diminished access include membership, potential liability, and fares (Howland et al. 2017; McNeil et al. 2017). McNeil et al. (2017) learned that cost concerns are linked to low-income status and race. They revealed that in their study,

with disabilities. In response to the economically costly nature of AVs, Paddeu et al. (2020) suggested implementing government subsidies to pay for free rides for older adults and riders with disabilities.

Beyond the financial cost burdens, AVs threaten to impose quality-of-life costs in low-income neighborhoods. When exploring the potential for AVs in the Boston area, Kaplan et al. (2017) proposed that parking for AV fleets could be located in “inexpensive areas” of the region. People who cannot afford to access AVs may experience missed time-saving opportunities. Some authors forecast the potential for AVs to help users leverage their time more effectively by allowing other activities to take place during travel (Kockelman et al. 2016); however, this benefit will only be available to those who can afford the transportation option. The indirect societal costs of AVs/SAVs include the potential for increased urban sprawl by making travel more convenient. Emory et al. (2022) also noted the concern that governments could channel funds from transit and active transportation to AVs, decreasing low-cost and accessible transportation funding and negatively impacting urban, low-income populations that are often transit-dependent. The authors pointed to the additional fears of job loss, decreased incomes, and decreased job opportunities for TNCs or taxi operators, which are more likely to affect low-income populations and people of color. To counteract this barrier, the authors suggested transitioning drivers to roles as security attendants in AVs/SAVs and setting aside retraining and education funds for former drivers. Additionally, some persons with disabilities may feel safer using AVs/SAVs if an attendant with a valid driver’s license and/or maintenance expertise is available to assist in the event of an incident or emergency (Feeley et al. 2020; Fraade-Blanar et al. 2021). However, this may not be a sufficient solution to cover the indirect costs of AV/SAV deployment, especially for low-income populations.

Despite the significant costs introduced by automation in the MOD market, these technologies are also expected to benefit society if implemented strategically. AVs, and especially SAVs, have the potential to create time cost savings and transportation cost reductions, particularly for low-income populations. A Stantec and Applied Research Associates (2020) report outlined the time savings potential of AV/SAV trips, including riders’ ability to work, sleep, or relax, thereby increasing productivity and convenience. The deployment of convenient and efficient AVs/SAVs could lead to an increased number of routes, increased speed, and improved reliability.

Although cost barriers will likely not be reduced by personal AVs, Emory et al. (2022) pointed toward the reduction in individual AV costs by encouraging shared AV use, which can improve accessibility for low-income populations without increasing cost. To this end, SAVs have the potential to decrease the cost of transportation and reduce the disproportionate cost on U.S. households in which housing and transportation expenses exceed 45% of income (true for two-thirds of American households). This reduced cost burden can be attributed to cost reductions in fuel, insurance, and ultimately AV/SAV technology. Road user charges and additional usage-based fees can also be used to reduce costs for low-income populations and rural populations (if progressively tiered user charges are implemented).

For travelers who are transportation disadvantaged, AVs/SAVs have the potential to increase access to essential services (e.g., jobs, medical care, grocery stores), as well as increase employment opportunities by providing convenient and reliable trips for those with limited transportation options, especially in urban settings (Stantec and Applied Research Associates 2020). In particular, Emory et al. (2022) noted that travelers with disabilities and older adults can benefit from AVs due to lower baseline access to transportation than the general population (if travelers are unable to drive and/or take transit or TNCs). To this end, AVs/SAVs have the potential to help narrow the access gap for these populations.

Banking

A large proportion of the population is considered underbanked, meaning individuals may have a bank account but lack access to or choose not to access mainstream financial services. Consequently, a significant number of potential mobility users are deprived of adequate banking facilities, with younger, undereducated, unemployed, and elderly individuals most at risk. This situation creates a challenging societal inequity through poor access to new mobility services.

According to the 2021 Federal Deposit Insurance Corporation (FDIC) National Survey of Unbanked and Underbanked Households, 28.5% of households did not have a credit card (FDIC 2021). Moreover, 4.5% were unbanked, and 14.5% were underbanked, with much higher rates among marginalized communities. Unfortunately, a major barrier to accessing bikeshare services is not having access to a debit card or bank account. Bikeshare systems often require a credit or debit card to sign up for a membership or require cash payers to make payments at their office (McNeil et al. 2019). For low-income individuals, additional requirements to sign up with some micromobility systems may include government-issued photo IDs, proof of enrollment in assistance programs, and access to a computer to upload documents. These additional barriers could be relaxed/augmented to help potential users access the service (Frias-Martinez et al. 2021).

Credit card and banking requirements have consistently been shown to be a barrier to use of new mobility options in various research projects since 2014. According to this body of research, limited access to the banking system is often related to low-income status, low or nonexistent credit scores, or limited trust in the financial/governmental systems associated with banking—which may be potentially related to immigration status (Cohen and Cabansagan 2017; Committee for Review of Innovative Urban Mobility Services 2016; Goodman and Handy 2015; Kodransky and Lewenstein 2014; Shaheen and Cohen 2018).

Banking challenges extend to automated technology too. Stantec and Applied Research Associates (2020) pointed out that a lack of access to smartphones, credit cards, cell phone data, and the internet can make AV/SAV use challenging or impossible for low-income populations and/or populations who are underbanked or unbanked.

To address these barriers to economic accessibility, Emory et al. (2022) suggested that AV/SAV providers consider cash and subscription-based payment options for low-income populations and populations who are underbanked and unbanked. Digital accessibility can be improved and ensured through the use of the U.S. Department of Justice webpage accessibility checklist (https://www.ada.gov/access-technology/guidance.html). This checklist can be used for webpages and apps to determine the extent to which the content is accessible to most people with disabilities (auditory, visual, cognitive, and ambulatory).

Physiological Barriers

People with disabilities, older adults, and those unfamiliar with current technology may experience barriers to new mobility options related to physiological access—ranging from app usage limitations to a dearth of WAVs (Cohen and Cabansagan 2017; Shaheen et al. 2017). New mobility users who are blind have experienced challenges using ridesourcing services, including trip denials and abuse of their service animals (Wieczner 2015). Wheelchair users also experience ridesourcing trip denials because the drivers cannot carry the wheelchairs (Wieczner 2015), and considerations such as accessible vehicles and apps have not been adopted by all service providers (Shaheen and Cohen 2018). Older adults, whether or not they use a mobility device as an aid, can be less comfortable entering and exiting shared mobility vehicles (Fraade-Blanar et al. 2021). Families with children experience limitations related to the service’s supplied vehicle options (Goodman and Handy 2015) or the need to bring child seats for use in ridesourcing or

carsharing services. TNCs typically provide WAVs through a third-party provider; in most cities, TNCs have been able to operate without providing equivalent accessible service, even though the U.S. Department of Justice has sided with disability rights organizations in some lawsuits (Ruvolo 2021).

Hyun and Cronley (2019) found that carsharing companies also lack considerations for accessible vehicles and services. The authors determined that carsharing operators

. . . need to consider how accessible their cars are for individuals with physical disabilities . . . [and should] maintain fleets of cars that have wheelchair accessibility [as well as] advertise their accessibility features and educate transportation professionals so that people are aware of these features.

People with disabilities are commonly underserved by bikeshare systems. Ruvolo (2021) noted that persons with disabilities may be averse to trying micromobility due to the difficulty in imagining what types of accessible vehicles might be available if the bikeshare or scooter-share systems have not introduced many adaptive vehicles for persons to consider. MacArthur et al. (2020) found that 10 of 70 bikeshare systems responding to their survey had adaptive bikes in their systems. Challenges for operators to implement adaptive vehicles can include specialized parts and maintenance costs as well as an inability to spread vehicles throughout the system. Types of adaptive bicycles include tricycles/quadcycles (to improve balance), tandems (for people to ride together), handcycles (for people with limited/no lower-body movement), recumbent bicycles/tricycles (to allow for riding while seated), heavy-duty cruiser or cargo tricycles (for larger weights and carrying capacity), and electric-assisted vehicles (MacArthur et al. 2020; Yaffe 2020). According to Benedict et al. (2020), the City of Milwaukee, Wisconsin, launched a 6-month pilot of adaptive bicycles through its Bublr bikeshare program, including upright tricycles, handcycles, and two-person side-by-side bicycles, which could all be located and reserved through the system’s app. Both BIKETOWN in Portland and MoGo in Detroit have adaptive bikeshare programs that include “tricycles, side-by-side tandems and hand-cycles, for people who are not able to ride a standard bike share bike” (McNeil et al. 2019). Additionally, Ford GoBike, in Oakland, California, conducted a 6-month pilot rental program to provide accessible bicycles for recreational use (Baldassari 2019).

People with disabilities and older adults thus face many difficult barriers to accessing the vehicles used to provide new mobility services, and the systems used to request and pay for services present similar barriers (Shaheen and Cohen 2018). Leistner and Steiner (2017), when researching a dynamic ridesharing program, found that a significant portion of the participants (about 15%) did not have smartphones when they enrolled in the service and that people who are not already familiar with the concept of dynamic ridesharing may be less likely to use the service. Chang et al. (2019) reported limited effort to design technology that can accommodate users with visual or cognitive disabilities. An evaluation of an AV demonstration in Phoenix found that users with visual disabilities thought the smartphone app for the service was less useful (than other users thought) and that accessibility of the app worsened over the pilot period (Stopher et al. 2021). Limited access due to technological advancements that preclude certain populations from traveling has the potential to result in health inequities and negative health outcomes—a risk that is particularly crucial for vulnerable populations, such as those with disabilities or older adults (Hanzlik and Schweninger 2019).

Wong et al. (2020) looked at perceptions of residents in California about using shared mobility services for transportation during emergency events (specifically wildfires) by conducting focus groups of identified underserved populations. Older adults did not have a positive perception of shared ride services due to concerns about the reliability of drivers and their availability to be in the area. Participants were also skeptical that ridehail drivers would willingly accept a ride to drive toward an emergency (i.e., into harm’s way). Focus group participants with disabilities had negative perceptions of using TNCs in an evacuation event due to viewing

the services as not disability-friendly; issues included not having WAVs readily available, poor communication and lack of assistance from drivers, and past experiences with canceled rides. Low-income participants were likewise skeptical about the viability of TNCs for evacuation trips, concerned with surged pricing, and worried about the inability to pay without a bank account (Wong et al. 2020).

Although the lack of a driver in AVs cuts costs and removes human driver error, it can also make AV participation difficult for riders requiring assistance to enter and exit private AVs by themselves, particularly older adults and people with ambulatory and visual disabilities. Older adults who lose their licenses or are unable to drive have limited mobility, especially in car-dependent rural areas or transit-scarce urban areas. Faber and van Lierop (2020) noted that older adults may have limited active transportation capabilities and may be dissuaded from using public transit if transit is difficult to access physically, is unreliable, or has poor FLM connectivity, thereby limiting mobility options. Poor sidewalk conditions can be additionally prohibitive to navigate to and from transport vehicles for older adults and persons with visual or ambulatory disabilities (Ruvolo 2021). Mobility devices (e.g., walkers, canes, wheelchairs) can make boarding, alighting, and storing devices difficult (Faber and van Lierop 2020). These concerns are echoed when considering accessible AV design for older adults and riders with disabilities.

The National Association of City Transportation Officials (2019) also pointed to barriers outside the vehicle, including limited wheelchair-accessible infrastructure (e.g., level boarding platforms or ramps), which can create additional challenges for riders with ambulatory disabilities or older adults in both urban and rural areas. Such challenges can especially be a problem if AVs lack designated drop-off areas and if vehicle/curb heights differ. Furthermore, a concern exists that AVs may pose a perceived or real safety threat to pedestrians with visual disabilities navigating near or around AVs (Stantec and Applied Research Associates 2020).

Kuzio (2021) noted that the current cost of paratransit can be a deterrent for riders with disabilities and older adults and that both the cost and extent of the accessible design of on-demand SAVs are concerns for these populations. Specifically, Kuzio expressed concerns that fewer wheelchair-accessible AVs will result in longer response times, reliability challenges, and a higher price for riders, particularly in low-density, rural areas. Considering that current AV design is occurring in the private sector without significant input from paratransit riders, scholars speculate the lack of inclusion during the design process will widen the accessibility gap for people with disabilities.

Tabattanon and D’Souza (2021) evaluated retrofits of early test deployments for shared AVs specifically because many first-wave deployments were not designed to accommodate people with various types of disabilities. Retrofitting AVs to include ramps can sometimes have the negating effect of making exit and entry of the vehicle more difficult for ambulatory users while improving accessibility for users of wheelchairs and walkers. The authors noted that subsequent deployments have improved accessibility through additional features such as ramps, audio cues, and passenger announcements. Research also recommends accounting for accessibility accommodations early in the design process and moving toward universal design in AVs to ensure accessibility for most passengers; the other benefit of universal design in AVs is that their benefits can be used by all people (Feeley et al. 2020; Fraade-Blanar et al. 2021). Focus groups working with persons with autism and other disabilities stated that AVs should be designed as fully accessible and include features available in modern transit buses, such as kneeling, ramps, and/or lifts (Feeley et al. 2020).

Despite the challenges involving physical accessibility in private AVs, companies have begun researching and developing more inclusive AV designs through participatory processes. Stantec and Applied Research Associates (2020) noted that Lyft and Aptiv are working with the visually

disabled community to include AV features such as braille guides in future vehicles. The AV company May Mobility designed a wheelchair-accessible AV prototype to be tested for deployment, while AV company Local Motors (no longer in business) developed a prototype designed to communicate with passengers that can be used by people with disabilities.

To ensure accessibility and safety, Kuzio (2021) suggested that accessible AVs could include extra maintenance and safety checks (already required under federal law) to make sure that entry/exit infrastructure (e.g., ramps) is working properly in both urban and rural areas. Federal law has additionally integrated paratransit AVs into MaaS networks that combine public and private entities. An early example of this is the partnership between the City of Grand Rapids, Michigan, and May Mobility to provide wheelchair-accessible AV/SAV shuttles on request. However, inequities remain in the implementation of integrating paratransit AVs, and work still needs to be done in this area to attain equitable transportation outcomes.

To provide convenient, predictable, and safe accessibility options for older adults and riders with disabilities in urban and rural areas, the demand-response nature of current paratransit could transition to AV/SAV networks via automated shuttles, online and app-based booking, and real-time tracking (Kuzio 2021). However, if AV/SAV networks rely on webpage and app-based bookings, an accessible interface that accommodates visual, auditory, and cognitive disabilities is necessary for an inclusive and accessible AV/SAV design.

The U.S. DOT establishes accessibility requirements under the ADA for vehicles used in service to the public, whether operated privately or publicly. Existing requirements for buses, vans, and systems apply equally to SAVs. For new types of vehicles heretofore unseen, U.S. DOT ADA regulations require that standards be developed by the U.S. Access Board and U.S. DOT in concert before deploying such vehicles in use.

The accessibility of public rights-of-way is under the jurisdiction of the U.S. Department of Justice, via its regulations implementing Title II of the ADA. State DOTs have the power to implement regulations requiring accessible pedestrian infrastructure for designated drop-off spots in urban and rural areas before AV/SAV deployment. Additionally, a 2016 federal policy set out by U.S. DOT and the NHTSA stated that AV/SAV riders do not require a license, which should be enforced at the state level to remove an additional barrier to access. AVs must be ADA-compliant to be able to provide inclusive and accommodating travel services for older adults and travelers with disabilities in the United States. Some cities, such as Seattle, which requires a certain percentage of shared AVs to be ADA-compliant, have made additional regulations beyond what is required by the ADA for AVs/SAVs to ensure accessibility for all travelers (Emory et al. 2022).

Social Barriers

Social barriers to new mobility services include challenges such as user perception and understanding, safety concerns, and issues of service provider neglect.

Perception and Understanding

McNeil et al. (2017) conducted a survey and discovered a range of barriers related to limited or incorrect knowledge of new mobility services:

• “34% of lower-income respondents of color said that not knowing enough about how to use bike share was a big barrier to using it, compared to 19% of higher-income respondents of color and 7% of higher-income white residents.”
• “18% of respondents thought that a helmet was required to use the bike share system [but none of the cities included in the study required helmets]. Lower-income respondents were more likely to make this incorrect assumption.”

An additional fear related to sharing AVs noted in the Stantec and Applied Research Associates (2020) report is the concern that shared or subscription-based SAVs may create problems due to riders’ issues of perceived safety with other riders. Studies have shown that African American male riders experience longer wait times and higher cancellation rates for TNC services, while women are more likely to be driven on longer, more expensive routes. Language and cultural barriers can also create safety and discrimination concerns for riders in shared SAVs, particularly for non-English-speaking riders.

To normalize shared mobility before AVs/SAVs are launched, shared mobility for current transportation modes should be integrated and emphasized (Emory et al. 2022). This action can reduce the perceptions of feeling unsafe associated with sharing vehicles with strangers and can decrease the cost of eventual SAV services.

To combat the perception of feeling unsafe, Emory et al. (2022) suggested implementing developer/design safety requirements and considerations for those with perceived or real safety risks (e.g., women, racial/ethnic minorities, non-English-speaking populations) in shared AVs. The risks can be ameliorated with safety designs such as cameras, designated safe-space drop-offs, attendants in vehicles, or seat controls/voice controls to send alerts in an emergency (Feeley et al. 2020). A programmed safe word for vehicles to direct AVs to the closest police station could be an additional mechanism for voice controls to improve safety for passengers (Wu et al. 2021). Research with persons with disabilities found that in a hypothetical AV service setting, voice instructions and controls were preferable to braille writing (Brewer and Ellison 2020). Tabattanon and D’Souza (2021) noted that accommodations to allow service animals and personal care attendants in vehicles may be particularly important to some persons with disabilities who use AVs in the future. Furthermore, Paddeu et al. (2020) noted that low-speed SAVs helped improve users’ perception of safety.

Regarding improved safety outcomes, AVs/SAVs may help prevent avoidable traffic crashes. In rural areas, in particular, there are higher rates of traffic fatalities on rural roads, largely resulting from speeding or alcohol use. If significant investments were made in AV/SAV network deployment in rural areas, these fatalities could be reduced (Emory et al. 2022). These efforts have already begun in Florida and Washington, where a Florida MPO and Washington State DOT have enacted policies to promote agency deployment of SAV shuttles in rural areas (Emory et al. 2022).

Emory et al. (2022) also pointed to the indirect health benefits of electric AVs/SAVs, which include reducing congestion and pollution effects from added VMT, thus protecting low-income and minority populations who are disproportionately exposed to air pollution created by the burning of fossil fuels. The authors suggested incentivizing electric AVs/SAVs, removing parking minimums and instituting parking maximums to control vehicle usage, and aiming to decrease negative environmental externalities that influence vulnerable populations. City governments that have already signed on to electric AVs/SAVs include Seattle DOT and the Association of Bay Area Governments, which require AVs/SAVs operating in the region to be fully electric, and the Louisville Metro Government, which is formulating plans to expand existing EV charging stations for future AV/SAV deployment.

Participatory research and community engagement can be used to inform equitable AV/SAV policy and determine the level of technology accessibility, especially for transportation-disadvantaged populations (Emory et al. 2022). For populations with disabilities and older adult populations, participatory design can be used to ensure the design matches the needs of all users (Paddeu et al. 2020). Although some state and regional agencies have mobilized equitable AV/SAV policy, many more have considered but not yet implemented these policies (Emory

et al. 2022). It is vital for both AV/SAV success and equitable AV/SAV outcomes to launch early-stage, inclusionary development and design efforts. These efforts can include multilingual public engagement/user-experience events, pilot studies with transportation-disadvantaged populations, partnerships with transportation advocacy groups, and educational information for government and community leaders alike (Stantec and Applied Research Associates 2020). Empowerment models of public engagement help community members have the information they need to make informed decisions and provide input to projects; this model can help overcome barriers in community engagement for AVs that have historically been caused by systemic racism, sexism, ableism, and other exclusionary attitudes (Steckler et al. 2021).

A study by Kadylak et al. (2021) discovered that, as of December 2021, 75% of American adults were hesitant about AV/SAV usage. The most common reasons for AV/SAV hesitancy include a lack of trust in AV/SAV technology, safety concerns, and cost concerns (Zmud et al. 2016). Studies have also established that sociodemographic characteristics can inform people’s perceptions and acceptance levels of AV/SAV technology (Zhigang et al. 2018). For example, a lack of trust in AVs/SAVs and safety concerns about driverless vehicles can be barriers to use for older adults and people with disabilities (Zmud et al. 2016).

Ward et al. (2017) examined the perceptions of AV/SAV risks and benefits across different age groups. The authors discovered that driving behavior and feelings toward driving technology differed by age group and that older adults felt that driving was risky and involved human error. Alternatively, younger adults felt that car safety features made them feel safer, and they possessed more AV/SAV knowledge than older adults. The authors also discovered that increased knowledge about AVs/SAVs was positively correlated with trust in technologies because the perception of risk decreased and the perception of benefits increased after watching an informational video about AVs/SAVs.

To increase AV/SAV acceptance by older adults, Ward et al. (2017) noted that the Baby Boomer generation has a significant amount of purchasing power and can buy personal AVs to prolong their independence. Although this may be true in some cases, marginalized older adults and older adults with less disposable income might not experience this benefit.

Ward et al.’s (2017) findings of older adult AV/SAV perceptions were echoed in a study by Kadylak et al. (2021), who discovered that older adults with higher educational attainment, limited transportation options, and positive attitudes toward technology were more willing to use AVs/SAVs. However, the study’s participant composition was primarily composed of White and urban populations. To make findings about older adult AV/SAV perceptions more comprehensive, education efforts and outreach to non-White, non-urban, older adults who are (a) typically not new technology adopters and (b) more skeptical of technology might be beneficial in increasing acceptance and willingness to use.

The study by Kadylak et al. (2021) additionally found that the group most passionate about AV/SAV adoption was of urban, educated, younger men. Because educational attainment is correlated with income, it can be inferred that lower-income populations are less passionate and accepting of AV/SAV technology, as are rural populations, both of whom face higher transportation cost burdens.

Considering that user acceptance of AVs/SAVs and trust in autonomous technology is essential for successful AV/SAV deployment, an emphasis must be placed on public engagement, the sharing of information, and educational/user-experience opportunities for transportation-disadvantaged groups (Paddeu et al. 2020). Furthermore, demographic differences must be considered in designing and implementing AVs/SAVs, and more studies are needed to evaluate acceptance by varying social and economic groups (Zhigang et al. 2018).

Personal Safety

New mobility services often rely on unwritten social contracts that hinge on trust between users, service providers, and the community at-large—contracts ranging from accepting rides from strangers (ridesourcing and ridesharing) to traveling along sidewalks and roadways (bikesharing and scooter sharing) exposed to other people and the elements in ways not familiar to users of traditional transit or personal autos. Naturally, some users may not feel safe under these conditions (McNeil et al. 2021).

Studying bicycling as a mode, McNeil et al. (2017) found that, regardless of race/ethnicity or income, safety concerns are cited by nearly half (48%) of the people they surveyed as the biggest barrier to riding. Similarly, safety concerns influenced by poor infrastructure are also prevalent (Goodman and Handy 2015; Howland et al. 2017; Stewart et al. 2013). Some efforts to remove barriers to bikeshare include adding bikeshare stations in underserved communities to provide easier access to the system; however, according to Chavis et al. (2018), this type of access improvement is only one piece of a larger puzzle that hinges on safety: “Common concerns such as worrying about personal safety, not having a helmet, or being unsure about bicycle/scooter liability are significant barriers that are not overcome by more equitable placement of bikes or scooters in low-income neighborhoods.”

Other personal safety concerns among new mobility users are related to gender and race. Women and people of color have been identified as especially vulnerable when using transit and new mobility options. In a review of a Los Angeles Metro report on women’s travel, Halais (2020) highlighted how women, more likely to have lower incomes and to travel during times when transit service is slow (forcing them to wait in potentially dangerous locations), experience extreme risks when traveling. For example, according to Metro’s data (which do not account for unreported assaults), 14 rapes were reported on transit lines between October 2017 and September 2018.

People of color experience frequent risk as a result of racism and bias that includes police violence and harassment by the general public. According to Brown et al. (2016), potential interactions with police while using bikeshare are cited as concerns by equity communities in several studies. Focus groups consisting of people of color and people with low incomes in Philadelphia documented uncomfortable or intimidating interactions with police, even in instances where the focus group participants were victims of theft or crashes (Hoe and Kaloustian 2014). The Community Cycling Center (2012) in Portland revealed that Latino/Hispanic and African immigrants fear racial profiling and the possibility of deportation as a result of their exposure when cycling. McNeil et al. (2017) further outlined the personal safety concerns experienced by people of color according to income level and compared to White residents:

Race is an important factor in whether respondents feel their personal safety could be compromised, either as a victim of crime or as a target for police attention. For people of color, being lower-income further exacerbated safety concerns. For example, 22% of lower-income respondents of color stated that a big barrier to riding [a bicycle] was that doing so could cause them to be harassed or a victim of crime. This compared to 17% of higher-income people of color and 7% of higher-income white residents.

The implementation of AVs also presents the possibility for further racial inequity and safety issues due to existing biases. According to Wu et al. (2021), AV algorithms have been primarily trained for the technology to recognize pedestrians in the roadway using images of White people; this process has resulted in the side effect of the technology more accurately identifying White pedestrians than pedestrians with darker skin tones during AV testing. Naturally, if unchecked and not proactively corrected, such technology would increase the risk of non-White pedestrians being hit by AVs during a crash incident (Wu et al. 2021). Consequently, the implementation of facial recognition technology in AV/SAV design must consider discriminatory and systemic bias to ensure that certain groups of people are not excluded or put at risk as a result of the software.

Neglect

Neglectful operating practices—which are actions that give little attention or respect to service recipients—among ridesourcing drivers and companies result in racially motivated denial of service and preclude some new mobility users from accessing services. In a study analyzing disparities in wait times for ridesourcing trip requests, the authors discovered a correlated pattern of discriminatory behavior that resulted in wait times for African Americans that were as much as 35% longer and a higher rate of driver cancellations for riders who requested rides with names that were interpreted to be names associated with an African American customer (Knittel et al. 2016). In Boston, Uber drivers were found to cancel rides requested by male users with “black-sounding names” at twice the rate of cancellations for “white-sounding names.” Ironically, in an indication of the potential for corporate culture to possibly bias service provision, the same study found that Lyft drivers did not show the same levels of discrimination (Cohen and Cabansagan 2017).

Inclusion-Focused Policies and Goals for New Mobility

New mobility providers initially deployed vehicles and operated in cities independently of local government controls and initiatives. However, popularity of services such as TNCs and scooter share has increased in urban areas, and cities have begun to consider the ramifications to both their transportation systems and the inclusivity of new mobility options for all residents. Some cities primarily focus on transportation costs as the driving factor for investment and planning decisions, but this thinking can lead to an inequitable distribution of mobility benefits to different system users (DeGood and Schwartz 2016). Meanwhile, city transportation systems and underserved populations are affected by the disruption of new mobility providers to previously existing services. Taxicab companies operating in cities are much more likely to allow cash payment and provide WAVs as part of their fleet than TNCs. As TNCs take a greater share of the for-hire transportation market from the taxi industry, the result may mean fewer travel options available for persons with disabilities or lower incomes (Committee for Review of Innovative Urban Mobility Services 2016).

Meanwhile, many transit agencies have engaged in partnerships with TNCs and other new mobility companies to provide service alternatives to transit. These alternatives can include cutting bus lines with low ridership and replacing them with subsidies for TNC fares. These partnerships are often done as a pilot project to prove the concept of service, which can mean they involve conducting an equity analysis or providing full inclusion to the service (Cohen and Cabansagan 2017). Likewise, the gradual introduction of AVs might result in lower ridership for transit, meaning available funding to support transit will likewise decrease (Kaplan et al. 2017). Because decision-making power is unevenly distributed across communities, the result of these initiatives can produce side effects in mobility inclusion even when there is not ill intent (Tibbits-Nutt 2019).

Stated Policy Goals

Some cities have begun to develop definitions of inclusive transportation and/or policy goals to increase the inclusivity of transportation for underserved populations. Policy goals can include bridging the digital divide, extending access to unbanked and underbanked users, or mandating accommodations for passengers with special needs (Shaheen and Cohen 2018). A report from EDR Group (2019) states that comprehensive accessibility definitions will address three dimensions:

1. “The user group, defining the perspective of the measure (including the level of spatial aggregation).”

A study by Freemark et al. (2020) measured local government officials’ perceptions of AV/SAV policies supported by literature but not often used in practice. Local officials supported AV/SAV policies focused on the right-of-way (e.g., expanding pedestrian infrastructure), equity (e.g., increasing access for low-income populations), and land use (e.g., reducing sprawl). Officials pointed to these policies as being within local municipal power but highlighted a gray area of bureaucratic limitations beyond these policies due to constraints in local power. However, local governments did feel that the political feasibility of AV policies would gain support over time and would help foster the integration of other equity-adjacent policies that have yet to be implemented, such as congestion pricing.

State governments have the power to regulate AV/SAV piloting and testing but not most vehicle design or safety standards (states can regulate window tinting, for instance) (Kuzio 2021). Similar to local governments, state governments have also begun to introduce legislation related to self-driving vehicles. As of May 2021, 38 states had passed AV/SAV legislation and 23 had created AV/SAV requirements for operators, testing, and public road use (Center for Strategic and International Studies 2021). However, the state-by-state patchwork of different rules and regulations makes standardization difficult. Currently, no federal mandates or regulations exist that specifically focus on AVs/SAVs. Instead, interpretations of NHTSA/U.S. DOT vehicle safety and operating mandates inform federal guidance. This guidance includes U.S. DOT and NHTSA’s Automated Driving Systems 2.0 and NHTSA’s Vision for Safety 2.0, both of which feature voluntary AV/SAV safety policy guidelines (Brinkley et al. 2019).

In terms of accessibility, both public and private entities are subject to U.S. DOT ADA regulations, though the specific requirements differ according to whether or not the private entity is primarily engaged in the business of transportation. SAVs themselves are governed by existing U.S. DOT ADA requirements for buses and vans, codified at 49 CFR Part 38, Subpart B. New technologies may, however, include service models and vehicle types that may not fall under existing regulations.

Lewis et al. (2017) cited additional AV/SAV policy considerations. In terms of fostering trust among populations who are distrustful of technology in general or in AVs/SAVs specifically (e.g., older adult populations), U.S. DOT can institute preemptive privacy regulations, and state and local governments can institute data-sharing agreements. Regarding improving AV/SAV safety and reducing the perception of AV/SAV dangers for populations concerned about the lack of a driver (e.g., populations with disabilities or non-English-speakers), the federal government can include AV/SAV technology implementation in the eligible list of federal safety programs. Staffing for SAVs can also be made an eligible operating expense for public entities under FTA grant programs.

Last, when considering concerns about degradation to the environment and decreased accessibility, particularly for populations that disproportionately carry pollution burdens (e.g., low-income populations, minority populations, and limited-English speakers) and populations with poorer transportation accessibility (e.g., populations with disabilities, rural populations), the federal government can create a discretionary grant program focused on transportation projects that meet environmental and accessibility goals. For example, federal grant programs such as U.S. DOT’s Smart Cities Challenge and MOD Sandbox incentivize state and local governments to partner with private industries to research and implement ways in which AVs/SAVs can improve environmental protection and enhance accessibility (Lewis et al. 2017).

Policies or Regulations Facilitating Goals

This section of the review focuses on examples of specific city or other government organization policies or regulations aimed at facilitating equity-initiative goals. Equity initiatives can

fall under the category of either focused equity programs or inclusive access initiatives (Peterson et al. 2019). Focused equity programs are services specifically designed to meet the needs of a specific group. These can include ensuring eligibility of low-income customers or providing on-demand neighborhood services in places without transit. Inclusive access initiatives aim to make regular shared mobility services more accessible to everyone. Such initiatives include offering discounts on fares or enforcing distribution requirements for shared vehicles.

Inclusive Access

Inclusive access can start with ensuring that new mobility vehicles and drivers are available in areas that are underserved by transportation options by developing city policies that provide and stage vehicles in these areas; this process can include requiring adaptive micromobility vehicles for persons with disabilities (Reinhardt and Deakin 2020). For older adults and persons with disabilities, locating services near senior housing facilities, group homes, and other similar housing places (potentially with designated stops or stations at these locations) is an important first step for improving access; Valley Go in the Central Valley region of California is an example of a carsharing program with this strategy (Yaffe 2020).

Many local governments build equity objectives into their permitting processing by requiring targets to be met through annual permit processes or licenses (Peterson et al. 2019); the same could be done in the future with the implementation of AVs (Wu et al. 2021). Examples of the incorporation of equity objectives into the permitting process include the following:

• The City of Denver’s Department of Public Works passed a regulation requiring carshare companies to place vehicles in areas with 30% or more of the population below the poverty line (called “opportunity areas”) (Kodransky and Lewenstein 2014).
• Chicago’s e-scooter pilot program requires scooter companies to distribute half of their e-scooter fleets within the designated priority areas to ensure accessibility to underserved community areas. The city’s scooter report noted that none of the 10 companies have consistently met the requirement for their first deployment in the morning, but rebalancing requirements have appeared to increase e-scooter availability in these communities (City of Chicago 2021).
• Washington, DC, has required carshare providers to locate vehicles in low-income neighborhoods as well (Committee for Review of Innovative Urban Mobility Services 2016).

Likewise, funding can also be a key mechanism for enabling more service in underserved communities, including subsidies to place vehicles or stations in lower-income neighborhoods (Ursaki and Aultman-Hall 2015). One example of an innovative funding mechanism is a California Air Resources Board initiative to help fund carsharing pilot projects in disadvantaged communities using funds from the state’s cap-and-trade program (California Air Resources Board 2015; Committee for Review of Innovative Urban Mobility Services 2016).

Carsharing programs have traditionally limited membership to persons with a current driver’s license so that all members can drive the vehicle. However, this limits access for historically underserved populations. Allowing membership to non-drivers and assigning drivers to their accounts can improve access. In this structure, non-driving members can be responsible for making reservations, billing, and all other non-policies and procedures while taking on legal responsibilities assigned to the drivers; these members will still undergo background checks and approval processes. The dedicated drivers will be covered under member policies and insurance from the operator and thereby incentivized to participate (Yaffe 2020).

Cities and community partners in bikeshare programs have some of the most visible examples of inclusive access in new mobility. The Better Bike Share Partnership (BBSP) is a collaboration between Philadelphia and community partners aimed at bringing bikeshare to underserved communities, particularly communities of color and lower-income individuals. BBSP placed

bikeshare stations in lower-income and racially diverse neighborhoods and conducted targeted outreach in underserved communities. Just Rides is a program between Charleston, South Carolina, Gotcha Bikes, and community advocacy agencies aimed at ensuring access to bikeshare for low-income residents, including by adding stations to targeted neighborhoods and prioritizing rebalancing at these stations (McNeil et al. 2019). The SoBi program in Hamilton, Ontario, has an explicit equity initiative, “Everyone Rides,” that added new stations and bicycles in low-income neighborhoods (Hosford and Winters 2018).

Outreach and rider education can be key for enabling underserved communities to learn how to use services. Just Rides uses community ambassadors, group rides, and other educational programming to engage and open opportunities for lower-income residents (McNeil et al. 2019). The Everyone Rides initiative includes education and outreach activities as well (Hosford and Winters 2018). Private companies sometimes do outreach and marketing directly to underserved areas. In 2019, Uber launched a strategic marketing campaign targeted at lower-income neighborhoods in the New York City outer boroughs, specifically messaging to areas with limited transit access (Atkinson-Palombo et al. 2019). However, independent programs like this are typically either the exception or limited to the launch period rather than being an ongoing component (similar to the accessible bikeshare pilots discussed elsewhere in this report). To further encourage supportive programs, Shaheen et al. suggested that cities and DOTs should also consider adopting a customer bill of rights to help ensure equitable service (Shaheen et al. 2017).

Late-Night Access

Another advantage of newer shared mobility options is the ability to provide additional travel options at later times in the day and evening than traditional public and private transportation services, particularly through the ease of hailing or finding vehicles via mobile technology. Public transit agencies typically reduce or cut service later in the day after peak commuting periods end due to a lack of demand and increased costs, creating a coverage gap for shift workers in the evening. The Massachusetts Bay Transportation Authority (n.d.) conducted a study that listed goals to (a) establish funding for programs dedicated to supporting late-shift transit operations, (b) embrace innovative partnerships to meet late-shift mobility needs, and (c) create frameworks to allow employers to subsidize late-shift period transit. Pinellas Suncoast Transportation Authority (PSTA) and New York Metropolitan Transportation Authority (MTA) are both conducting pilot programs to provide rides for late-shift workers through private providers. PSTA’s TD Late Shift provides free rides on Uber for qualified riders (Uber 2017). Washington Metropolitan Area Transit Authority (WMATA) began a 1-year pilot program with Lyft in 2019 to help with coverage in the early-morning service on WMATA rail for qualified riders at a capped rate of 40 free trips per month (Glambrone 2019).

Shared mobility providers can also provide bonuses/incentives for drivers to service low-income or other underserved areas. A study of ridehail data in Chicago found that Uber used “boost zones” and Lyft used “personal power zones” to increase rates for drivers to service trips in neighborhoods identified by the platform. This strategy for higher pay to drivers may help encourage service to areas away from higher-demand places and provide more equitable service for riders, so long as the additional incentive for the driver is not passed through to the rider as a higher fee for service (Pandey and Caliskan 2021).

Equity Policies

Equity policies and goals help ensure increased access to service for underserved or discriminated-against individuals, either through equality of opportunity or equality of outcome (Peterson et al. 2019). For example, people of color and women can experience significantly greater challenges when traveling on public transit or using shared mobility services. Some

cities have created equity mandates and/or requirements within permitting that require micromobility operators to locate a specified portion of vehicles in underserved areas (Samsonova 2021). San Francisco includes an equity component in its permitting program for scooters; the city requires providers to have a low-income membership option, encourages deployment of scooters in underserved areas, and requires that multilingual information be posted online (Maguire 2018). Transport of London has adopted an equality and inclusion policy with measurable goals for improving experiences for transit users (Halais 2020). Indego in Philadelphia uses a station report card to grade each station via performance measurements, including measures on equity that are based on the percentage of riders with reduced rate passes, people of color, or low-income individuals (McNeil et al. 2019). Private mobility companies themselves frequently have policies on equity and inclusion. A 2017 study found that one in four bikeshare systems had written policies on equity, while many more systems consider equity as part of their system in terms of station siting, fee structure, payment systems, and promotion and marketing (Howland et al. 2017).

Discounting Memberships/Fares

Some programs for increased access to new mobility offer reduced and lifted fares or membership/application fees, including monthly subscriptions or pay-per-trip options (Peterson et al. 2019). These options help offer lower initial entry cost for trying the service, mimic fare pricing of public transportation, or reduce the burden of fees for longer rental times; moreover, offering free trials to users can be a low-risk method for underserved populations to learn about the service and see whether it works for their trip needs (Pan and Shaheen 2021). Just Rides and BBSP have both used discounted membership options or changes in pricing structures to help low-income riders (McNeil et al. 2019). SoBi in Hamilton, Ontario, and Baltimore’s bikeshare has also offered free or subsidized memberships for low-income residents (Chavis et al. 2018; Hosford and Winters 2018). The Jobs Access Program between Lyft and nonprofit Philadelphia Works issues $10 ride credits for clients who are looking to secure employment or training but are outside of public transit service (Murphy 2019).

New mobility services can also offer options for customers to pay either per single trip or through a weekly or monthly membership, as is provided by Baltimore’s bikeshare program (Chavis et al. 2018). Bikeshare users in low-income neighborhoods may have higher average trip costs due to their travel distance needs, causing financial barriers for further use and membership; operators can help alleviate these burdens with reduced fees, early-stage promotions, or longer time limits for free rides (Qian and Jaller 2020). Monthly memberships help reduce the large upfront cost of an annual membership to a new mobility service (Howland et al. 2017; Pan and Shaheen 2021). Allowing low-income individuals to pay for annual memberships in installments helps overcome large one-time payments that present an overburdensome barrier (Goodman and Handy 2015; Martin et al. 2020). Simplifying pricing structures for fares/memberships to be more straightforward and understandable is another policy to consider for further system adoption (Deakin et al. 2020). Additionally, discounted memberships/fares to shared mobility for transit users can help encourage use by historically underserved populations; the Bike + Bus Pass program in Kansas City, Missouri, offers unlimited bikeshare rides (up to 60 minutes each) for users with a monthly bus pass (Patterson 2020). Some transit agencies have used the practice of trip capping—setting a maximum amount for fare purchases in a month and allowing free travel once a rider has reached the monthly pass cost—to help curb inequity caused by fare pricing structures. The literature search found no examples of explicit fare-capping practices by new mobility companies.

Some private companies also have discount programs to help incentivize low-income riders to use their services. These programs, which include free memberships or discounted passes,

are based on proof of eligibility or participation in a state or federal assistance program by the applicant rider. McNeil et al. (2019) documented the following examples of discount programs from new mobility companies:

• Lyft’s Community Pass program offers unlimited rides (up to 30 minutes) for a $5 monthly membership.
• Lyft, in partnership with Uninterrupted, provides free 1-year bikeshare memberships to qualified teenagers through YMCA locations (Lyft 2020).
• Lime’s Lime Access program offers 100 rides for $5 for a standard bike or 50% off rides on e-bikes.
• Pace/Zagster offers $5 monthly memberships for low-income riders.

Some micromobility programs offer eligibility for discounted memberships/rides based on current participation in government benefit/assistance programs (Samsonova 2021). Boston’s bikeshare system, Bluebikes, introduced its SNAP Care to Ride program in January 2018, which offered discounted membership to SNAP participants at $5 per month, or $50 per year, for unlimited trips up to 60 minutes in duration. SNAP Care to Ride was subsequently expanded to include participants of other local government assistance programs; participants could either sign up online or in person at guided enrollment centers. A study of the program found that bikeshare access increased across the city following the program launch, but gains of access were proportionally smallest in communities with the highest need. However, results still represented a 27% increase in access for these communities (compared to 50% in other parts of the city). The authors speculated that additional stations are necessary for high-need communities to make the dock bikeshare service have better system connectivity (Soto et al. 2021).

In San Francisco, affordable membership plan options have helped boost participation in dockless bikeshare programs by low-income riders. Ford GoBike reported that 20% of its members took part in its Bike Share for All Program. JUMP Bike Boost Plan members took three times as many trips on average, while 55% of trips began or ended in an identified “community of concern” (Qian et al. 2020).

An evaluation of shared micromobility programs in Seattle recommended adopting a living income-based discount program that considers the cost of living in the city rather than the federal poverty level standard to account for the higher impact of transportation costs on low-income populations (Beale et al. 2022). Broadening the historically narrow definition of low-income populations in discount programs to include rent-burdened and other similar characteristics of financial strain can help to address barriers for populations with the highest need (Pan and Shaheen 2021).

Alternative Payment

In addition to cost, cities and companies have offered alternate forms of payment to help address challenges that unbanked or underbanked riders face when accessing new mobility services that require use of a credit card (Howland et al. 2017). These alternatives include integrated options for customers to use preloaded debit cards to pay for fares, thus eliminating the need for a debit card or bank account to use the service (Reinhardt and Deakin 2020; Shaheen et al. 2018). PayNearMe is a service used by Indego bikeshare in Philadelphia that allows riders to purchase a barcode for ride payment with cash at local convenience stores (McNeil et al. 2019). Dallas Area Rapid Transit (DART) uses a similar cash payment option for its GoPass mobile app. This solution does not allow the use of direct cash payment for fares, which is rare for new mobility services. However, Baltimore’s bikeshare program, BikeArlington in Virginia, Lime, and Zagster allow for cash payment on their systems (Chavis et al. 2018; Committee for Review of Innovative Urban Mobility Services 2016; McNeil et al. 2019). Similarly, Just Rides

facilitated cash enrollment efforts to help underbanked customers sign up for the service (McNeil et al. 2019).

Bundled mobility programs allow users to pay for shared mobility services using a travel wallet that pays for multiple types of services; integrating fare payment options between public transit and shared mobility services can also help encourage utilization by low-income and other historically underserved populations (Pan and Shaheen 2021; Patterson 2020). Portland’s Transportation Wallet program integrated micromobility services to the public transportation payment platform in this manner while providing prepaid cards to low-income individuals to use multiple services (Beale et al. 2022; McNeil et al. 2021). Importantly, public agencies and shared mobility providers should look for opportunities in which compatible technologies may allow integrated fare payment between services (Beale et al. 2022).

In addition to offering cash payment options that allow customers to pay for fares in a workaround to mobile apps and smart platforms, transit agencies and new mobility companies can help customers get easier access to electronic payment options that the platforms typically use (Committee for Review of Innovative Urban Mobility Services 2016). In Los Angeles, Bird allows customers to unlock vehicles using SMS text-messaging along with cash fare payment (Samsonova 2021). Baltimore provided free checking accounts for customers and removed holds on debit cards for casual users of its bikeshare system (Chavis et al. 2018). Some cities have formed partnerships with banks or credit unions to simplify access to checking accounts and provide a better way for unbanked riders to use new mobility services, though these partnerships are small and have limited impact on customers nationwide (Kodransky and Lewenstein 2014).

Community Education and Feedback

Lack of adequate information about how shared mobility services work and the benefits of using such services can be a barrier for many underserved populations (Leister et al. 2018). Language barriers for non-English or English-as-second-language speakers may be the key challenge in understanding the service when using the available customer education materials. Even simple measures such as advertisements with different types of users (backgrounds, age, fitness levels, etc.) can be useful to show that services are available to everyone (Martin et al. 2020). Programs to educate riders on how to sign up for and use new mobility services can help clear up misconceptions and answer questions about using the services. These programs may include classes and events conducted by staff, ambassadors, or community partners (McNeil et al. 2019). Such programs may be offered directly by the city/company or in partnership with local community organizations. For bikeshare programs, learn-to-ride classes can help novice riders learn the skills to ride the vehicle in their community environment. Educational programs can also discuss options for payment plans, cash payments, and adaptive/accessible needs available to riders (McNeil et al. 2019). Information centers with in-person customer service can be a tool to help educate people about new mobility options and safe use of the system (Auckland Council 2019; Ursaki and Aultman-Hall 2015). Likewise, community education should be offered in other languages to ensure equal access to information in the community (Kodransky and Lewenstein 2014).

Community outreach and education can help build relationships with local partners and improve the visibility and public support for new mobility options in communities that are not as willing to use them (Chavis et al. 2018; Ruvolo 2021). In some communities, additional investment in customer education about available shared mobility service options and discount programs can be more impactful for adoption by low-income populations than simply increasing the amount/presence of vehicles (Pan and Shaheen 2021). Detroit’s MoGo program trains ambassadors on the bikeshare system and uses them to engage and encourage peers in the community to use and adopt bikeshare, coordinate group rides, and assist with customer surveys

(McNeil et al. 2019). Another use of education programs is to promote safety of new mobility options, such as providing free helmets for bikeshare users (Chavis et al. 2018). Education and orientation events can also provide an opportunity to get people signed up for a service, particularly services that require online access to set up memberships, by providing computer terminals and staff or volunteers to assist in the process (Chavis et al. 2018).

Community feedback can also be instrumental in helping cities learn about specific barriers to accessing new mobility services for lower-income individuals or other target populations. These outreach programs should be culturally appropriate for the groups targeted to use the service (Peterson et al. 2019). OakMob 101 in Oakland, California, is an example of a particularly dynamic program that engaged residents in the city’s lower-income areas with low public transit and carshare service availability to learn about their needs for shared mobility companies. Workshops educated residents about shared mobility and collected feedback on barriers to accessing carshare and bikeshare services (Brown et al. 2017). New York City DOT crowdsourced recommendations for new bikeshare station placements to better locate stations in lower-income areas where residents recommended them (Kodransky and Lewenstein 2014). BBSP conducted interviews with community partners to build relationships, learn about the target user community, and build better connections to employment opportunities (Chavis et al. 2018). Civic inclusion models for participation and input into program designs can lead to buy-in from traditionally underserved populations to use the shared mobility service once it is available (McKinney 2020).

Planning

Effective cities use the planning process to develop goals and strategies for improving transportation in their communities. Because of the emergence of newer shared mobility options, transportation planning has started to include these travel modes and vehicle types, too. Successful planning for emerging mobility discusses how technology can help promote the goals of the city/agency, determine what culture change is needed, quantify potential costs and impacts, facilitate communication between stakeholder groups, and develop requirements for operators. Communities within the city/region, including underserved communities, should have the opportunity to participate in the decision-making process as well (Peterson et al. 2019). Planning principles to improve equity in decision-making and goal-setting include reframing traditional viewpoints of planning conversations, allocating funding and resources equitably, improving diversity in leadership, prioritizing underserved and underrepresented communities, and investing in communities without displacing people (Hanzlik and Schweninger 2019).

Beyond representation among decision-makers, data representation can also limit inclusion. If low-income, unbanked, and older adults, along with other underserved groups, are not using new mobility services, data on their travel patterns and needs may remain undetected and unreported. Connectivity via sensors or other communication technologies may help in gathering data to understand the precise travel needs of the underserved, as well as their preferred travel origins and destinations. On-demand services may then be tailored to address these mobility requirements. On the other hand, data emanating from connectivity may lead to inequitable outcomes. For example, service providers may price routes differently, depending on willingness to pay. Opting to take a cheaper route may take passengers on longer journeys, whereas those willing to pay more could access faster premium routes with better roads and more reliable journey times. A corollary impact of this differential pricing may be an increase in traffic volumes being routed through low-income neighborhoods, with associated congestion and collision risk.

Service planning and implementation are challenged by preemptive legislation (such as the statewide TNC laws in Texas), poorly developed service contracts, absence of service following the pilot period, and policies that lack consideration for local conditions.

Partnerships

Shared mobility partnerships have been a mechanism for cities and transit agencies to pilot new services and address existing transportation gaps in the community. Partnerships with private mobility companies work best when the public agency has clear objectives that are used to develop a detailed plan for the partnership. For example, some cities are already subsidizing ridesharing services as an effective way to address the scarcity of mobility options for low-income workers on night shifts. Other public agencies are working with technology companies to use real-time data to make bus routes more efficient and redirect the flow of public transit to the neighborhoods that need it most.

Baltimore formed a partnership with Lyft and community groups to improve access to grocery stores in areas of the city identified as food deserts. The pilot provides qualified residents with eight rides a month using the Lyft app for travel to grocery stores at a flat $2.50 rate (Descant 2019). Dakota County, Minnesota, partnered with Lyft to provide qualified residents on Medicaid waivers free rides to and from work. The partnership is focused on helping people with disabilities find work, particularly in areas without available or frequent transit service (Clarey 2019).

Cities should also involve third parties, such as community-based groups, businesses, and local champions, in the partnership to help with successful dissemination and diversity of investment (Peterson et al. 2019). Intermediaries or third-party brokers (such as advocacy groups or city departments) can help bridge the barriers to serving lower-income or disability communities by identifying barriers, providing solutions, and offering outreach to the community (Kodransky and Lewenstein 2014).

Technology Access

Another frequent barrier for underserved communities is technology access, which has been exacerbated as new mobility options have driven the use of smartphones and availability of internet access and data plans to be nearly required for riders to use shared mobility services. However, cities and companies have found some solutions for easing the barriers in technology created by new mobility services. Concierge services for customers to make trip reservations and payments can also help overcome other technology and payment barriers for customers; tracking of completed pickups and drop-offs on a ridehailing service can be an additional step to ensure customer safety (Deakin et al. 2020).

Many partnerships for ridehailing or microtransit include a call center option to allow customers to request rides by phone rather than through a mobile app, while Lime’s Lime Access program in New York City allows riders to unlock e-scooters using a text-to-unlock feature, which means they can simply send a text message to a designated phone number for unlocking, eliminating the requirement of owning a smartphone (Lime 2019). If call centers/concierge services are established for new mobility services, having call-takers in common non-English languages is another important step to ensure access for historically underserved populations (Deakin et al. 2020). Improving accessibility within the technology for additional features, such as voice-activation compatible features in the smartphone application for persons with disabilities, is also necessary (Yan et al. 2021).

In-person enrollment or paper application processes such as those at Ithaca CarShare or dockless bikeshare in San Francisco allow applicants without internet access to sign up (Kodransky and Lewenstein 2014; Martin et al. 2020; Qian et al. 2020). However, little is found in the literature about improved access to internet and cellular networks in areas with poor coverage, which in turn lessens the ability of affected residents to request rides on demand.

Incentives and Subsidies

Subsidizing the cost of AVs/SAVs and providing a more equitable pricing structure for AV services can benefit low-income and rural populations and can be used to incentivize the shared use of AVs. Allowing for multiple payment options will also remove barriers to access for low-income populations, unbanked/underbanked populations, and populations without access to a smartphone or credit card. AV/SAV operators can additionally be required to provide service to transit-scarce urban areas and rural areas as a public right-of-way operating condition. The enactment of these regulations can be used to distribute service more equitably, particularly for low-income and rural populations (Stantec and Applied Research Associates 2020). Incentives for current shared mobility providers to locate carsharing or micromobility systems in low-income communities are another tool for cities to reduce spatial inequality (Deakin et al. 2020; Martin et al. 2020). Another incentive may be reducing or eliminating taxes (such as car rental taxes that are applied to carsharing services) so that operators will locate their services for local residents (Martin et al. 2020).

Policies or Regulations Preventing Goals

Some policies and regulations put into place by city and state governments on new mobility options can have negative consequences by creating or reinforcing barriers to underserved populations wanting to use the services. City intervention in the supply of vehicles via permitting without specific requirements to ensure service in all parts of a city can create issues with limitations on the span of service or preferential service provision by the private mobility company due to such companies’ pursuit of the highest level of demand and user income, consequently leaving a lack of service provision or available vehicles in lower-income areas (Cervero 2017). Caps on shared vehicles and high insurance requirements can also limit the number of providers available to users, thereby facilitating monopolies for larger private companies and decreasing the amount of competition available to improve equity of shared mobility options. Insurance options for shared mobility services that alleviate costs for operators can be another way to lessen barriers to entry into local markets (Martin et al. 2020).

For ridehailing/TNCs, a majority of state legislation preempts the local city authority’s ability to regulate, tax, or impose rules on TNCs (Moran et al. 2017). These state regulations with preemption limit the ability of cities to require service provision and accessible vehicle options in underserved neighborhoods. Currently, most states do not have legislation in place governing micromobility modes such as bikeshare and scooters; passing similar preemptive regulations for micromobility could lead to additional barriers in service equity created by policy. For carsharing and bikesharing, cities with their own programs often have membership fees required to use the service. Any membership fees that are set at an annual payment rate present a barrier for lower-income individuals, who may struggle to afford the service at a high one-time cost (Goodman and Handy 2015).

In general, jurisdictional rules and requirements related to new mobility vary, which can potentially challenge service providers and limit or hinder access for some groups (Deakin et al. 2020). However, despite some existing literature that describes jurisdictional issues associated with new mobility, none of the work deals with jurisdictional issues from the perspective of underserved communities. The following is a list of some of the existing literature:

• TRB Special Report 319: Between Public and Private Mobility: Examining the Rise of Technology-Enabled Transportation Services discussed issues of jurisdictional rules on TNCs but is not related to equity (Committee for Review of Innovative Urban Mobility Services 2016).
• TCRP Research Report 202: Handbook for Examining the Effects of Non-Emergency Medical Transportation Brokerages on Transportation Coordination discussed issues of non-emergency

Peterson et al. (2019) recommended focusing new mobility services on specific target objectives for equity, including defining user groups or neighborhoods and building the service (with involvement from those communities) to meet their needs. Cities should balance the needs of existing ridership while attracting new riders to services by using metrics that focus on transportation happiness or performance-based community engagement to improve access and should allow for flexibility to ensure broad participation (Shaheen et al. 2017).

Infrastructure for New Mobility Options

Infrastructure for new mobility includes facilities or technology assets that assist in the provision of vehicles and facilitate (or bar) access for underserved populations. For carshare and bikeshare options, there is associated infrastructure for vehicle parking and docked stations. Dockless bikeshare and scooter options may also have associated infrastructure, such as designated parking areas for vehicles to be staged for new customers that are free of sidewalk clutter. Curtailing clutter from micromobility vehicles particularly helps persons with disabilities and older adults successfully navigate sidewalks because it removes additional physical obstructions that impede access to transportation (Chen et al. 2020; Ruvolo 2021).

Infrastructure as a Hindrance to Access

Siting supportive infrastructure for new mobility options, such as docking stations, bike lanes, and designated pickup/drop-off areas, is critical in promoting the availability and adoption of service in low-income areas. Providing limited docking station sites in underserved areas likewise offers limited capacity for returning vehicles. Researchers found that the availability of bikeshare stations in Philadelphia and Chicago was correlated to income levels, with greater availability in higher-income areas (Niemeier and Qian 2018). Similarly, an analysis of CitiBike stations in New York City in 2019 found that the greatest proportion of bikeshare stations were in wealthier census tracts, with no significant changes in station distribution during the service’s recent expansion (Babagoli et al. 2019). Micromobility companies with bikeshare and scooter-share vehicles sometimes include equity considerations in station siting processes, but not all companies have explicitly included equity in their process (Shaheen et al. 2014). Additionally, equitable distribution of shared mobility vehicles and stations does not guarantee that underserved populations will use them (Chavis et al. 2018; Clewlow et al. 2018). For micromobility vehicles traveling in urban environments, protected bike lanes provide a safe and comfortable riding experience for customers. Investment in infrastructure for non-car modes of transportation is helpful for historically underserved populations to maintain access to needed mobility options during emergency events with reduced service availability from public transport and on-demand vehicles (Brown and Williams 2021).

Bike lanes provide a sense of security for micromobility users by limiting exposure to automobiles on roadways through separated bike paths or bike boxes at intersections (Ursaki and Aultman-Hall 2015). Goodman and Handy (2015) found that safety concerns due to poor bicycle infrastructure can be a barrier for low-income individuals to use bikeshare in their communities. Moreover, these communities often have fewer advocates for improvement of infrastructure, thus contributing to less investment and siting of lanes and stations than in more affluent areas (Howland et al. 2017). Addressing problems of a lack of or deteriorating infrastructure can help some user groups feel more comfortable using bikeshare and e-scooter share services (Reinhardt and Deakin 2020).

Local regulations on parking minimums or access to right-of-way can sometimes be outdated and discourage shared mobility operators (like those of carsharing and bikesharing systems)

from deploying their services in certain neighborhoods. Cities should look at these regulations and determine the right balance of need for application versus relaxing requirements to make services available for local residents (particularly traditionally underserved populations) (Martin et al. 2020).

Internet and cellular network access is a key component of transformational technologies in transportation. Limited internet availability and reliability in underserved communities, particularly rural areas, make it difficult for residents to access new mobility services, which require high-speed data connections for real-time data and location services for locating users and processing transactions. A study on autonomous and connected vehicles for rural, isolated, tribal, or indigenous (RITI) communities stated that major barriers to service equity include lack of communication infrastructure, lack of electrical power, high costs of expanding communication and power networks, and lack of local champions in the community to support these expansions (Sorour et al. 2022). Since new mobility services rely on mobile apps for customers to hail trips or unlock vehicles (Dias et al. 2017), the lack of internet or cellular coverage can limit the ability of users to access the service. This infrastructure can also provide better up-to-date travel planning information for users before they access the vehicle (Koeppel 2017). Kodransky and Lewenstein (2014) found that the requirement of internet access for participation in carshare service was less likely to be met by low-income persons. In 2016, the Pew Research Center found that 13% of adults in the United States did not use the internet, and lack of internet access was predominant among persons older than 65 with incomes below $30,000 who lived in rural areas (Shaheen et al. 2017).

Infrastructure-Specific Policies and Regulatory Strategies

No literature was found on specific policies or strategies being used to improve internet infrastructure in underserved areas. One solution to internet access issues for new mobility services is installing kiosks for users to unlock vehicles, make trip requests, and pay for fares and subscriptions. Kiosks have been used with docked bikeshare systems to allow users to pay for subscriptions under payment plans or for the cost of a single-day membership (Goodman and Handy 2015). However, these kiosks have the potential to provide better service in denser urban areas through better internet infrastructure and improved cellular coverage (Shaheen et al. 2017).

Another potential solution to limited internet availability is designing “lite” versions of smartphone applications that use less cellular data for essential processes and then download supplemental data once Wi-Fi is available (Shaheen et al. 2017). New mobility services can also use paper membership applications to enroll persons without smartphones. Ithaca CarShare allowed paper application processing through its Easy Access membership plan for persons without internet access (Kodransky and Lewenstein 2014).

Targeted siting of new mobility services and associated infrastructure in underserved areas requires measuring potential demand for service and incentives or subsidies (Kodransky and Lewenstein 2014; Shaheen et al. 2017). Transit agencies and government entities can also work with both private new mobility companies and community organizations to locate infrastructure in underserved neighborhoods. Los Angeles Metro conducted crowdsourcing efforts with community members to gather input on locating bikeshare stations, including distributing informational flyers and available phone numbers for providing feedback (McNeil et al. 2019).

Populus (2018) stated that the presence of micromobility companies can help cities gain increased public support for improved active transportation infrastructure. Shaheen et al. (2014) noted that bikeshare operators should work with city agencies to improve bicycle infrastructure that would generate additional users. Another issue for bicycle and scooter-share supportive infrastructure is that management of these systems is separated from city transportation

departments, yet these functions need to work in harmony to determine safe micromobility routes of travel and facilitate upgrades to infrastructure in needed areas (McNeil et al. 2019).

For AVs, there is a need to develop a standardized policy framework for AV/SAV design and deployment (Emory et al. 2022). Without standardized regulations, AV/SAV design, deployment, and networks can lack coordination. Some cities and states have already implemented their own AV/SAV frameworks, such as Portland, Oregon, which developed a framework (Fleets of Automated Vehicles that are Electric and Shared [FAVES]) that requires AVs to be electric and shared. Hawaii’s AV policy framework (Accessible, Automated, Connected, Electric, and Shared [A2CES]) requires that AVs be electric, shared, accessible, and connected. Further, Minnesota DOT identified a set of goals and associated strategies to inform Minnesota state policy to require the accessible and equitable design and deployment of AVs/SAVs (Emory et al. 2022). The standardization of an AV/SAV framework informed by equity and accessibility goals will ensure that transportation-disadvantaged populations are included in the design and deployment of AVs/SAVs.

Elements of Successful Business Models

Business models of shared mobility systems require improvements in structure and public agency coordination to ensure quality infrastructure in underserved communities. Shared mobility companies focus their services in areas with the highest demand, meaning they are unlikely to target low-income communities. A better understanding of regulations and incentives to improve access to services can improve the sustainability of shared mobility business models (Kodransky and Lewenstein 2014). Public agencies must understand the value proposition of for-profit companies to provide service in underserved communities, including subsidies needed and efforts of businesses to reach these populations. Including a mix of partners from public, nonprofit, and for-profit sectors is a mechanism for expanding new mobility services to low-income communities since public and nonprofit entities can help identify barriers and alternative revenue sources, thereby enabling private companies to scale up service in these communities (Kodransky and Lewenstein 2014).

Shaheen et al. (2017) found that equity outcomes in shared mobility business models were difficult to assess, particularly in cases where private operators received monetary support from agencies with federal funding as part of pilot partnerships. Subsidies from federal funding may not be a substitute for removing existing financial barriers in shared mobility business models but instead may function to help subsidize operational expenses. Eventually, some shared mobility business models for ridesharing/ridesourcing services will adjust to AVs instead of available drivers.

Shared mobility business model advantages in providing service in underserved communities include flexibility in filling current coverage gaps, service in late-night and off-peak periods, lower upfront costs to users, tailored services to specific physical and cognitive disabilities, and flexible payment and access options (Shaheen et al. 2017). New mobility services can also make access to systems for low-income and disadvantaged users more like public transit models, with specific access mechanisms for these user groups. Healthy Ride in Pittsburgh is an example of a bikeshare system that developed a pricing model that complements public transit pricing (McNeil et al. 2019).

References