Private Transit: Existing Services and Emerging Directions (2018)

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

8A variety of services could reasonably be included in the definition of “private transit,” includ- ing employer and university shuttles, jitneys, turn-key private services provided under contract to public transit agencies, and new on-demand shared services that have been made possible by widespread adoption of mobile computing platforms. This report generally focuses on transportation services that meet the following criteria: • Private and Commercial – The services are owned and operated by private entities (including both for-profit and non- profit entities).1 – The services are provided as part of a commercial transaction between private entities (such as a service that gives rides in exchange for fares paid by individual riders or a service sponsored by an employer or building manager, in which the transaction is between the sponsor and the vehicle operator) or are supported by a public-private partnership with a substantial contribution from the private members. – The driver is paid for driving (rather than volunteering or taking others on a trip the driver would be making anyway). • Transit-Like – The services are shared concurrently by unrelated users. – The services provide intraregional trips (i.e., within a single metropolitan area, although often across municipal or state boundaries). – The services are available for regular, ongoing use (i.e., not related only to a special event or for sightseeing). Vendors and operators in the transportation sector often provide a range of services that meet only some of these criteria at a given time, with the public/private distinction being espe- cially fluid for some contracted services. While this report describes a few cases where the lines between public and private entities are not quite clear, it primarily focuses on services that meet all the criteria listed above. These criteria are described at greater length in Appendix B, which also includes a longer discussion of university transit systems, which are related to, but not within the scope of, this project. Taxonomy of Private Transit Services Despite the reorganization of some markets through the introduction of innovative hailing, dispatch, and routing technologies, the basic forms of services that make up the transporta- tion system have not changed significantly since the 1990s. Consequently, this taxonomy draws S E C T I O N 2 Defining Private Transit Services 1This includes the provision that the services are not primarily provided as part of a contract to a public transit agency, but this is not a categorical exclusion from consideration.

Defining Private Transit Services 9 heavily on a structure established by Robert Cervero (1997, 14), adding or adjusting categories where necessary. Most of the service dimensions outlined below possess largely the same options as they did 20 years ago. There is continuity in the options for passenger capacity (no funda- mentally new vehicle configurations have been introduced), stop configuration, service areas, and access requirements. One notable change, however, has been the increase in options for requesting rides and collecting fares (and behind the scenes, for organizing the market for drivers). These changes have reinvigorated the private transportation market in recent years and introduced sev- eral new players. Small-vehicle, taxi-like services have grown in usage due to the ability— popularized by TNCs but not exclusive to them—to request a ride from anywhere at any time and the ease of paying for that ride electronically, usually within the same mobile application. This research identifies five primary private transit service types: • On-demand pooled services • Prearranged route- or zone-based services • Flexible route-based services • Employer-based commuter services • Property-based services At the broadest level, this taxonomy categorizes these services based on the primary commer- cial relationship underlying them: • Private market services: The service provider does business directly with the rider, generally for a fare (business to consumer). • Sponsored services: A property owner or employer (or consortium made up of a mix of these and, at times, a public-sector partner) contracts with the service provider for the benefit of some other party, such as tenants or employees (business to business). Broadly speaking, private market services include any shared private transit service that is arranged for directly by the passenger where the commercial relationship is primarily between the passenger and the operator (usually in the form of fare payment). This con- trasts with sponsored services, where the service is paid for by some other entity, such as an employer, developer, or property management company. In sponsored services, the sponsor- ing entity typically pays for or heavily subsidizes the service, although riders may pay some fare to offset this cost. In the case of property-based services, the level of sponsorship may vary among a property’s commercial tenants; that is, on the same vehicle, users from one tenant may pay a fare while users from another tenant may have their costs fully covered. On a demand-responsive to fixed-route spectrum, private market services generally fall toward the demand-responsive end (with the notable exception of prearranged route-based services, or microtransit) while sponsored services are almost uniformly on the fixed-route end, at least with respect to stop locations. This is more of a guideline than a rule, however, and is changing as DRT-enabling routing algorithms and real-time hailing and tracking become more common for all types of transit services. Table 1 categorizes each service type on five dimensions, and the “Private Market Services” and “Sponsored Services” sections below provide greater detail about the various services included in these two categories, including examples of each service. In Table 1, where multiple values are listed, this indicates a range of possible configurations for an attribute. Further descriptions of the attributes within the taxonomy are presented in Appendix C.

10 Private Transit: Existing Services and Emerging Directions Private Market Services Transit services, whether public or private, thrive in dense environments where the costs (in time and money) of owning and using a private automobile are higher. Private market transit services generally appear in the same markets and often concentrate their service on the same corridors as public transit services. Private market services often provide a premium level of service compared to some aspect of public transit service, such as frequency, direct connection between specific locations, or vehicle amenities (Goldwyn 2017, interviews with Goldwyn 2017, interview with King 2017). Costs compared to a public transit ride can range from roughly equal (in the case of jitneys) to many times as much. The private transit services that primarily do business directly with riders can be grouped into three categories: • On-demand pooled services (such as shared taxicabs and similar TNC-based pooled services). • Prearranged route- or zone-based services (including airporters and “microtransit” services). • Flexible route-based services (including the several varieties of street-hailed jitney). On-Demand Pooled Services This category of services comprises shared taxicab and taxi-like services, including TNC- based services such as Lyft Line and UberPool that generally use smaller passenger vehicles, as well as services operating larger passenger vehicles [sport utility vehicles (SUVs) or minivans], such as Via. While this category has recently been expanded by the addition of the application-based services, shared taxicab services, in which unconnected passengers share a portion of their ride, have been a feature of the transportation landscape for decades. However, taxicab regu- lations in many jurisdictions prohibit sharing or limit the area where it is permitted (to rides Service type Stop coniguration Service coniguration Ride request format Fare collection Access requirements Private Market Services On-demand pooled services (shared taxis or TNCs) - Many-to- many -Demand responsive - Zone route - On demand - Online/app - Cash - Credit - Technology restrictions Prearranged route- or zone- based services (microtransit) - Few-to-few - Few-to-one - Fixed route - Zone route - Prearranged - Ticket - Online/app - Technology restrictions Flexible route- based services (jitneys, dollar vans) - Many-to-one - Few-to-one - One-to-one - Fixed route - Route deviation - Request stop - Street hail - No arrangement - Cash - Ticket - General public Sponsored Services Employer-based commuter services - One-to-one - Fixed route - Request stop - No arrangement - Free/subsidized - Online/app - Ticket - Employee only Property-based services - Few-to-one - One-to-one - Fixed route - Request stop - No arrangement - Prearranged - Free/subsidized - Online/app - Ticket - Afiliated-only Table 1. Taxonomy of private transit services.

Defining Private Transit Services 11 originating at airports or at particular venues, for instance) (Chicago City Code § 9-112-560, DC Mun. Reg. 31-808.2). The on-demand pooled services differ from other services in the private market category in that they travel between ad hoc stops anywhere within a defined service area, which can range in size from the densest portions of the urban core to the entirety of a metro area. The Uber and Lyft pooled services are offered around the clock in the markets where they operate, while Via operates at a limited capacity outside of peak hours. The rides are requested and paid for through a mobile application; thus people without a smartphone or without a credit card are generally unable to use these services. The major TNCs have recently begun programs to allow users to request rides on another’s behalf, which begins to address these limitations. However, thus far these programs appear to be marketed to specific enterprise clients, rather than being a feature of the basic application, and so are not yet available to all users. Only in the largest of the cities where the major TNCs operate do they offer the pooled ver- sions of their services, pointing to the large overall (non-pooled) TNC market needed to support this subclass of the business model. As of March 2016, more than half of Uber trips taken in several cities were within the UberPool fare class, and the company told the New York Times that month that more than 100,000 people used the service class each week in New York, Chicago, San Francisco, and Los Angeles; that number has likely grown since (Manjoo 2016). The same article reported that more than half of all Lyft rides in San Francisco and New York City use the Lyft Line class. In summer 2017, Lyft stated that “over 40% of rides are shared rides in markets where Lyft Line is available” (Walker 2017). However, as Schaller notes in a study of New York City TNC rides, these pooled-ride fig- ures represent a top-end estimate because “customers selecting pooled options are not always matched with other riders, so the actual number of shared trips is lower” (2017, 7). Via currently operates in New York City, Chicago, and Washington, DC. The service focuses only on the densely traveled corridors or zones of the cities, rather than offering availability throughout the region like the TNCs. Via told the researchers that it provides 250,000 rides per week across the three markets (the clear majority in New York City) through roughly 5,000 drivers who take part in its network (Via interview 2017). The company also appears to be working toward more partnerships in which it provides only the ride-matching platform, rather than the vehicles and rides themselves. In June 2017, Via announced a new partner- ship with the taxicab network Curb, allowing users to arrange and pay for shared rides in New York City yellow cabs at a discount to the metered fare for the shared portions of rides (Hu 2017). That same month, the company also announced a pilot with Austin’s Capital Metro in which the transit agency would use Via’s software to provide on-demand shared rides within specific service zones in the city. Consolidating rides into fewer vehicles appears to be central to the long-term viability of the TNC business model: the major ride-hailing companies are expending considerable resources on maximizing the use of these classes of service (Anand 2017, Heath 2016). Uber and Lyft have both experimented with variations on more constrained versions of their pooled services that work similarly to Via, using their regular network of drivers to offer lower-cost shared trips between limited, predetermined stops along a few set routes.2 To date, none of these offerings appears to have lasted beyond the testing phase in the largest markets. 2Between 2015 and 2017, Uber announced trials of products called uberHop, Smart Routes, and Express Pool, while Lyft announced HotSpots and Shuttle, all of which offered lower prices for shared rides along set routes in the downtowns of a few of their largest markets. As of October 2017, only Lyft Shuttle appeared to still be available.

12 Private Transit: Existing Services and Emerging Directions Prearranged Route- or Zone-Based Services (“Microtransit”) The term microtransit is used to describe prearranged services that, in contrast to the com- muter services arranged by employers, involve a direct transaction between the transportation service and the rider. Aside from the airporters, these services are generally confined to the commute market in both hours of operation and route or zone arrangement. These services operate during peak com- mute (or flight) hours and serve routes or zones, often with a limited set of possible stops, with one endpoint generally at a major employment center. These rides must be arranged and paid for in advance (lead time can range from minutes to weeks), and specific stops and routes may be created in real time, depending on rider requests and traffic conditions. Airporters operate whenever flights are offered and usually have a defined service buffering around major airports. Competing Definitions of Microtransit Although in this report, the term microtransit is used in a narrow sense, the meaning of this relatively new term is still in flux. The various usages seem to cluster around three main senses, none of which has yet become dominant: • Private services that operate passenger vans or small cutaways and develop routes and stops in response to customer input and demand, generally with tech-enabled booking and payment. This is the narrow sense used in this report as well as in TCRP Research Report 188 (Feigon and Murphy 2016, 5) and TRB’s Special Report 319 (Committee for Review of Innovative Urban Mobility Services 2016, 17). • Public or private flexible services using vehicles larger than sedans but smaller than transit buses and including technology-driven route or stop flexibility. Los Angeles Metropolitan Transportation Authority (Metro) used the term in this sense—which includes existing demand-responsive formats already in use by many public transit agencies—in a solicita- tion announced in August 2017 for a vendor to provide both vehicles and technology for a demand-responsive transit service to be operated by Metro personnel. • At its most broad, almost any flexible service in the space between taxis/TNCs and fixed-route transit, regardless of the mix of public or private ownership or operation. As with the on-demand pooled services, most microtransit services require a credit card for payment and a smartphone for arranging the ride. Prices may vary by the intensity of demand, with discounts for off-peak rides and for customers who prepay for a larger number of rides. Because the commute-centered services operate in larger (six-plus passenger) vehicles, these ser- vices qualify for the use of pre-tax transit benefits and commuter check programs under Internal Revenue Service guidelines. Microtransit Operators Three microtransit operators are discussed herein: Chariot, Bridj, and the Bridj/KCATA RideKC program. Chariot, which was acquired by Ford Smart Mobility in September 2016, is a microtransit company operating routes in San Francisco (see Figure 1), Austin, and New York City as of October 2017. The company announced plans to expand to Seattle and several more cities by the end of 2017 (Bomey 2017). It uses 14-passenger vans, and drivers are employees of the company. Chariot establishes some of its routes under a “crowd-sourced/crowd-funded” model, where potential riders suggest new routes in the cities where it operates. Other users then vote to sup- port proposed routes by pledging to ride on them if they are established. If enough supporters (usually about 50) vote to create a given route, within a few weeks it is established on a trial basis.

Defining Private Transit Services 13 Routes with enough ongoing ridership become permanent “public” routes. The company also provides “private” routes that more closely resemble sponsored commuter services for specific employers, with the option to open the routes to other riders to help underwrite their cost. As of this writing, Chariot is the last remaining representative of a transportation business model that for several years was generating much interest and venture capital investment. Among the microtransit services that have begun and then ceased operations since 2012 are Bridj (based in Boston, MA), Split (based in Washington, DC), and Leap and Loup (based in the Bay Area) (TransitCenter 2017b, Keshani 2017, Gaus 2015, Cagle 2015). Bridj operated independently in Boston, and briefly in Washington, DC, before it shut down in April 2017. Starting in March 2016, Bridj operated a 1-year pilot service jointly with the Kansas City Area Transportation Authority (KCATA) in Kansas City, MO, and Johnson County, KS. Bridj operated within two zone pairs in Boston, between which riders could travel inbound in the morning and outbound in the evening, and one internal travel zone comprising several adjacent neighborhoods. The Bridj/KCATA RideKC service was a pilot program with KCATA that began in March 2016 and, as a public-private partnership that amounted to contracted service provi- sion, is somewhat beyond the scope of this study. However, it is included here because of the level of interest the partnership generated and the fact that it used KCATA’s union drivers to operate the service’s vans. The pilot did not meet ridership expectations. Although the RideKC Bridj service was projected to carry around 200 riders per day after its first 6 months (Zipkin 2016), it had provided fewer than 600 rides total by the 6-month mark, according to KCATA reports Figure 1. San Francisco routes and stops of Chariot microtransit service, October 2017. (Credit: Route data ©2017 Chariot, map data ©2017 Google)

14 Private Transit: Existing Services and Emerging Directions (TransitCenter 2017b), and fewer than 1,500 rides over its 12-month course (Marshall 2017). The Transportation Sustainability Research Center at the University of California, Berkeley, released a midpoint evaluation of the program (Shaheen et al. 2016). The results were limited by a small sample size (due in part to the very low ridership levels), but pointed to issues with marketing, high cost barriers, and a lack of public awareness about the service. Although KCATA did not continue a relationship with Bridj after the pilot, the authority incorporated the lessons learned, especially on marketing new service formats, into another on-demand service offering that launched in May 2017. Microtransit Challenges Of the business models reviewed for this study, microtransit appears to have the hardest time surviving as a viable private service model. While it’s impossible to understand exactly why these services are failing without knowing more about their operations and finances, there are a few possible explanations for their difficulty: • A capital-intensive business model. Even in the world of shared mobility, microtransit is a relatively young and untested concept, with only a few companies having attempted to compete in this space. So far, operators have relied on a business model that involves leas- ing vehicles and, for the most part, paying drivers as employees, as opposed to the relatively nimbler TNCs, which rely on contractors who supply the vehicles they drive. The more capital-intensive model used in microtransit may make it more difficult to expand and sustain compared with the very rapid growth that’s possible under the TNC model. • Too much similarity to transit to succeed without subsidy. Few public transit routes are profitable. In addition to farebox recovery, transit systems depend on funding from a variety of public sources to support their capital investments and operation. Microtransit operators provide a transit-like service, but at a lower volume and without public support. While private funding has helped to fill that gap, it is less dependable (as illustrated by Bridj’s failed talks with an automaker as a potential partner) and likely not a long-term solution. • Insufficient demand. In line with the previous reasons, it appears that this new service type is thus far appealing to so few riders (even within its relatively modest expectations) that most routes are unable to remain in operation long enough to establish the mode’s viability. This may change as the service becomes more widely understood and accepted outside of a few areas. Flexible Route-Based Services (Jitneys) This category includes all shared private transit services that can be used without prearrange- ment. This includes formal and informal and legal and illegal services that pick up passengers through systems of known fixed-route stops or by street hail. These services have many names— including jitneys, dollar vans, minibuses, raiteros, and camionetas—which vary depending on the service’s location and predominant user base. Although many jitneys operate on well-known routes, very few have fixed schedules. Most private jitneys in the United States rely on word of mouth or, at most, routes and schedules posted at major pick-up points such as the Port Author- ity Bus Station in Manhattan. Characteristics of Flexible Route-Based Services Jitneys are most common in U.S. metro areas with large immigrant populations (particu- larly from the Caribbean), but do not exclusively serve immigrant communities. Jitneys operate a largely fixed-route configuration and are the only private transit service that accepts street hails. Street hails are often illegal for services other than taxicabs, but this is nonetheless central to the business model. Most jitneys accept only cash and may charge fares either at a flat rate per ride or according to the distance traveled.

Defining Private Transit Services 15 Business relationships between vehicle owners and operators resemble those in the taxi- cab business, with drivers who are a mix of owner-operators, independent operators who lease vehicles daily or weekly from fleet owners, and employees of fleet companies. In New York City, commuter van drivers are required to possess a chauffeur license appropriate to the class of vehicle they are operating, along with undergoing fingerprinting, background checks, and drug testing. The most prominent U.S. systems exist in New York City (mostly in Queens and Brooklyn); Miami; Atlanta; Southern California; and between Hudson County, NJ, and Manhattan. Local transit services in Atlantic City, NJ, have long been provided by highly regulated and subsidized private jitneys. Houston, TX, has provisions for local jitney services and has developed a few operators in specific markets. Few formal studies exist for most of these systems, by either public authorities or scholars. Eric Goldwyn’s research on dollar vans in New York’s outer boroughs estimates that 100 licensed vans operate on Brooklyn’s Flatbush Avenue every day, a corridor that is also one of Brooklyn’s highest ridership bus routes (Goldwyn 2017). For every licensed van on Flatbush, he estimates that up to three vans are operating without licenses (Goldwyn interview 2017). Goldwyn esti- mates that 15,000 passengers, heavily drawn from West Indian and Caribbean immigrant com- munities, use the Flatbush dollar vans every day. (See Figure 2.) Flatbush Avenue is one of the most heavily trafficked jitney routes in New York, but is by no means the only route, as documented by Reiss (2014). Figure 2 shows Reiss’s map of the New York City jitney routes as they existed in 2014. In addition to the largely corridor-based services provided by the West Indian dollar vans, another group of vans provides express service between concentrations of Asian immigrants in the Chinatowns of Manhattan and the outer boroughs. New York Taxi and Limousine Commission (TLC) data indicate that as of mid-2017, 54 dif- ferent jitney services were licensed to operate a combined 443 commuter vans in the city. These numbers are only a fraction of the total number of vans, and even the licensed vehicles are often operating outside of regulatory bounds by working on Metropolitan Transportation Authority (MTA) routes and picking up street hails. There are no systematic data on the number of unlicensed vans in New York City or else- where in the United States, although estimates exist for a number of systems. A study com- missioned by the North Jersey Transportation Planning Authority (NJTPA 2011) found approximately 20 operators serving nine routes. Most of these vans travel into New York, which means they are regulated at the federal level under interstate travel laws and thus beyond the reach of local authorities like TLC on either side of the Hudson. Goldwyn’s (2017) and Reiss’s (2014) dollar van estimates, although they are anecdotal, are the closest thing to actual data that exist for the full picture of New York’s jitney market. Informal Systems Beyond the New York City Area Several informal systems have been described: • Miami-Dade County. A study administered by the county estimated in 1992 that Miami had 400 licensed and unlicensed jitneys serving 45,000 daily passengers, before the market was regulated to provide service on fixed routes (King and Goldwyn 2014). See more about the Miami-Dade jitneys below. • Chicago. A 2013 article on raitero services in Chicago’s Little Village neighborhood estimated that the services shuttled as many as 1,000 workers a day in about two dozen buses to work- sites in Chicagoland (Grabell 2013). • Atlanta. A 2005 news story on Atlanta’s Buford Highway jitneys estimated that one driver alone served 2,000 riders in a day (Ramos 2005).

16 Private Transit: Existing Services and Emerging Directions Scope estimates for the full extent of markets with large informal components are generally anecdotal and sporadic. King and Goldwyn (2014) estimated the scale, at their height, of legal jitney systems in several cities and years (see Figure 3) based partly on occasional official probes of the systems. These estimates are only the visible points in a system that appears to have ana- logues in many major U.S. cities and several rural immigrant communities. Legal Jitney Systems Beyond the New York City Area Below are brief summaries of other flexible route-based services that currently operate with official sanction in several jurisdictions. Miami-Dade County, Florida Jitneys in the Miami metro area have origins in, and continue to serve, transit-disadvantaged communities. As such, jitneys have acted as indicators for where public transit could support higher levels of service. At times, they have also played a role in disaster-recovery transportation. Figure 2. Major jitney corridors in the New York City area. (Reiss 2014)

Defining Private Transit Services 17 Currently, the Miami-Dade Municipal Code regulates jitney vehicles as “any motor vehicle having a maximum seating capacity of fifteen (15) or less, transporting passengers for compen- sation on a semi-fixed route between fixed terminals not on a fixed schedule basis.” Drivers must hold a chauffer’s license. Additionally, the designated routes cannot compete with rail or bus service, producing what is, by definition, a complementary set of services [Miami-Dade Code of Ordinances §31-102(j)]. Today, the jitneys appear largely to serve a Caribbean immigrant community and are statu- torily limited to operating route-based service on corridors that do not substantially duplicate frequent public transit routes. After a rapid expansion of unregulated jitneys in the early 1980s, regulatory efforts in the region focused largely on avoiding duplication of public transit routes as well as addressing vehicle safety. One successful effort was implemented in 1986, when the City of Hialeah partnered with a jitney company to provide first-/last-mile connections to the city’s Metrorail station (Corsa 2000, Figure 3. Selected legal jitney programs in U.S. cities. Daily passenger estimates refer to the size of the ridership at the time of each system’s greatest extent; only New York City is likely currently at this scale. [From Table 1 in King and Goldwyn (2014)]

18 Private Transit: Existing Services and Emerging Directions Hialeah 1998). This private service still exists and now provides reciprocal transfers with Miami- Dade Transit’s fare system (MDTA 2017). After Hurricane Andrew devastated the region in 1992, jitneys filled extra transit demand through a 1-year contract from the Federal Emergency Management Agency (FEMA). Under a $46 million grant agreement, four contracting companies were paid $28 per vehicle-hour to hire more than 220 jitney operators, who in turn received around $21 per hour to provide free service on 12 routes in South Dade County. Drivers were required to possess a chauffer’s license, insurance, and proof of a vehicle safety inspection. They also had to pass a criminal background check (MDTA 1994). At its height in 1993, the FEMA-funded program provided around 20,000 rides per day. This arrangement temporarily resolved the problem of unauthor- ized jitney operation by providing a predictable revenue stream for operators, largely in areas not serviced by frequent transit. Due to the jitneys’ absence from their typical corridors, it also provided insights into where the private network had been competing with public bus routes and where it had complemented them (Dade County Metropolitan Planning Organization 1993, MDTA 1994). The use of jitneys as an indicator for areas capable of supporting improved public transit ser- vices was borne out in several studies, where jitney demand was used to propose service improve- ments for transit that resulted in increased ridership (Dade County Metropolitan Planning Organization 1993, MDTA 1994, Cervero 1997). Atlantic City, New Jersey The Atlantic City jitney system has operated continuously since jitneys’ emergence in the early 20th century. The jitneys are a sanctioned private transit service, providing much of the local transit service, and operating in the city with occasional injections of public subsidy (Cohen 2015). The not-for-profit Atlantic City Jitney Association (ACJA), which has “unique and exclusive operating rights within certain areas of Atlantic City” (NJT 2016, 7), contracts to allow owner- operators to purchase uniform vehicles that seat 13, include wheelchair lifts, and are fueled by compressed natural gas. The jitney services have periodically received public grants to upgrade their fleets, which are statutorily limited to 190 vehicles in the city (Atlantic City Mun. Code §233-27, Kent 1996, DOE 2013). The jitneys have statutory requirements as in other jurisdictions, with the additional, some- what usual, stipulation that operators must also belong to the ACJA. Municipal code regulates the number of vehicles and specific locations of routes and stops and requires that jitneys run 24 hours a day, year-round, on most routes, as well as dictating fares and discounts. In times of “transportation crisis,” the city may direct jitney operations. The system was used in this emergency capacity during Hurricanes Irene (2011) and Sandy (2012) (Atlantic City Mun. Code §233-27, DOE 2013). In addition to four statutorily designated routes on which jitney operators charge a cash fare of $2.25, New Jersey Transit (NJT) has an operating agreement with the jitney association to run five free routes in the city from NJT’s Atlantic City Rail Terminal station at a cost to the agency of roughly $5,000 per day (NJT 2016, NJDOT/NJT 2008). While Atlantic City’s economy is largely based on gaming driven by visitors to the city, the jitney system covers more of the city than do NJT bus routes and serves both tourists and residents. Houston, Texas Houston currently has two licensed jitney companies that operate under standards like those in other cities. Both feature 9–15 passenger vehicles, bear required trade dress, and travel pre- determined routes, but not on a fixed schedule (Houston Department of Administration and

Defining Private Transit Services 19 Regulatory Affairs 2017). However, unlike the utilitarian transportation focus of other cities’ jitneys, those in Houston appear to serve primarily as evening circulators in the city’s enter- tainment districts, selling all-night tickets good for multiple rides between venues, rather than single-ride fares (Houston Department of Administration and Regulatory Affairs 2017, Gordon 2015). While this system appears to be built largely around transporting revelers around night- life areas and from the city’s sports venues, the services do fill a niche by operating on routes and at times that would generally not make for productive public transit service. Sponsored Services In private market services, generally riders pay directly; in sponsored services, generally an employer or property manager (or a group of them) contracts for the service and makes it avail- able to the riders either for free or for a heavily subsidized fare. Sponsored services are generally provided as a benefit or amenity for employees or tenants. These services may fill gaps in public transit service and cover regional trips that public transit agencies cannot provide. Frequently, sponsored services are offered to employees and tenants who would otherwise drive alone. These services often complement public transit by providing last-mile rides from commuter rail stations in low-density areas or by driving long distances spanning the territories of multiple transit agencies. Several of the employers and property managers interviewed for this research used the word “amenity” and explained that their services provided a competitive advantage over other employers or properties vying for the same employees or tenants. This was particularly the case for a younger demographic that is increasingly embracing urban living and rejecting auto- mobile commutes and the expenses of car ownership. In addition, sponsored services may be a response to very tight housing markets that force employees to live far from their workplaces. These services are often an important component in the fulfillment of TDM requirements in jurisdictions where sponsors are located. Little information is available about the services and ridership of sponsored services as a whole since operations are dispersed among many providers and the participants (both sponsors and vendors) are not eager to share details that they regard as trade secrets. Furthermore, in many jurisdictions, licensing and regulation of these services occurs only at the federal level, which means that they operate largely independent of oversight from, or coordination with, public agencies at the city or state level. A notable exception to this is in the Bay Area and San Francisco in particular, which have gone further than any other U.S. region in building cooperative structures for understanding and mitigating the impacts of these services on the public realm. (See Case Study I in Section 5.) Employer- and property-based commuter services are becoming a common amenity for large corporate campuses and high-end multi-unit residential developments. The services are generally offered in one of three forms: • Fixed-route, scheduled services that make the entire journey between home and workplace • Transit feeder (last-mile) services • Circulator services Circulator services are usually internal to an employment campus or operate between nearby business locations. Fixed-route, scheduled services and transit feeder (last-mile) services are usually provided between the employment center and a residential area, a major transit stop, or a park-n-ride. Property-based services usually connect a large residential site with common transit, shopping, entertainment, and work destinations. It is reasonable to assume that these services exist in most major U.S. cities, but without a census of employers there is no way to

20 Private Transit: Existing Services and Emerging Directions estimate how many vehicles operate or how many riders they serve. This is an important area for future research. A hybrid sponsored model, which seems to be more common in suburban environments, falls somewhere between the employer- and property-sponsored services, and often takes the form of a public-private partnership. This model is centered on groups of employers or property owners who voluntarily associate by way of an entity such as a transportation management association (TMA) or a business improvement district and contribute toward the sponsorship of transpor- tation in their area. Public entities, such as municipalities or TDM agencies, may also take part in these partnerships. The resulting consortiums may either contract directly with a private transportation provider or pay a subsidy to a public transit agency for serving their locations. The services are generally either a local circulator or a connection to high-capacity public transit or a combination of both, rather than a longer fixed route. Examples of consortium-sponsored services, in several different forms, may be found in many regions across the country. Since consortium-sponsored services are more distinguished by the underlying partnership than the transportation service itself, this model is explored at greater length in Case Study II. Sponsored services use a range of vehicle types, from sedans and small vans for intracampus services, to larger 14–30 passenger cutaways, to 30–40 passenger transit buses, and 40–60 passen- ger motor coaches. On longer routes, operators may deploy the very large double-decker class of motor coach, which can hold 60 or more seated passengers. These services often include onboard amenities such as Wi-Fi and power outlets so that employees can more productively use their commute time, which may last 2 or more hours one way on some Bay Area routes. Several private operators provided the researchers with information on their operations in metro areas around the country, including the following: • WeDriveU, which is typical of the large commuter charter operators for corporate clients. The company reports that it provides 5.5 million passenger trips annually (about 460,000 per month) on some 400 daily routes in several cities. WeDriveU currently operates in Boston, Austin, Portland, Seattle, Denver, the Bay Area, and Washington (DC) with 1,200 total employees. The company runs both fixed-route and feeder services and reported 70–80% capacity utilization3 for fixed-route services and 60–70% capacity utilization for feeder/last-mile ser- vices (WeDriveU interview 2017b). The company notes that many service sponsors, especially in the highly competitive Bay Area market, prefer exclusive-use vehicles, in part so that their employees are not potentially mixing with competitors’ employees. • American Coach Lines, a subsidiary of Coach USA, which primarily operates in Atlanta, GA, and transports around 15,000 passengers a month. The company runs shuttles for employers or property managers in downtown Atlanta and suburban job centers (American Coach Lines interview 2017). Several of the suburban shuttles are last-mile routes connecting office parks to outlying rail stops in the Metropolitan Atlanta Rapid Transit Authority (MARTA) system, as well as a network of circulators and last-mile routes in dispersed suburban activity nodes, supported both by self-funding community improvement districts and by independent busi- nesses or groups of businesses. (See Case Study II in Section 5 for more on the consortium sponsorship model.) • MV Transportation, one of the largest operators of commute shuttle services, handling oper- ations for major employers in the Bay Area and Puget Sound regions, as well as providing 3“Capacity utilization” measures the utilization of available seats on the non-deadhead portion of trips.

Defining Private Transit Services 21 contracted fixed-route and paratransit services to public agencies and public-private partner- ships in many parts of the country. The company also provides turn-key shuttle operations for many consortium-sponsored services (MV Transportation interview 2017). Employer-Based Commuter Services The operators described above are among the dozens of companies hired to operate com- muter service for individual employers like Google, Microsoft, Genentech, Facebook, and Apple, all of which contract for extensive commuter shuttle networks to, and between, their corporate campuses. These contracts are commonly for turn-key operation, in which all operations and vehicle leasing/ownership are outsourced to a transportation provider. In what appears to be a less common model due to its much greater capital demands, the sponsoring companies may own or hold the leases to their own vehicles and contract only for operations and maintenance, a model followed by both Google and Microsoft. (It is unclear how many other employers purchase their own vehicles, but few companies would likely have the financial capacity for such a major capital investment.) Since at least 2012, Microsoft has operated some 80 buses on 22 routes for around 220 runs each day in the Puget Sound region (Lee 2017, Peterson 2012, Microsoft Connector 2017). Google has offices in many cities, but its largest commuter shuttle operation is in the San Francisco Bay Area. The Mountain View campus boards up to 9,000 passengers each way on 250 fixed-route commuter buses each day. (This figure equates to about 390,000 weekday boardings per month.) Google also runs an intercampus circulator that boards 3,000 passengers per day (about 65,000 per month)4 (Google interview 2017). The company contracts with five operators to drive the com- pany’s buses to a total of 120 stops on 800 daily fixed-route runs in the Bay Area. Google was one of 35 shuttle sponsors and operators that took part in the MTC and Bay Area Council’s Shuttle Census, published in 2016 (MTC 2016). The shuttle survey focused on com- muter and feeder services, excluding airporters and non-commute charters (many of which were run by the same companies providing the commuter shuttles). Table 2 summarizes the growth in shuttle ridership and capacity over the period covered by the census, 2012 to 2014. The growth of the fleet kept pace with passenger boardings, with average daily capacity growing from 18,000 to 34,000 seats during this period. Property-Based Services Not every company is a tech giant requiring an extensive shuttle fleet. More common is the offering of a property shuttle, which is shared by multiple tenants in a commercial property. Prudential Plaza and the Aon Center in downtown Chicago are two such examples. The adjacent 4Monthly figures were derived by multiplying daily boardings by 21.75. 2012 2013 2014 Annual boardings (thousands) 6,600 8,200 9,600 Avg. boardings/month (thousands) 550 683 800 Shuttle vehicles 473 536 765 Daily capacity (thousands) 18 23 34 Annual VMT (millions) 16 20 25 Table 2. Growth in private shuttle ridership and capacity in the Bay Area, 2012 to 2014. (MTC 2016)

22 Private Transit: Existing Services and Emerging Directions buildings’ property management companies jointly contract with bus operator SP Plus and electric bus manufacturer Proterra to lease all-electric feeder shuttle buses traveling to and from three downtown commuter rail stations and Prudential Plaza and the Aon Center at the opposite corner of the Loop. Ten vehicles in total, configured as 44-seat transit buses, serve about 3,000 riders daily from each building—roughly one-third of each building’s occupancy (Prudential interview 2017). As might be expected in Chicago, the ridership greatly varies by season with more people opting to walk in fair weather. The Prudential/Aon buses are one service among dozens that can be observed working a peak-hour circuit between rail terminals and workplaces around the Loop. In contrast to most commercial services, these sponsored services generally require no pre- arrangement or fare. They come on a regular schedule and usually require an ID badge or pass provided by the employer or property manager to board. Not every ride is free to the rider, often requiring at least some nominal fare. In some cases (as with Prudential/Aon in Chicago), the rides are subsidized on multiple levels, often differing from tenant to tenant. First, subsidies are offered by the property company, which provides the service to make the property more appeal- ing. Second, subsidies are offered by individual commercial tenants, that may or may not pass the cost onto employees. Any remaining cost is borne by the employees themselves through the purchase (via application or paper ticket) of daily fares at rates specific to each employer.