Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) (2019)

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.

17 C H A P T E R 4 This study includes 20 case studies of transit agency partnerships with TNCs based on the findings of the transit agency survey and interviews with key staff. Partnerships between transit agencies and TNCs are an active market; information captured here reflects the state of partnerships as of mid-2018. These case studies offer in-depth information not only on the parameters of the result- ing service offerings, but also insights into the process of developing and evaluating part- nerships. The Partnership Playbook (Chapter 6) synthesizes lessons learned from these 20 case studies into specific, practical guidance to enable transit agencies to be more pro- active in their approach to partnering with TNCs. Figure 4-1 summarizes highlights of each case study. Key takeaways include: • Twenty case studies represent slightly less than half of known transit agency–TNC partnerships as of early 2018, with research conducted through mid-2018. • To date, the transit industry has been more reactive than proactive in its approach to partnering with TNCs. • The vast majority of case studies are formal in nature (i.e., they involve an exchange of funds between transit agencies and TNCs) and there is significant variation in the amount and structure of ride subsidies. • Transit agencies express a strong desire to evaluate their programs, but rarely have developed an evaluation plan, and, lacking that, have been under-prepared in negotiating with TNCs in regards to data sharing. • Transit agencies often receive some data from TNCs, but it is almost always aggregated by time or geography. • Transit agencies often look at cost and ridership metrics as indicators of success, though there is a wide range of success metrics in the case studies. • Though partnering with TNCs is often suggested as a way to attract a new population of riders to transit, it is rare that transit agencies measure mode shift or net new ridership resulting from their pilots. • Unique or novel approaches in the case studies include: – Leveraging an external firm to evaluate a pilot so as to avoid a public records request of sensitive information; – Setting means-based eligibility for subsidies to target low-income communities; and – Noting positive publicity as an indicator of success. • It is not common but not unexpected to see transit agencies re-contract with a new TNC in the middle of a pilot. Case Studies

LA Metro MOD Pilot Los Angeles County, CA First Mile/ Last Mile Subsidized MOD Recipient Existing and new transit customers, vulnerable populations, geographic diversity Urban, Suburban Via (previously Lyft) Formal MOD Sandbox Grant, inter-regional collaboration, vendor transition Awarded the MOD Sandbox Grant to offer first mile/last mile services to transit stations TBD January 2018 – January 2019 Under development Yes FTA MOD Sandbox Grant Allocated: $1.35 million ($600,000 for Metro, $350,000 for Puget Sound, and $400,000 for research) Expended: ~$20,000 on research – as of Aug. 2018 LAVTA GoDublin! Livermore, CA Suburban Mobility First Mile/ Last Mile Subsidized Transit customers Suburban Lyft, Uber, DeSoto Cab Company Formal Multi-vendor, ADA provision Discounted shared TNC trips due to a comprehensive review that eliminated bus routes that did not meet productivity thresholds 50% subsidy (up to $5) for shared trips that start and end in Dublin January 2017 – June 2019 Monthly ridership Average cost per trip Origin/destination Yes Grant from the Alameda County Transportation Commission and LAVTA marketing funds – both stemmed from Measure BB funds (local sales tax) Allocated: $200,000 Expended: $60,000 – as of FY18 Transit Agency Partnership Name Location Partnership Intent & Design Target Market Service Area Context Vendor/ Partner(s) Type of Arrangement Key Themes Partnership Scope at-a-Glance Subsidy/Discount Duration Indicators of Success Access to Data Funding Source(s) Budget BBB Mobility on Demand Everyday (MODE) Santa Monica, CA Paratransit Subsidized Seniors and people with disabilities Urban Lyft Formal Cost savings, extensive marketing/outreach Replaced existing DAR program with on-demand service through Lyft Unlimited subsidized rides, customer copays $0.50 July 2018 – Present (permanent service offering) Increased utilization over fixed-fleet model Lower cost per ride Customer satisfaction Yes Local return state dollars Allocated: $600,000 annually Expended: $100,000 – as of Aug. 2018 CapMetro Exposition Area Innovation Zone Pilot Austin, TX First Mile/ Last Mile Subsidized Transit customers in an area with limited transit access Urban RideAustin Formal Non-profit TNC, robust data sharing Subsidized rides in an area that had no transit service due to the elimination of a local bus route Unlimited, fully subsidized rides to and from two bus stops June 2018 – December 2018 Overall cost of pilot vs. operating cost of eliminated route Cost per trip Overall ridership Geographic areas served Customer satisfaction Yes CapMetro General Fund Allocated: $24,900 Expended: $99 – as of Aug. 2018 CET None Central Oregon, OR Special Event/ Late Night Marketing Transit customers Suburban Uber Informal No contracting agreement Discounted rides home on First Fridays due to limited late night transit service Initially provided free rides up to $15, then transitioned to a 50% discount (up to $30 off) June 2017 – September 2017 None identified No, but received total trips per month N/A; no exchange of funds N/A CPTA dba RabbitTransit Paratransit for Seniors and People with Disabilities Central Pennsylvania, PA Paratransit Subsidized Seniors and people with disabilities Urban, Suburban, and Rural Lyft, Uber Formal Overflow paratransit demand Understanding which paratransit customers are appropriate for TNC service Supplemented existing paratransit services and providers with TNCs Fully subsidized rides when more cost-effective for CPTA than other options Mid-2017 – Present (permanent service offering) Monthly ridership Yes Pennsylvania’s Lottery-Supported Shared Ride Program funds Allocated: Part of funds allocated for subcontracted paratransit service. Used: $11,032 – as of FY17 DCTA Highland Village Lyft Discount Program Denton Region, TX Suburban Mobility First Mile/ Last Mile Subsidized Suburban residents in Highland Village with limited transit access Suburban Uber, Lyft Formal Vendor transition, ADA provision Provided first mile/last mile services in Highland Village through Uber; later replaced Uber with Lyft and is working to design new pilot services $2 subsidy for all trips within Highland Village and to/from the nearby hospital and A-train regional train October 2016 – Ongoing None Uber: No, but later received weekly ridership and number of unique participants Lyft: Yes DCTA's General Fund (local sales tax revenue from member cities) Uber Partnership Allocated: $20,000 Expended: $1,500 – as of Sep. 2018 Lyft Partnership Allocated: $20,000 Expended: $120 – as of Sep. 2018 GRTC Care On- Demand Richmond, VA Paratransit Subsidized ADA paratransit eligible customers Urban, Suburban UZURV and RoundTrip Formal Cost savings, use of “reservation network companies” to ensure provision of a call center Provided same-day or advance reservation service to ADA riders through reservation network companies Customers initially copays $6, GRTC subsidized up to the next $15, and customers pay any excess over $21 August 2017 – August 2018 (option to extend for an additional year) Percentage of ADA trips taken on CARE On- Demand, goal of 10% Miles, hours and ridership Yes State and local funds Allocated: $224,000 Expended: $187,503 – as of June 2018

Pierce Transit Limited Access Connections Seattle Region, WA Suburban Mobility First Mile/ Last Mile Late Nite Existing transit customers, college students, and residents with limited transit access Suburban Lyft (initially selected Uber) Formal MOD Sandbox Grant, vendor transition Awarded the MOD Sandbox Grant to offer first mile/last mile and late night services. Pierce Transit initially selected Uber, but was unable to enter a contract during negotiation. Fully subsidized, limited to 48 rides per month May 2018 – May 2019 (or until funding runs out) Many indicators are tracked; including perception of transit service quality, ridership, new customers, and cost effectiveness Yes MOD grant (80%) and local match (20%) Allocated: $206,000 (MOD); $51,500 (local) Used: ~$12,500 – as of July 2018 Subsidized MOD Recipient PSTA Direct Connect, TD Late Shift, P4- MOD Pinellas County, FL First Mile/ Last Mile Late Nite Paratransit Transit customers, paratransit customers, late night workers Suburban Direct Connect: Uber, United Taxi, Wheelchair Transport TD Late Shift: Uber, United Taxi, Care Ride P4-MOD: Lyft, United Taxi, CareRide, Goin, Wheelchair Transport Formal Multi and Title VI provisions, means-based eligibility Direct Connect: first/last mile connections for general population TD Late Shift: late night commute for low-income shift workers P4-MOD: same- day, on-demand rides for paratransit riders Direct Connect: $5 subsidy, $25 for Wheelchair Transport rides TD Late Shift: $20/month for a transit pass and 25 rides P4-MOD: fully subsidized Direct Connect: February 2016 - present TD Late Shift: August 2016 - present Increasing ridership Lower response times Number of rides per month Number of unique users PSTA tracks NTD metrics for rides done on Wheelchair Transport, Care Ride, and United Taxi No, only from non- TNCs Direct Connect: PSTA operating cost savings from previous circulators ($40,000 per year; varies by year) TD Late Shift: Florida Commission for the Transportation Disadvantaged ($507,000; varies by year) P4-MOD: FTA MOD ($625,000) Allocated: The funding for Direct Connect and TD Late Shift vary each fiscal year. Expended: Unknown Subsidized MOD Recipient SacRT RT Station Link Sacramento, CA First Mile/ Last Mile Subsidized Transit customers Suburban Lyft, Uber, Yellow Cab Company Formal Multi-vendor, timing of program launch Provided first mile/last mile service to/from six designated light rail stations. $5 subsidy, limited to 10 rides per user October 2016 – March 2017 Ridership Customer/public feedback Yes, but only from one TNC partner Grant from the Sacramento Metropolitan Air Quality Management District and Sacramento Area Council of Governments Allocated: $50,000 Expended: $4,554 – as of Mar. 2017 MARTA None Atlanta Region, GA First Mile/ Last Mile Marketing Transit customers Urban Uber, Lyft Informal Marketing/outreach, cost savings, unique contracting agreement Provided first mile/last mile connectivity to and from MARTA rail stations TNCs offered discounts ranging from 20-50% July 2015 – Present None No N/A; no exchange of funds N/A MBTA The RIDE Boston Region, MA Paratransit Subsidized People with disabilities (customers of ADA Paratransit) Urban Uber, Lyft Formal Multi-stage pilot, multi-vendor, ADA and Title VI compliance Supplemented ADA paratransit service to reduce costs Rider copays $2 and MBTA subsidizes the next $13 (later increased to $40). Trip determined by average trips made in customer’s first six months. September 2016 – Present Monthly ridership Cost savings Customer mobility Yes MBTA Operating Budget Allocated: Unknown Expended: $2.2 million - as of Aug. 2018 NYCT Access-a- Ride E-hail New York City Region, MA Paratransit Subsidized People with disabilities (customers of ADA Paratransit) Urban Uber, For-Hire Vehicles, NYC Taxi and Limousine Commission, Verifone, Limosys Formal (under negotiation) Cost savings, ADA provision Supplemented paratransit service with “E-hail” services – contracted with taxis but still negotiating with Uber as of May 2018 TBD TBD TBD TBD Access-a-Ride (AAR) program TBD Omnitrans RIDE Taxi & Lyft Program San Bernardino, CA Paratransit Subsidized Seniors and people with disabilities Urban, Suburban Lyft, RIDE Taxi Formal Alternative same- day transportation option for seniors and people with disabilities Supplemented ADA paratransit service to reduce costs 50% subsidy in the form of a $40 purchase for $80 worth of rides per month July 2016 – Ongoing Monthly ridership Initially no, but later received ridership, date of trips, and number of riders Measure “I” county ½ cent sales tax, federal funds, JARC grants (federal) Allocated: Unknown Expended: $11,000 during CY 2017 Figure 4-1. Case study highlights.

SamTrans TNC Partnership Pilot (in planning) San Mateo County, CA First Mile/ Last Mile Subsidized Transit customers, commuters Suburban TBD Formal First mile/last mile connections to transit, replacing or supplementing low productivity fixed- route transit SamTrans is in preliminary discussions with TNCs and has completed a feasibility study, but no partnership has been established. TBD TBD Specific metrics would need to be developed once the partnership is finalized. They may include ridership shifts from paratransit and improving first mile/last mile connectivity to transit stations. TBD SamTrans operating budget Allocated: None to date. Expended: N/A Transit Agency Partnership Name Location Partnership Intent & Design Target Market Service Area Context Vendor/ Partner(s) Type of Arrangement Key Themes Partnership Scope at-a-Glance Subsidy/Discount Duration Indicators of Success Access to Data Funding Source(s) Budget SEPTA None Philadelphia, PA Suburban Mobility First Mile/ Last Mile Marketing People driving and parking at SEPTA rail stations Suburban Uber Informal Marketing partnership, limited data sharing, no exchange of funds Addressed first mile/last mile connectivity to suburban rail stations 40% discount, up to $10 May 27, 2016 – Sept 5, 2016 Net new SEPTA ridership Positive publicity No N/A; no exchange of funds N/A SORTA None Cincinnati, OH Guaranteed Ride Home Suburban Mobility Marketing Transit customers Urban Uber Formal Marketing/outreach, transit advertising Provided Uber free advertising on transit vehicles and the agency’s website in exchange for offering discounted rides to first- time customers 20% discount for first-time customers March 2016 – March 2017 Rides taken in CMAQ’s Guaranteed Ride Home program, trip origins and destinations by time of day No CMAQ (for GRH trips only) N/A, no exchange of funds outside of GRH STA First Mile/Last Mile Pilot Solano County, CA First Mile/ Last Mile Subsidized Employees at selected worksites within 2 to 5 miles of a rail station Suburban Lyft Formal Transit use for commuting, first mile/last mile service for targeted employers Provided first mile/last mile connections to the nearby Amtrak station for employees working in the Solano Business Park area Customer copays $2 and STA subsidies the remainder May 2017 - present Monthly ridership Yes California’s Transportation Fund for Clean Air (TFCA) Allocated: $100,000 Expended: $7,000 – as of Aug. 2018 WMATA Abilities-Ride - An Alternative to MetroAccess Washington, D.C. Region Paratransit Subsidized ADA paratransit customers Suburban Two local taxi companies Formal Same-day service for customers of ADA Paratransit, TNCs encouraged to participate, but none were selected Provided same-day ADA service through taxis Customer copays $5 and WMATA subsidizes the next $15, customer pays any amount over $20 September 2017 - present Average monthly ridership Number of customers Impact on ADA paratransit demand Yes WMATA’s ADA paratransit budget. Allocated: No funds specifically allocated for pilot. Expended: Unknown Figure 4-1. (Continued).

Case Studies 21 Big Blue Bus (BBB) MODE – Santa Monica, CA Partnership Development and Implementation Why Enter a Partnership? In 2017, BBB overhauled its DAR program by rebranding to MODE and providing on-demand DAR service. The previous DAR service was costly to operate, highly subsidized, and over-utilized during weekday mornings, yet under-utilized at all other times. The DAR service also required prior sched- uling of rides of between one and six days in advance. The purpose of MODE was to replace the existing service, improve customer experience, provide on-demand mobility options for DAR customers, and reduce the cost of DAR service for the transit agency. Partnership Development In 2017, BBB released an RFP for “On-Demand Transportation Services.” The purpose was to solicit a vendor to provide four on-demand services, two of which were optional: (1) DAR service for Santa Monica residents, (2) reser- vation and dispatch service using city-owned vehicles, (3) optional DAR service to the public, and (4) optional motor coach operator field relief service. While the two optional services have yet to be executed operationally, BBB selected Lyft to provide DAR service and Campus Saferide to provide reservation and dispatch service for unbanked and ADA customers. The agreement allocated a budget of $600,000 for one year, with three additional one-year renewal options of equal amount. The total four-year period cannot exceed $2.4 million in spending, and future year funding is contingent on Council budget approval. Basic Parameters In order to ease the onboarding process and transition for DAR customers, BBB pro- vided a temporarily overlap of service between the new and old program. Price, eligibility, and service area requirements through MODE are consistent with the previous DAR pro- gram. All trips cost $0.50; however, customers may travel with up to three companions at an additional cost of $0.50 each, or one personal car attendant for $0.25. Customers must be age 60 or older, or age 18 or older with a disability, registered through MODE, and certified by WISE & Healthy Aging. Existing DAR customers are required to reregister with BBB to participate in MODE. Non-smartphone customers can request a Lyft ride by phone, while unbanked customers can also request a Lyft ride by phone by depositing funds in an account held by BBB. ADA customers can reserve a BBB-operated vehicle by phone. The service area consists of the Santa Monica city limits, select shopping destinations in Venice, and three nearby major medical facilities. Trips are limited to the following hours of service: • Monday to Friday: 8:00 a.m. to 6:00 p.m. • Saturday: 8:30 a.m. to 3:30 p.m. • Sunday: 8:00 a.m. to 1:30 p.m. Case Study Highlights Dial-a-Ride Subsidized Target Market Seniors and people with disabilities Service Area Context Urban Vendor/Partner(s) Lyft Type of Arrangement Formal Key Themes Cost savings, extensive marketing/outreach Partnership Scope at-a-Glance BBB is a municipal bus operator that serves Santa Monica and adjacent neighborhoods in Los Angeles. BBB replaced its existing DAR program with on-demand service through Lyft. Duration July 2018 – Present (permanent service offering) Indicators of Success Increased utilization over fixed-fleet model Lower cost per ride Customer satisfaction Funding Source(s) Local return state dollars Budget As of Aug. 2018 Allocated: $600,000 annually Expended: $100,000 (~$50,00 monthly)

22 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Partnership Details Data Sharing • BBB receives enough aggregated data from Lyft to meet NTD requirements. It receives frequent destinations and peak travel times, and individual trip data is available, but on a geographically aggregated basis. • Unbanked, and non-smartphone trips reserved through Saferide by phone will provide more fine-grained data for individual trips such as time of day and trip origin and destination. • Program evaluation will be informed by data received from Lyft and customer surveys. Indicators of success include higher ridership and lower cost per ride compared with the previous DAR program, as well as customer satisfaction. Regulatory Considerations • BBB was able to address ADA and Title VI concerns by providing service through the transit agency itself. Customers can request a WAV or pay with cash on BBB-operated vehicles. Marketing/Outreach Methods • BBB engaged in a robust outreach process to promote the program. Outreach methods included in-person outreach events, physical mailers of informational material, and ads through multiple mediums (e.g., news, on transit, etc.) • BBB staff scheduled one-on-one in-person meetings with all 3,000 existing DAR customers to assist with MODE registration and use. BBB fully subsidized transportation to these meetings through Lyft to familiarize customers with the new service. Project Wins and Areas of Further Development Project Wins • Local engagement and support from senior and disability commissions have helped promote the program and engage new and existing customers. Areas of Further Development • The program has highlighted a need for door-to-door assistance. Nearly 25% of customers report needing occasional assistance. Many people who do not use wheelchairs request WAVs because they need some level of assistance from the driver such as loading groceries, folding a walker, or getting into and out of a standard sized vehicle. Documenting WAV requests does not fully illuminate the situation when switching to a TNC model. • Approximately 50% of the customers do not have smartphones, which has made transition to Lyft more challenging. Though Lyft’s call center and dispatch solution—Lyft Concierge— offers a solution for these riders, it is not a feature of this pilot. Three-quarters of these cus- tomers live below the poverty line, which indicates issues with smartphone ownership beyond those that are cultural and cognitive. Because nearly 80% of all rides in the program are booked with Lyft vehicles, streamlining the reservation process is key to success. Lessons Learned • The transition period into using TNCs is more challenging with older adults and people with disabilities. In addition to meeting with customers individually, BBB found it was critical to overlap MODE with the previous DAR service during the initial stages. • During outreach, BBB had to overcome challenges that were specific to the senior popula- tion. For example, the Congress of California Seniors (CCS) advises against credit card

Case Studies 23 payment. In addition, some customers already had Lyft accounts with poor rider ratings, which lowered their driver acceptance rate. BBB had to work with these customers to close their accounts, open a new one, and educate them on good customer behavior to maintain better ratings. • Non-smartphone customers can overwhelm call centers and experience difficulty identifying the correct vehicle assigned to them. • Although trips booked through Lyft are one-third the cost of traditional DAR trips, the MODE program may overwhelm budgets because customers can book as many rides as desired. In addition, BBB ridership has increased over 50% from FY 2016–17 to FY 2017–18.

24 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Capital Metro (Capmetro) Exposition Area Innovation Zone Pilot – Austin, TX Partnership Development and Implementation Why Enter a Partnership? In June 2014, Capital Metro (CapMetro) completed a service change plan that eliminated and rerouted several fixed bus routes that did not meet productivity thresholds. In 2017, a subsequent transit study focused on reorganizing bus routes and substituting low-performing service with six “Mobility Innovation Zones.” These innovation zones were envisioned to serve as testing grounds for alternative transportation pilots. The Exposition Innovation Zone was chosen as the first pilot due to the elimination of a bus route that left a one-mile stretch of Exposition Boulevard without transit service. As a result, CapMetro developed a first mile/last mile pilot to provide service between the Exposition Innovation Zone and two of the nearest bus stops. The pilot will run through December 2018, with the possibility of extension to June 2019. Partnership Development CapMetro issued a formal RFQ in May 2018, and all three TNC providers in the area (RideAustin, Uber, and Lyft) submitted proposals. After discussion with all three vendors, CapMetro selected RideAustin, a local non-profit TNC that was willing to meet CapMetro’s data-sharing requirements. Basic Parameters RideAustin provides unlimited, fully subsidized rides between the Exposi- tion Innovation Zone and two of the closest bus stops. The subsidy is auto- matically applied to weekday rides between 7 a.m. and 7 p.m. that begin and end in a geofenced service area. The service area consists of a one-square mile area in the innovation zone, and a quarter-mile radius surrounding each bus stop. Partnership Details Data Sharing • CapMetro receives exact origin and destination, distance traveled, trip cost, time of day, day of week, driver name, response time, and customer name for each ride from RideAustin. In a public records request, the transit agency would only reveal that a trip was taken and the time of day of that trip; all other details are protected. • CapMetro intends to conduct a program evaluation in December 2018. • CapMetro’s primary indicator of success is the comparison between the overall cost of the pilot to the cost of the eliminated bus route ($600 per day). Other indicators of interest include cost per trip, overall ridership, geographic areas served, and customer satisfaction. • Program evaluation will be informed by data collected from RideAustin, an online survey distributed by known participants, and customer feedback. Regulatory Considerations • WAVs are provided by CapMetro’s paratransit service. • Customers without smartphones are able to reserve rides by calling CapMetro’s dispatch system. Case Study Highlights First Mile/Last Mile Subsidized Target Market Transit customers in an area with limited transit access Service Area Context Urban Vendor/Partner(s) RideAustin Type of Arrangement Formal Key Themes Non-profit TNC, robust data sharing Partnership Scope at-a-Glance CapMetro provides public transit service to Austin. CapMetro partnered with RideAustin, a non-profit, to fully subsidize rides in an area that had no transit service due to the elimination of a local bus route. Duration June 2018 – December 2018 Indicators of Success Overall cost of pilot vs. operating cost of eliminated route Cost per trip Overall ridership Geographic areas served Customer satisfaction Funding Source(s) CapMetro General Fund Budget As of Aug. 2018 Allocated: $24,900 Expended: $99

Case Studies 25 Marketing/Outreach Methods • CapMetro conducts the majority of marketing and outreach in-house. Methods include local news media, flyers at bus stops, and public meetings at the local middle school and library. Project Wins and Areas of Further Development Project Wins • Having launched the first TNC pilot with RideAustin, CapMetro is now exploring options to expand to other innovation zones. Areas for Further Development • CapMetro experienced challenges with attracting more customers. As of August 2018, there was only one customer, which only costs the transit agency $6 per day. CapMetro was hopeful that parents or assigned guardians would use the service to accompany students to and from the nearby middle school. However, no students have used the service thus far. • CapMetro’s call center handles dialed-in trip requests. Dispatch representatives initially were not familiar with booking a trip with a TNC, which resulted in unanswered reservation calls and booking of luxury trips that did not qualify for the subsidy. Lessons Learned • Begin by determining the goal of the partnership (e.g., first mile/last mile, paratransit, etc.). • Provide sufficient lead time to develop the partnership and get the word out. CapMetro only had one month to develop and launch the partnership. • Prioritize and conduct extensive public outreach to understand travel patterns and determine what type of service is desired and suitable.

26 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Cascades East Transit (CET) of the Central Oregon Intergovernmental Council Central Oregon, OR Partnership Development and Implementation Why Enter a Partnership? CET had limited transit service due to limited access and short service hours. CET staff were thus interested in seeking alternative transportation options that would operate outside of service hours. Partnership Development Uber reached out to CET and expressed interest in entering a partnership almost immediately after it began service in the area. In May 2017, CET entered into a marketing collaboration with Uber for the four First Friday events held in the summer. The intent of the collaboration was to encourage attendees to take transit to First Fridays and provide another mobility option to return home out- side of CET service hours. CET had no financial agreements with Uber—Uber provided promotional material. Basic Parameters The initial collaboration provided a free ride after 4 p.m. of up to $15 for attendees to return home from First Fridays. As a means to inform customers of nearby bus stops, pickups were limited to designated bus stops. The intent was to provide customers a ride back home outside of service hours as well as to promote CET. After the first event, the subsequent First Friday promotion codes provided a 50% discount (up to $30 off) as opposed to an entirely free ride. Partnership Details Data Sharing • CET received data on the total number of trips provided each month. Twenty trips were com- pleted in August 2017, and none in September. Regulatory Considerations • A formal agreement was not signed. Because the pilot was more intended for marketing and promotional purposes, CET did not have to abide by any Title VI regulations. Marketing/Outreach • Uber assisted with marketing via social media and printed materials. CET also posted on social media and websites as well as distributed flyers to transit customers on buses. Project Wins and Areas of Further Development Project Wins • Having established a relationship with Uber already, CET now has a contact to explore other types of TNC partnerships. Case Study Highlights Special Event/Late Night Marketing Target Market Transit customers Service Area Context Suburban Vendor/Partner(s) Uber Type of Arrangement Informal Key Themes No contracting agreement Partnership Scope at-a-Glance CET provides public transit bus service to Deschutes, Crook, and Jefferson Counties in Central Oregon. CET partnered with Uber in 2017 to provide discounts to First Fridays in the summer. Duration June 2017 – September 2017 Indicators of Success None Funding Source(s) N/A; no exchange of funds Budget N/A

Case Studies 27 Areas of Further Development • The partnership did not seem to attract more ridership and may have been more effort than it was worth. CET did not offer the same promotion in 2018 due to lack of staffing, but is considering resuming the partnership in 2019. • The intent was to get people to ride the bus by promoting multiple modes, but the larger issue at hand is that many in the area are resistant to taking transit altogether. Lessons Learned • There is a general reluctance to ride transit in the area. • Providing a more direct incentive to transit customers such as only providing codes to existing transit customers, may have been more effective in promoting alternative mobility options.

28 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Central Pennsylvania Transportation Authority (CPTA) dba RabbitTransit Paratransit for Seniors and People with Disabilities – Central Pennsylvania, PA Partnership Development and Implementation Why Enter a Partnership? CPTA uses a network of various subcontracted providers that supplement its directly operated paratransit service, meeting demand across a large 10-county service area. With a philosophy of continually exploring mobility solutions to expand options for its customers, the transit agency reached out to both Lyft and Uber to see if the TNCs might fit as additional providers for its paratransit provider network. Partnership Development CPTA worked with both Lyft and Uber to establish service agreements. At the outset, CPTA spent considerable efforts to allow customers to provide a copay for part of the trip fee. However, the TNC model did not allow a copay option. Basic Parameters CPTA uses a specific funding source for the TNC trips—Pennsylvania’s Shared Ride Program, which is funded through the Commonwealth’s lottery. This funding program does not involve all the requirements of FTA funding, such as drug and alcohol testing. CPTA realized the TNCs would not be able to meet all FTA requirements. The transit agency schedules Lyft and Uber trips directly on behalf of cus- tomers when (1) the TNC trip is cost effective, that is, equal to or less than reimbursement from the Commonwealth’s Shared Ride Program and (2) the customer is ambulatory, can travel without supervision, and understands that an un-branded private vehicle will arrive for the trip. The transit agency provides capacity in its scheduling software for Lyft and Uber. When the transit agency determines that a particular trip meets the requirements for a TNC trip, the dispatcher requests the trip for the customer, using a desktop application for a Lyft trip or a smartphone for an Uber trip. When a TNC driver accepts the trip, the dispatcher contacts the customer with the trip information: “Today your trip is assigned to [Lyft/Uber], your driver is [name] and will arrive in a [car description, etc.].” CPTA has a corporate account with the TNCs and is billed at the end of the month for trips provided. Partnership Details Data Sharing • The TNCs provide data on trips they provided and “some data” on the trip particulars with the monthly billing statement. The transit agency would prefer to receive actual data for passenger miles and revenue hours, but it is not dependent on the TNCs for data needed for NTD reporting. The transit agency has the trip origin and destination data in its paratransit software and can generate data for revenue miles/time from mapping software. Case Study Highlights Paratransit Subsidized Target Market Seniors and people with disabilities registered for “Shared Ride” service, which includes several specialized programs (e.g., ADA paratransit, Medicaid non-emergency transportation, senior transportation) Service Area Context Urban, Suburban, and Rural Vendor/Partner(s) Lyft, Uber Type of Arrangement Formal Key Themes Overflow paratransit demand Understanding which paratransit customers are appropriate for TNC service Partnership Scope at-a-Glance CPTA partnered with Lyft and Uber as new providers to add to its network of subcontracted paratransit providers, which supplement the transit agency’s directly operated service. Duration Mid-2017 – Present (permanent service offering) Indictors of Success Monthly avg. ridership: rough estimate: 300–600 trips Funding Source(s) Pennsylvania’s Lottery-Supported Shared Ride Program funds Budget Allocated: Part of funds allocated for subcontracted paratransit service Used: $11,032 during FY17 for TNCs

Case Studies 29 Regulatory Considerations • CPTA knew the TNCs could not meet all the requirements for FTA funding, so the transit agency uses less restrictive state transit funds for these trips. • CPTA schedules its paratransit trip demand so that its directly operated paratransit fleet and drivers are used first. This ensures that the transit agency’s workforce is fully productive. Marketing/Outreach Methods • CPTA used local news media and social media to inform the public and customers at the outset of using Lyft and Uber for its Shared Ride service. There were at least a few members of the public who did not fully understand the planned role for TNCs and actually called the transit agency asking to schedule an Uber trip. Project Wins and Areas of Further Development Project Wins • The two TNCs provide another mobility option for the transit agency to use to meet peak demand for its paratransit service—“another tool in the toolbox.” Areas of Further Development • The transit agency continues to work on the customer copay option—trying to find a way so that customers provide a copay for their TNC trips. Lessons Learned • Given the transit agency’s large service area and demand for paratransit, CPTA relies on a multi-provider network of subcontractors to support its directly operated paratransit service. Lyft and Uber have proved to be useful additions to the network. • Using TNCs for public transit service can be facilitated by using a funding source that is less restrictive than FTA funding. • Only certain paratransit customers are appropriate for TNC service—those who are ambula- tory, can ride without supervision, and have no cognitive impairments. • The new service has received a mixed response from the customers: some “love it” and others “hate it,” with about a 50–50 split. Those on the negative side prefer the dedicated RabbitTransit service, with a branded vehicle and a driver whom they have come to know. Those customers also like the social aspect of riding with other seniors to their destinations. And if they have a complaint about the TNC service, they are not sure where or to whom they should complain. • If a transit agency wants to use TNCs, it should be willing to follow their rules. The TNCs have not been particularly flexible. CPTA understands and generally accepts the TNC business model. The transit agency has tried to change the model—by allowing customers to provide a copay for their trip—but so far has not been successful.

30 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Denton County Transportation Authority (DCTA) Highland Village Lyft Discount Program – Denton Region, TX Partnership Development and Implementation Why Enter a Partnership? In 2016, the DCTA completed a comprehensive operational analysis of its two fixed-route bus networks. These networks served two of DCTA’s three member cities (Denton and Lewisville), but provided limited access to Highland Village— a lower density suburb with less demand for public transit. The DCTA board directed staff to explore additional mobility options for Highland Village resi- dents, including a potential TNC partnership. Partnership Development DCTA staff worked with Uber and the contractor conducting the bus network analysis to design a pilot partnership program, which began operation in October 2016. In January 2017, DCTA released an RFP for “On-Demand Rideshare Services.” The intention of the RFP was to procure a contractor that would help design and operate pilot on-demand services on an on-call basis. No funding was allo- cated in the initial RFP; the RFP was a qualification process to get vendors under contract to provide service proposals that would be implemented via a task order. DCTA selected Lyft and a local taxi company as contractors through this RFP process. Lyft replaced Uber as the vendor for the Highland Village first mile/last mile program, and is currently working with DCTA to design new pilot services. Basic Parameters DCTA’s initial Uber partnership program provided a $2 discount for all trips within Highland Village, as well as trips between Highland Village and a nearby hospital in Lewisville. Customers could use the service to connect with A-train regional rail service to Dallas and Denton or local bus service to Lewisville. Trip requests and payment were completed directly through the Uber app, which dis- played a DCTA menu option when customers were within the Highland Village pilot area. DCTA also continued to provide a directly operated on-demand service, which customers could request by phone at least two hours in advance, as well as ADA paratransit service. Lyft replaced Uber as the program vendor after being selected through the on-call RFP process. Partnership Details Data Sharing • DCTA’s initial contract with Uber did not include any data-sharing require- ments. Uber initially did not provide any data to DCTA, but eventually provided weekly ridership numbers, plus the number of unique customers that participated in the program. DCTA did not receive specific origin and destination trip data, but did receive ridership numbers in certain areas that helped the transit agency make decisions on expanding the discount zone. Case Study Highlights Suburban Mobility First Mile/Last Mile Subsidized Target Market Suburban residents in Highland Village with limited transit access Service Area Context Suburban Vendor/Partner(s) Uber, Lyft Type of Arrangement Formal Key Themes Vendor transition, ADA provision Partnership Scope at-a-Glance DCTA is a suburban transit agency serving Denton, Lewisville, and Highland Village, and facilitating transportation options in several adjacent counties and municipalities. In 2016, DCTA contracted with Uber to provide first mile/last mile services in Highland Village. In 2017, DCTA released a RFP for on-call transportation pilot services—leading to a contract with Lyft and expanded service options. Duration October 2016 – Ongoing Indicators of success None Funding Source(s) DCTA's General Fund (local sales tax revenue from member cities) Budget As of Sep. 2018 Uber Partnership Allocated: $20,000 Expended: $1,500 Lyft Partnership Allocated: $20,000 Expended: $120

Case Studies 31 • DCTA’s on-call services RFP did not include a specific requirement for data sharing, but did include “the ability of [the] contractor to provide ridership reports and detailed data” as a proposal evaluation criterion. In specific task orders, Lyft agreed to provide data on a monthly basis using a template they created. • In partnership with Lyft, it receives monthly reports with anonymized origin and destination data for each trip at the Census Tract level, along with the weekday and general time period (e.g., midday and PM peak) for each trip. It also receives overall mileage and hours of service each month. Regulatory Considerations • DCTA has not used federal funding as a revenue source for its existing partnerships with TNCs, but will be exploring ways to use federal dollars in the future. • DCTA considers its Highland Village first mile/last mile program to be a premium service. The transit agency continues to provide paratransit service in Highland Village, which allows customers to book trips by phone and use cash for payments. Project Wins and Areas of Further Development Project Wins • DCTA provides an example of an early TNC partnership pilot that has transitioned to a more permanent service offering. Areas of Further Development • DCTA will continue to develop the program to meet additional market needs. As of August 27, 2018, to address an increase in regional fares and allow Highland Village residents to have internal trips almost fully subsidized, it had expanded the program to offer a sub- sidy of up to $10 for Lyft trips in the same zone. • DCTA also started using the program to replace the Highland Village Community On-Demand Program, which had previously been operated using DCTA fleet vehicles and was deemed inefficient. Lessons Learned • On-call contracts may be an effective mechanism for quickly deploying new TNC pilot services. • TNC partnerships provide a good alternative to traditional transit solutions for a lower density suburban environment.

32 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Greater Richmond Transit Company (GRTC) CARE On-Demand – Richmond, VA Partnership Development and Implementation Why Enter a Partnership? GRTC wanted to provide another mobility option for its ADA paratransit customers and also to help steer the increasing demand for ADA paratransit to a service more cost effective than ADA paratransit. Partnership Development GRTC reached out to several other transit agencies that had implemented ADA paratransit alternatives with TNCs to learn about their experience. Addi- tionally, one of the transit agency’s senior managers brought direct experience with a same-day taxi- and sedan-based service for ADA customers from a prior position in a different city. In developing its pilot, GRTC was very focused on ensuring compliance with ADA and Title VI requirements following FTA guidance. The transit agency con- tacted its FTA Regional Office for further guidance. GRTC was then deliberate in including ADA and Title VI requirements in its procurement document for the pilot, in particular, including a requirement for a call center and accessible vehicles. GRTC’s planning efforts explored use of taxis and the “traditional” TNCs for the same-day service. Taxis did not seem a feasible option given the local taxi industry, and the TNCs did not seem interested in providing a call center, a feature GRTC specifically required. Basic Parameters GRTC negotiated agreements with two “reservation network companies.” The first, UZURV (founded by two Uber drivers and based in Richmond) provides a call center and a mobile app. Since late 2017, UZURV has transitioned to a TNC, calling itself an “Adaptive TNC.” The company has characteristics of a “traditional” TNC but also meets certain ADA requirements, e.g., providing all drivers with ADA and sensitivity training and providing door-to-door service when needed by riders. The second company, RoundTrip (co-founded by a Richmond resident and Virginia Commonwealth University graduate), is a web-based reservation service and has agreements with transportation providers in its network to serve the requested trips, including over 15 with accessible vehicles. This second company is formally known, per Commonwealth of Virginia regulations, as a Transportation Broker/Dispatcher. ADA eligible customers can contact either of the two companies for a same-day trip at least two hours in advance. Trips can also be booked up to 30 or more days in advance. Customers can request favorite drivers, at least with UZURV. Service includes door-to-door assistance if needed by the customer. A personal care assistant (PCA) rides without charge, and companion(s)—on a space available basis—also ride without charge. The customer pays the first $6, with GRTC paying up to an additional $15. If the trip costs more than $21, the customer pays the remainder. The companies charge for trips based on mileage, and the cost to the customer is also subject to an additional charge for peak time of day demand. Case Study Highlights Paratransit Subsidized Target Market ADA paratransit eligible customers Service Area Context Urban, Suburban Vendor/Partner(s) Two hybrid TNCs—UZURV and RoundTrip Type of Arrangement Formal Key Themes Cost savings, use of “hybrid network companies” to ensure provision of a call center and aspects of ADA paratransit Partnership Scope at-a-Glance GRTC partnered with two hybrid network companies that provide transportation service, including accessible service. ADA riders can book same-day or advance reservation trips. Duration Pilot began August 1, 2017, for one- year period with option to extend for an additional one year. Indictors of Success Percentage of ADA trips taken on CARE On-Demand, goal of 10% Miles, hours, and ridership Funding Source(s) State and local funds Budget As of June 2018 Allocated: $224,000 Expended: $187,503

Case Studies 33 Partnership Details Data Sharing • GRTC identified data and documentation requirements in its RFP and reports that the two companies provide the required information on a monthly basis. The information is needed for reimbursement. • According to the RFP for the pilot, the data and documentation include, among others: – Customer information including the GRTC Client ID number – Number of companions – Full pick-up address with phone number – Destination address – Starting and ending mileage – Money collected from the customer and total trip cost – Daily dispatch log, which shows all program reservations taken, cancelled, dispatched, and no-shows – Complaint log, which identifies customer complaints received directly and through GRTC customer service as well as the complaint resolution • The main indicator of success is percentage of total trips by ADA customers taken on CARE On-Demand, with the original goal of 10%. As of April 2018, approximately 1,500–1,600 trips were completed, representing 7–8% of total ADA paratransit and pilot monthly ridership. GRTC also tracks miles, hours, and ridership. Regulatory Considerations • GRTC was careful to address the regulatory requirements of ADA and Title VI in the planning and design of its pilot. Marketing/Outreach Methods • GRTC’s marketing department has worked with the two companies to advertise and market the pilot. Efforts included information on the transit agency’s website and through social media as well as flyers and a mailing to ADA paratransit customers. • More recent marketing has been word of mouth in the community. Project Wins and Areas of Further Development Project Wins • GRTC is pleased with the pilot. • The pilot is diverting ADA paratransit demand and cost: GRTC’s cost for the pilot trips are a maximum of $15, while ADA paratransit trips cost, on average, about $30. Areas of Further Development • GRTC is looking at some possible synergies that may exist with the current ADA paratransit program and the pilot program. Lessons Learned • GRTC believes that the time it took to plan the pilot, including careful research to ensure compliance with ADA and Title VI requirements, and the additional time to conduct the procurement and contracting process were important in launching a successful pilot. As the transit agency stated, “We did our homework.”

34 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) • The transit agency also reports that its pilot benefited from a slow and gradual build-up of ridership, allowing GRTC to gain experience and learn from that early experience. • Information for eligible customers about the cost of trips should have been more explicit, specifi- cally that the cost is not determined only by trip mileage but it can include an additional charge at certain times due to increased demand. There was some confusion in the early months about trip costs as customers did not understand that the cost for the same trip could vary.

Case Studies 35 Los Angeles County Metropolitan Transportation Authority (Metro) MOD Pilot – Los Angeles County, CA Partnership Development and Implementation Why Enter a Partnership? Los Angeles County Metropolitan Transportation Authority (“Metro”) identified an acute need for enhanced first mile/last mile access, particularly as the regional rail/BRT network expands with new funding from the passage of Measure M. Despite a relatively strong bus network, Metro has struggled to provide high-performing transit service in low-density corridors and near stations with poor pedestrian and bicycle access. In October 2016, Metro was awarded, jointly with Sound Transit and King County Metro (Seattle area transit agencies), a MOD Sandbox Grant of $1.35 million from the FTA to provide enhanced first mile/last mile access to select transit stations. The part- nership aims to improve mobility by increasing ridership for the transit agen- cies and to test a new business model. The specific goals of the project are to: • Improve mobility by increasing ridership for the transit agency through the pilot service • Provide a reliable, high-quality customer experience • Ensure optimal utilization of pilot vehicles through efficient aggregation of customers • Ensure access for disadvantaged populations • Ensure access to ADA-compliant accessible vehicle service • Create cost efficiency for the transit agency and the contractor Partnership Development Eager to leverage emerging mobility solutions to improve transit agency operations, in late 2015, Metro created an internal Office of Extraordinary Innovation (OEI), which manages this partnership. OEI had previously engaged in a two-week marketing partnership with Uber, but currently has no actively deployed service delivery models with them. While Lyft had not yet formed any partnerships with transit agencies, it wanted to test true ser- vice delivery partnerships and share data. As such, Metro selected Lyft as its partner for the MOD grant with local partners (Access Services, Foothill Transit, and the Los Angeles Department of Transportation) and the two Puget Sound-area transit agencies. Though Lyft was included as a partner in Metro’s original MOD grant application, the two parties were unable to negoti- ate a mutually beneficial partnership contract once awarded the grant. The FTA allowed Metro to identify another partner that could deliver the accessibility and data-sharing components as originally proposed. In November 2017, Metro signed a new term sheet with Via, an on-demand shared rides provider, envisioning a project from January to June 2018. As of August 2018, Metro is negotiating terms and conditions with Via and, like the Puget Sound-area transit agencies, does not yet have a fully executed contract. The partnership will provide first mile/last mile connec- tions to rail and BRT stations in the Seattle and LA, metropolitan areas; in LA three stations and service zones of about six square miles were identified in collaboration between Metro and Via. Basic Parameters In a one-year pilot slated to start in early 2019, Metro (along with Sound Transit) will use its MOD funding to provide subsidized Via rides to customers whose origin or destination is one Case Study Highlights First Mile/Last Mile Subsidized MOD Recipient Target Market Existing and new transit customers, vulnerable populations, geographic diversity Service Area Context Urban, Suburban Vendor/Partner(s) Via (previously Lyft) Type of Arrangement Formal Key Themes MOD Sandbox Grant, inter-regional collaboration, vendor transition Partnership Scope at-a-Glance LA Metro is a regional transit agency serving Los Angeles County. In 2016, Metro was awarded the MOD Sandbox Grant to offer first mile/last mile services to transit stations. Metro initially selected Lyft, but was unable to enter a contract during negotiation. Metro has subsequently entered into a partnership with Via. Duration January 2018 – January 2019 Indicators of Success As part of ongoing negotiations with Via, Metro is developing specific key performance indicators associated with each of the project’s goals. Funding Source(s) FTA MOD Sandbox Grant Budget As of August 2018 Allocated: $1.35 million ($600,000 for Metro, $350,000 for Puget Sound, and $400,000 for research) Expended: Metro has expended about $20,000 on research tasks

36 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) of three selected rail/BRT stations: North Hollywood, Artesia, or El Monte. Customers will book shared rides in vehicles operated by independent contractors that pool customers who are going the same direction to each station. The exact subsidy for the shared rides has not yet been speci- fied; according to Metro, “The cost of the ride should be less than a similar privately operated and funded service.”12 Partnership Details Data Sharing • Data from the pilot will be shared, analyzed, and evaluated by UCLA’s Institute of Trans- portation Studies, the Eno Center for Transportation, and the University of Washington’s Transportation Research Center (TRAC). Research from these institutions will quantify potential benefits and challenges to incorporating TNC first mile/last mile services into public transit systems. This research will serve as a blueprint for transit agencies partnering with technology-enabled mobility companies. • Negotiations around data sharing with Via started with a term sheet—a non-binding agree- ment that specifies basic terms. A formal and binding data-sharing agreement is expected and being developed in ongoing negotiations. • Similar to other MOD grant recipients, a third-party consultant was hired by the FTA to evaluate the program based on data received from Via, customer surveys, and in-person counts conducted by Metro. Regulatory Considerations • Customers without access to smartphones will be able to call a call center to create a Via account and book a ride. Via will operate its own call center and subcontract to LionBridge, which is also a contractor for Metro, for non-English speakers. Via’s contract with LionBridge is entirely separate from Metro’s contract with LionBridge. • People with disabilities will be able to book WAVs. It remains to be determined whether Via will subcontract with a third-party accessible vehicle provider or whether Via can train and identify a set of independent drivers with accessible vehicles. • Unbanked customers will have the option to use prepaid debit cards, and Metro is exploring ways to make that option more accessible. • Metro believes that drug and alcohol testing of drivers, which is a requirement that conveys with FTA funding in certain circumstances,13,14 has been one barrier to implementation, though Metro expects the issue to be resolved with future state-level regulations. Within the MOD pilot with Via, Metro has required drug testing. Metro is focused on drug and alcohol policy as an impor- tant issue and is in support of identifying ways to effectively require drug and alcohol testing. Marketing/Outreach Methods • Metro believes that marketing has a significant impact on the level of awareness and adoption of services offered through transit agency/TNC partnerships. • Metro has not conducted any pre-marketing activities. A significant marketing and outreach campaign will launch when the service nears operation. Project Wins • The project offers a unique learning opportunity with peer transit agencies Sound Transit and King County Metro. • Metro and OEI have garnered significant media attention for the MOD project. • Most transit agency responsibilities involve existing system operations and maintenance, and rarely does any staff person or group have explicit responsibility for innovation. Creating the

Case Studies 37 OEI as a standalone office, with 12 full-time equivalents (FTE) as of August 2018, dedicated to innovation and allowed to take an iterative approach to projects, is seen as a critical factor in developing and carrying out the MOD pilot. • The MOD grant gave Metro justification to enter partnerships with mobility service providers. Without this context, a partnership of this nature may have generated political controversy due to the perception of Metro giving away public resources to a direct competitor. • Metro is setting a new framework for data sharing between public and private entities, enabling Metro to protect disaggregated data and personally identifiable information, which has been a challenge for many transit agencies to overcome. • Metro is creating a new business model, whereby Via is taking on operational risk of deliver- ing the service, utilizing principles used in public-private capital partnerships. • Metro has developed a justifiable methodology for selecting stations for these types of pilots and for creating service areas. • Metro has created a statement of work whereby the contractor is taking on full responsibility for the delivery of WAVs. Areas for Further Development • Metro is seeking data on how to best enforce regulations and plans to use OEI’s experiences to advocate for high-level regulatory changes at the regional, state, and federal levels. Lessons Learned • Be resilient and find a way to get to yes. We may be beyond the early days in which neither public nor private partner had experience working together and finding mutually acceptable terms. • Local, state, and federal regulations have significant bearing on the success of a partnership. Metro terminated its partnership with Lyft after the two parties were unable to agree on terms of regulatory compliance for accessibility of the service, among other concerns. • Exercise caution in entering into new partnerships to ensure that everyone’s key outcomes and needs are being met. Establishing a partnership agreement prior to garnering commitment to the transit agency’s expectations around data sharing and accessibility would be premature. • Remember that public transit agencies have significant power and public assets that can be leveraged in negotiations. Transit agencies can play an active role in shaping the terms of a partnership.

38 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Livermore Amador Valley Transit Authority (LAVTA) GoDublin! – Livermore, CA Partnership Development and Implementation Why Enter a Partnership? Due to declining bus ridership, LAVTA completed a comprehensive review of its Wheels fixed-route bus system in the summer of 2016. Bus routes that did not meet productivity thresholds were eliminated, leaving less populated areas in Dublin with limited transit service. LAVTA developed the GoDublin! pilot to pro- vide residents with affordable on-demand transportation and to increase access to the Dublin BART station and Wheels Transit Center. Partnership Development LAVTA initially explored other options, such as mini buses, to replace routes with low ridership. However, LAVTA ultimately entered a contract with Uber and Lyft, the only TNCs in the area, due to their popularity with customers. The DeSoto Cab Company, which has its own ride-hailing app, was selected as a third vendor to provide wheelchair-accessible rides and allow for cash payment and ride requests by phone. Although LAVTA intended to launch the pilot in August 2016 to coincide with service changes, contract negotiations delayed the launch by five months. The program was scheduled to end in June 2018, but was extended to June 2019 to allow more time to evaluate the program and analyze trends. Basic Parameters GoDublin! uses geofencing technology to provide a 50% discount (up to $5) for trips that start and end within Dublin city limits (a six-square-mile region). Only shared rides booked through UberPOOL, Lyft Line, or DeSoto Share are eligible for the promotion. Customers who do not have access to a smartphone, wish to pay in cash, or are in need of a WAV can request rides by calling the DeSoto Cab Company. Partnership Details Data Sharing • Each provider shares different datasets.15 While all data is anonymized, datasets differ in the level of detail and type of reporting. Due to the variation in data received, LAVTA has encoun- tered challenges with comprehensively evaluating the program. • LAVTA hired a consultant to evaluate the pilot and help determine if program continuation past June 2018 is warranted. • Indicators of success include average cost per trip, total ridership, and origin and destina- tion info. LAVTA staff had anticipated an average monthly ridership of 1,500 and trip cost of $5.00. Although the ridership estimate was initially met, technical complications with a provider’s discount code contributed to a significant decrease in ridership. Average ridership was approximately 1,000 and trip cost, $3.07. The most common trip pairings were to and from neighborhoods to BART stations, which confirmed that most customers were using the discount for its intended purpose of providing access to transit centers. • LAVTA hopes to obtain data on shared rides in order to determine the amount of trips that have multiple passengers traveling to a similar destination. Case Study Highlights Suburban Mobility First Mile/Last Mile Subsidized Target Market Transit customers Service Area Context Suburban Vendor/Partner(s) Lyft, Uber, DeSoto Cab Company Type of Arrangement Formal Key Themes Multi-vendor, ADA provision Partnership Scope at-a-Glance LAVTA is a suburban Bay Area transit authority looking to provide residents an easy, affordable transit solution. Wheels (operator within LAVTA network) pays for 50% of the rider fare (up to $5) for shared TNC trips that start and end in Dublin on Uber, Lyft, and DeSoto Cab Company. Duration January 2017 – June 2019 Indicators of Success Monthly ridership Average cost per trip Origin/destination Funding Source(s) Grant from the Alameda County Transportation Commission and LAVTA marketing funds—both stemmed from Measure BB funds (local sales tax) Budget As of FY 2018 Allocated: $200,000 Expended: $60,000

Case Studies 39 Regulatory Considerations • LAVTA was able to address ADA and Title VI concerns by partnering with the DeSoto Cab Company, which has a call-in request system, can process cash payments, and has WAVs in its fleet. In initial testing, WAV service was not guaranteed, so LAVTA renegotiated its contract with the DeSoto Cab Company to guarantee service. The agreement stipulated that LAVTA would pay $225 per WAV trip as a way to ensure similar response times to Uber and Lyft. One year into the pilot, the transit agency had not received any complaints regarding the responsiveness of WAVs. • LAVTA’s labor unions initially raised a concern that the partnership would replace union workers, but the pilot did not reduce service hours in the fixed-route network. The goal was to focus the transit agency’s existing resources on routes with higher ridership and rely on TNCs to serve areas where LAVTA is not competitive. Marketing/Outreach Methods • Although Uber and Lyft offered to assist with outreach and provide marketing materials, they were resistant to showing their logos side by side. As a result, LAVTA primarily developed marketing and outreach materials, which show Uber, Lyft, and DeSoto’s logos together. Project Wins and Areas of Further Development Project Wins • Weekday ridership for LAVTA has increased since the pilot launch, though LAVTA is uncer- tain if this is due to the partnership or other improvements it made to the system concurrently. Areas of Further Development • Customers access the GoDublin! promotion by inputting a promotion code manually. Uber, Lyft, and DeSoto notified app users of the promotion through in-app messaging. As a result, LAVTA has often been removed from direct customer engagement. The transit agency is unable to determine whether participants of the GoDublin! promotion are existing TNC users or had been transit customers of discontinued transit service. Lessons Learned • Coordination between three different providers with varying platform specifications requires strong program management. • Marketing efforts require a notable time investment to coordinate. Preferences for logo usage, and outreach, for example, vary by company. • Though data privacy was not a motivation for hiring a third party to conduct program evalu- ation, doing so could result in a more meaningful analysis as private firms may not subject to public records requests and TNCs could therefore be more willing to share data.

40 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Metropolitan Atlanta Rapid Transit Authority (MARTA) Atlanta Region, GA Partnership Development and Implementation Why Enter a Partnership? In a region experiencing high population growth but a decline in ridership, the Metropolitan Atlanta Rapid Transit Authority (MARTA) began exploring innova- tive ways to market their services and provide access to destinations out of reach by transit. MARTA initially wanted to leverage TNCs to provide customers with access to MARTA rail stations so they could complete their trips to the airport by transit. Partnership Development MARTA initially entered into a written agreement with Uber, with specific marketing commitments for both parties. After the agreement with Uber had ended, Lyft approached MARTA and entered into a similar partnership. MARTA and the TNCs worked together to create co-promotional visuals and common language to advertise the partnership and its benefits. While marketing was more extensive at the outset of the partnership, it evolved into social media and internet advertising. TNCs bought advertising from MARTA’s external advertising vendor. How- ever, there was no coordination between the external advertising vendor and MARTA’s co-marketing effort. To remain flexible and keep up with the need for quick implementation, MARTA and the TNCs typically make agreements by email and without a formal contract. MARTA had no financial agreements or contracts with the TNCs. Promotions were fully financed by the TNCs, who regularly communicated with MARTA about new promotional opportunities. Uber and Lyft were aware of the concurrent partnerships with MARTA. Basic Parameters MARTA’s partnerships with Uber and Lyft included discounts, ranging from 20 to 50%, for trips that began or ended at MARTA rail stations. The discounts were intended to encourage transit use to destinations such as the Hartsfield-Jackson Atlanta International Airport. The “MARTA on the Go” app also provides access to the Uber app platform. MARTA previously ran a promotion with Uber to provide a 50% discount on trips to and from the Mercedes Benz stadium, where the professional football and soccer teams play. At the time of this study, this offer had ended. However, MARTA partnered with Lyft to provide official drop-off and pick-up areas around the stadium. Trip requests and payment were completed through Lyft’s app platform. Partnership Details Data Sharing • MARTA’s contract with Uber and Lyft did not include any data-sharing requirements. It does not receive any trip data from either TNC. Case Study Highlights First Mile/Last Mile Marketing Target Market Transit customers Service Area Context Urban Vendor/Partner(s) Uber, Lyft Type of Arrangement Informal Key Themes Marketing/outreach, cost savings, unique contracting agreement Partnership Scope at-a-Glance MARTA is the public transit agency for the Atlanta metropolitan area, operating mostly in Fulton, Clayton, and DeKalb counties. In July 2015, MARTA partnered with Uber and Lyft to provide first mile/last mile connectivity to MARTA rail stations. Duration July 2015 – Present Indicators of Success None Funding Source(s) N/A; no exchange of funds Budget N/A

Case Studies 41 Marketing Methods • MARTA believes success with this type of partnership is largely dependent on a transit agency’s ability to engage in co-promotional efforts with simple informal agreements and a strong relationship with the TNC’s local representative. Project Wins and Areas of Further Development Project Wins • Customers were impressed with MARTA’s branding and commended the transit agency for being “current” and “moving with the times.” Areas for Further Development • MARTA would like to continue working on the following areas: – Technology integration with MARTA’s app platforms – Provisions for data sharing – Utilizing TNCs to strategically fill service delivery gaps Lessons Learned • Data is crucial. MARTA did not have access to concrete data to support whether the part- nerships have driven ridership. • Partnership outcomes are very dependent on the relationship with local TNC representatives. • Difficult to translate lessons learned from city to city; highly dependent on local relationships and ability for the transit agency to engage in co-promotion with simple informal agreements.

42 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Massachusetts Bay Transportation Authority (MBTA) The RIDE – Boston Region, MA Partnership Development and Implementation Why Enter a Partnership? In September 2016, the MBTA initiated a pilot through Lyft and Uber to provide an alternative to its traditional ADA paratransit service known as The RIDE. The pilot was intended to reduce the overall cost of providing paratransit, while also providing an on-demand transportation option. The MBTA has experimented with various copay amounts, subsidy caps, and trip caps, with the intention of understanding how program design affects overall cost and customer mobility. Partnership Development The transit agency released an RFP in summer 2016, to which both Uber and Lyft responded. After several months of negotiation and pilot development— focusing primarily on providing WAVs and services for those who do not have access to smartphones or credit cards. Both Uber and Lyft launched the pilot service in September 2016. Basic Parameters Customers who opted into the pilot could choose to use either Uber or Lyft for on-demand trips, providing an alternative to The RIDE’s one to seven day advance reservation service. Customers sign up for the pilot via Uber or Lyft’s website.16 Applications are verified by the MBTA, and then the pilot program is activated to the customer’s account by either Lyft or Uber. Customers can make any trip within The RIDE service area during The RIDE service hours, with certain restrictions, and the customer’s copays are automatically adjusted within the app. Both Uber and Lyft have worked with MBTA and/ or third-party vendors to provide trip request, service, and payment options that are ADA and Title VI compliant. The MBTA’s overarching goal for the pilot was to reduce the cost of providing ADA para- transit by shifting trips to lower cost TNC services, while increasing the mobility of its customers. The transit agency entered into the pilot with the intention of testing various service models, understanding that providing an on-demand, rather than next-day, service would likely induce additional trips. The MBTA also shifted from paying its paratransit vendors per trip to paying vendors per revenue hour during the pilot period, which made it more advantageous to shift longer trips, rather than any trip, from a dedicated paratransit vehicle to a TNC service. In chronological order, the MBTA used the following service designs: • Four hundred RIDE customers were initially allowed to sign up for the pilot, and were allowed to take an unlimited number of Uber or Lyft trips per month. Customers paid a $2 initial copay for each trip (less than the $3.15 or $5.25 per trip fare for The RIDE), the MBTA sub- sidized the next $13, and the customer paid for any costs above $15. All trips were served by either UberX or Lyft, meaning that customers did not have to share rides. • The MBTA found that some pilot participants that had previously only made a few paratransit trips each month began to take many more trips on the alternative TNC service. To limit the risk of subsidizing a significant number of new trips, the MBTA implemented a 20 trip per Case Study Highlights ADA Paratransit Subsidized Target Market People with disabilities (ADA) Service Area Context Urban Vendor/Partner(s) Uber, Lyft Type of Arrangement Formal Key Themes Multi-stage pilot, multi-vendor, ADA and Title VI compliance Partnership Scope at-a-Glance MBTA is the primary public transit agency in Greater Boston. In early 2016, the MBTA began investigating the potential for using TNCs to reduce the overall cost of providing its ADA paratransit service—known as The RIDE. Duration September 2016 – Present Indicators of Success Monthly average ridership Cost savings Customer mobility Funding Source(s) MBTA Operating Budget Budget As of Aug. 2018 Allocated: Unknown Expended: $2.2 million

Case Studies 43 month cap. This was later revised to 2, 20, or 25 trips and then 2, 10, 20, 30, or 40 trips, based on the number of RIDE trips each customer had made on average over the past 6 months prior to their joining the pilot. The pilot was also opened up to all RIDE customers during this period. • After the MBTA switched from paying its paratransit vendors on a per trip basis to per revenue hour, the maximum subsidy per TNC trip was raised from $13 to $40. The inten- tion of this shift was to encourage RIDE customers to take longer trips, which require more revenue hours and thus became higher cost, on TNC services than on traditional paratransit vehicles. • The MBTA recently introduced shared ride trips via UberPOOL as a service option for RIDE customers. Trips on UberPOOL cost $1, rather than $2, and also have a $40 trip subsidy. • In February 2018, the MBTA released interim results from the pilot program. It found that trips subsidized through the program cost an average of $15.27, compared with $40 per trip over the same time period. The pilot resulted in negligible overall cost savings, but a 43% increase in trips per month. By Q2 FY18, about half of all paratransit trips made by pilot cus- tomers were taken on The RIDE; overall, The RIDE currently provides about 150,000 trips per month in comparison with about 13,000 trips with the pilot providers. In other words, the pilot has not resulted in significant cost savings for the MBTA, but has allowed it to sig- nificantly increase the mobility of its customers. Partnership Details Data Sharing • The MBTA receives the following trip-by-trip data from Uber and Lyft, provided within monthly invoice reports: – The RIDE ID of the customer – Time and date of trip – Whether the trip was shared or solo – Customer cost – MBTA cost – Pick-up zip code – Drop-off zip code – Whether the trip was provided on a WAV Regulatory Considerations • During the procurement process, compliance with MBTA insurance requirements was a sticking point with Uber and Lyft. This issue was resolved after statewide regulation increased insurance requirements and mandated background checks. • The MBTA worked with Uber and Lyft to develop strategies to comply with ADA and Title VI regulations around trip requests, accessible service, and fare payment. Uber and Lyft took different approaches for each compliance focus area: – Trip requests: Lyft contracts with a call center that uses the Lyft Dispatch platform to book trips for customers without smartphones. Uber has provided a limited number of smart- phones free of charge to pilot customers to use for trip booking. – Accessible service: Lyft contracts with a third-party chair car provider for pilot customers that need a WAV. Uber provides WAV access by making arrangements with owner/operators of accessible taxicabs. Less than 2% of pilot trips are made using a WAV, even though roughly 12% of RIDE customers require a WAV, though this number has been increasing in recent months. – Customer copay payment: Both Uber and Lyft accept prepaid debit cards, gift cards, or PayPal for copay payments, which the MBTA views as satisfying Title VI requirements for customers without bank accounts.

44 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) • The RIDE’s service providers’ drivers are unionized. • Indicators of success include monthly average ridership, cost savings, and customer mobility. During Q2 of FY18, 7,989 trips were completed, nearly 50% of all RIDE trips. Project Wins and Areas of Further Development Project Wins • The TNC pilot has provided same-day, on-demand services for the MBTA’s ADA paratransit customers (and more trips overall) without a net increase in the cost of provided paratransit service. • Overall, pilot customers expressed higher levels of satisfaction with the pilot than with The RIDE across a range of factors—convenience, cost, reliability, comfort, travel time. On a scale of one to seven, the pilot typically beat The RIDE by two to three points. • Mobility for pilot participants was up for all types of trips—work, school, shopping, social, and medical. Areas for Further Development • The TNC pilot has not, however, resulted in a significant net decrease in the overall cost of providing ADA paratransit—which was one of the MBTA’s initial goals for the pilot. • Customers that require a WAV are using the TNC pilot service at a significantly lower rate than the MBTA’s paratransit service offerings overall. Lessons Learned • Customers that both regularly and infrequently use ADA paratransit will likely increase the number of trips they take each month when a same-day, on-demand service is made available through a TNC pilot. Monthly trip caps may be an effective mechanism for limiting a transit agency’s exposure to subsidizing additional trips. • Among the subset of customers who continue to take both The RIDE and Uber or Lyft, the most common reason for continuing to take The RIDE is that customers run out of their allotted pilot trips. • Having thorough ridership and cost data before and after the pilot allowed for a thorough comparative evaluation to assess the impact of ridership and operating costs.

Case Studies 45 New York City Transit (NYCT) Access-a-Ride E-Hail – New York City Region, NY Partnership Development and Implementation Why Enter a Partnership? In 2017, the Paratransit Division of NYCT began exploring ways to improve its Access-A-Ride (AAR) service to manage the growing demand. The existing AAR service was costly to operate, not user-friendly, and inefficient for same-day pick-up and shared rides. NYCT explored the use of E-hail services (TNCs and taxis) as a way to reduce operating costs, supplement existing service, improve customer experience, and provide flexible options. As of May 2018, the pilot pro- gram was in the development and testing phase. It should be noted that NYCT uses the term “E-hail” to refer to services provided by TNCs and traditional taxi companies. Partnership Development NYCT had previously implemented a broker service program, yellow taxi debit card program, and taxi/car service reimbursement to varying degrees of success. After receiving an unsolicited proposal from an E-hailing service, the transit agency garnered enough support from elected officials and the board of directors to move forward with pilot development. The pilot program used New York City’s (NYC) existing Green and Yellow Taxi service managed by the NYC Taxi and Limousine Commission (TLC). The web-based app that matched customers with drivers was provided by Verifone Transportation Systems. Uber had participated in the pilot proof of concept until August 2017. As of May 2018, NYCT was negotiating with Uber to reach a contract. Partnership Details Data Sharing • NYCT had a formal agreement or contract provision that required the E-hail services to pro- vide data. The public could also obtain the data (with some restrictions) through New York State’s Freedom of Information Law. • NYC TLC provided NYCT with access to data on a weekly/monthly basis and by request. Data included total trips subsidized/provided, number of unique customers, average cost and fare per trip, average trip length, data on trip time of day and day of week, percent of trips in WAVs, percent of trips requested using an alternative hailing option (e.g., call center, etc.), and trip origins and/or destinations by address or nearest intersection. Regulatory Considerations • All E-hail services under the Taxi Service Pilot were subject to TLC Rules and Regulations that included necessary liability insurance. NYCT’s contractor also provided additional umbrella insurance. • NYCT tailored the pilot program so that smartphones and credit cards were not required to use the service. The only valid method of payment was cash during the time of this study. Customers without a smartphone could dial-in to a call center to book trips. Case Study Highlights ADA Paratransit Subsidized Target Market People with disabilities (ADA) Service Area Context Urban Vendor/Partner(s) Uber, For-Hire Vehicles, NYC Taxi and Limousine Commission, Verifone Transportation Systems, Limosys Type of Arrangement Formal (under negotiation) Key Themes Cost savings, ADA provision Partnership Scope at-a-Glance NYCT is a municipal transit operator that serves New York City. The Paratransit Division of NYCT used E-hail services to improve the efficiency and operation of its Access-A- Ride (AAR) program. As of May 2018, it was still in negotiations with Uber and successfully negotiated contracts with Verifone to dispatch taxis. Duration Unknown (under negotiation) Indicators of Success Unknown Funding Mechanism The AAR program is funded by paratransit fares, the City’s Urban Tax, a subsidy from New York City, and an internal Metropolitan Transportation Authority (MTA) cross- subsidy. Budget Unknown

46 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) • No labor issues were reported, since E-hail drivers were regulated by a separate entity (For-Hire Vehicles), and Yellow/Green Cab drivers were regulated by TLC. Project Wins and Areas for Further Development Project Wins • Customer feedback was very positive primarily due to direct rides and flexibility. Areas for Further Development • NYCT reported challenges with fare collection, data access, and wait times. Lessons Learned • E-hail services resulted in increased demand for paratransit.

Case Studies 47 Omnitrans RIDE Taxi & Lyft Program – San Bernardino, CA Partnership Development and Implementation Why Enter a Partnership? Omnitrans provides a range of specialized transportation services for seniors and people with disabilities. As the transit agency is responsible for coordi- nating human service transportation with designation as the region’s CTSA (Consolidated Transportation Services Agency), its goal is to give seniors and people with disabilities service options beyond ADA paratransit. Other objectives include providing a same-day trip option and helping reduce the cost for ADA paratransit. Partnership Development Omnitrans has an arrangement with Lyft whereby eligible customers can take Lyft trips that the transit agency discounts. There is no formal agreement or contract for the arrangement. The transit agency also has a similar arrangement with a local taxi company. Basic Parameters Seniors age 62 and above and people with disabilities who live within the Omnitrans service area are eligible for the transit agency’s several specialized programs, which include RIDE Taxi and Lyft. Interested individuals complete a written application form to determine eligibility, which includes signing a “hold harmless” form where the applicant agrees that “being driven by others is an inherently dangerous activity . . . and I agree to indemnify and hold harmless Omnitrans, its officers . . . from any and all claims, losses and liabilities . . . arising out of or in any way connected with my participation in the [specialized transportation service] programs.” Trips through Lyft require the individual to have a smartphone and Lyft account. For those who need accessible service or do not have a smartphone (or bank account), the taxi service is available. To use Lyft, an eligible individual can purchase a code for $40 that gives $80 worth of Lyft transportation. The individual inputs the code into his or her personal Lyft account. Each month, the individual can purchase the same amount of Lyft service for $40 by calling the transit agency. Originally, the code had no expiration date but, with changes by Lyft as of April 2018, a code is valid only for the calendar quarter it is purchased. Customers are informed of the expiration date before they decide to purchase a code. To use the taxi service, an eligible new individual can buy $80 worth of taxi transportation loaded onto a debit card (which has a photo of the individual) for $40. The card is issued by the transit agency (obtained through CabConnect, a technology company supporting the ground transportation industry). Each month thereafter, the individual can “reload” the card for $80 worth of taxi service for $40 by contacting the transit agency. For both Lyft and taxi, eligible users can obtain the subsidy only once per month. Case Study Highlights Paratransit Subsidized Target Market Seniors and people with disabilities Service Area Context Urban, Suburban Vendor/Partner(s) Lyft and local taxi company Type of Arrangement Formal Key Themes Alternative same-day transportation option for seniors and people with disabilities Partnership Scope at-a-Glance Omnitrans is the public transit agency serving the San Bernardino Valley in Southern California. The transit agency subsidizes Lyft and taxi trips for eligible seniors (age 62+) and people with disabilities. Monthly subsidy cap of $40 per eligible participant. Duration July 2016–Ongoing Indictors of Success Monthly avg. ridership:151 trips (based on total program trips Oct–Dec. 2017) Funding Source(s) Measure “I”-Voter-approved county ½ cent sales tax, with portion dedicated to senior and persons with disabilities transportation improvements; federal JARC grants. Budget Allocated: Unknown Expended: $11,000 during CY 2017

48 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Partnership Details Data Sharing • Obtaining data from the TNC on the use of the TNC service has been difficult. However, this appears to be changing with the April 2018 revisions. With the new arrangement, Omnitrans is billed at the end of the quarter according to the use of Lyft trips by the customers during that time period. Along with the bill, the TNC has provided data on the number of trips provided, the date of the trips, and the number of customers who used the service. While the transit agency would prefer more detailed trip data, for example, data on trip origins and destinations and trip times (which would be useful for transit planning purposes, for example, exploring possible group trips), the transit agency appreciates receiving the new data provided. Regulatory Considerations • To ensure the same-day program is available to all, Omnitrans specifically includes a taxi company as a transportation provider, which allows trip requests by telephone and provides accessible service. Marketing/Outreach Methods • The program is marketed in several ways: through the transit agency’s website, the transit agency’s travel trainers, the ADA paratransit certification process, and also the regional Public and Specialized Transportation Advisory and Coordination Council. Project Wins and Areas of Further Development Project Wins • The transit agency is pleased with the program. • The customers have another option for mobility. • The program is diverting trips from more costly ADA paratransit service. • Though there is exchange of funds, the transit agency considers having no written agreement or contract an advantage. Areas of Further Development • The transit agency would like to see greater use of the taxi component of the program; appar- ently there are few trips on taxi. This suggests that those who are eligible and use wheelchairs are not taking advantage of the same-day accessible mobility option provided through the program. Lessons Learned • Obtaining data on program use from the TNC has been challenging, though this appears to be changing. • It is important to provide a program that is available to all eligible customers, allowing the customers to choose what is best for them.

Case Studies 49 Pierce Transit Limited Access Connections – Seattle Region, WA Partnership Development and Implementation Why Enter a Partnership? In 2016, Pierce Transit was awarded a grant for $205,922 through the MOD Sandbox Demonstration Program to implement a yearlong pilot partnership titled “Limited Access Connections.” The overarching goal was to increase transit con- nections by addressing challenges such as limited transit access to less dense areas, early service termination, long wait times, and highly utilized park-and-ride lots. Partnership Development At the time of the grant application, Uber was the only TNC operating in the area and selected as the vendor. However, Pierce Transit and Uber could not reach an agreement due to lack of agreement on the indemnification of guest customers. Pierce Transit subsequently reached out to Lyft and announced it as the new vendor for the Limited Access Connections program in May 2018. Pierce Transit also issued two different RFPs to solicit third-party providers to ensure accessibility, but failed to receive responses either time due to the uncer- tainty in the number of requests. Pierce Transit ultimately decided to provide accessible service itself through its own paratransit vehicles and drivers. Basic Parameters Pierce Transit provided free rides to customers in four specific zones within the transit agency’s service area. Each zone was issued a different code and could only be used at certain times and days of the week. Customers were limited to a maxi- mum of 48 rides per month. The four zones were: • Puyallup Commuter and Fife Commuter (Mon–Fri, 5 a.m.–7 p.m.): pro- vides commuter connection from East Puyallup and Fife to the Puyallup Sounder Station. • Guaranteed Ride Home (Mon–Fri, 8:30–10:30 p.m.): provides late night service for Pierce College Puyallup students to destinations within the East Puyallup zone and the South Hill Mall Transit Center. • Browns Point/NE Tacoma Connection (Mon–Sun, 9 a.m.–4 p.m.): provides connections to the Tacoma Dome Station multimodal facility when Route 63 is not operating. • Parkland/Spanaway Midland University Place (Mon–Sun, 5 a.m.–10:30 p.m.): connects pockets of densely populated areas that have limited transit service to the nearest transit service. Partnership Details Data Sharing • Pierce Transit received enough data to satisfy the FTA. Data includes general trip character- istics such as origin and destination by Census Tract level, trip duration in five-minute incre- ments, day of week, time period of day, trip distance in five-mile increments, promotional code zone, trip cost, and dispatch method (smartphone app vs. Lyft Concierge). • Pierce Transit had requested more granular origin and destination data, but Lyft was unable to provide data past the Census Tract level. Case Study Highlights First Mile/Last Mile Suburban Mobility Late Night Subsidized MOD Recipient Target Market Existing transit customers, college students, and residents with limited transit access Service Area Context Suburban Vendor/Partner(s) Lyft Type of Arrangement Formal Key Themes MOD Sandbox Grant, vendor transition Partnership Scope at-a-Glance Pierce Transit is a suburban transit agency serving Pierce County, Washington. In 2016, Pierce Transit was awarded the MOD Sandbox Grant to offer first mile/last mile, and late night services. Pierce Transit initially selected Uber, but was unable to enter a contract during negotiation. Pierce Transit consequently entered a partnership with Lyft. Duration May 2018 – May 2019 (or until funding runs out) Indicators of Success Many indicators are tracked; including perception of transit service quality, ridership, new customers, and cost effectiveness Funding Source(s) MOD grant (80%) and local match (20%) Budget As of July 2018 Allocated: $206,000 (MOD); $51,500 (local) Used: ~$12,500

50 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) • Though available, Pierce Transit elected to not receive passenger masked ID due to potential privacy concerns in the event that the transit agency could not protect such data from the Freedom of Information Act. • Similar to other MOD grant recipients, a third-party consultant was hired by the FTA to evaluate the program and submit findings to the U.S. DOT by June 2019. • The evaluation will be informed by data received from Lyft, rider surveys, Pierce Transit and Sound Transit ridership data, costs, Pierce College Puyallup enrollment, and stakeholder interviews. The evaluation plan17 determines success through many indicators that include, but are not limited to: – Improved perception of transit service – Increase in overall ridership on Pierce Transit – Customers of the new service riding transit more because of the new service – Increase in the number of people accessing the Sound Transit stations and Pierce Transit bus lines – Cost effectiveness – Reduced parking lot use – Reduced travel and wait times Regulatory Considerations • Non-smartphone customers can request Lyft rides online or by calling a number. • WAVs are provided by Pierce Transit. • Although drug/alcohol testing and background checks were a concern, the MOD grant was established with the understanding that TNCs did not have to comply with the FTA require- ments that are typically compulsory for transit agencies. Marketing/Outreach Methods • Pierce Transit collaborated with Lyft on marketing and outreach. Both parties released a joint press release to announce the pilot launch and posted on their respective Twitter feeds and websites. • Pierce Transit and Lyft also set up an outreach event at Pierce College Puyallup that was well attended. Lyft assisted with program promotion and also recruited drivers. Similar outreach was planned at community events such as farmers markets. • Sound Transit and Pierce College Puyallup also helped with program promotion to their respective audiences. • Lyft offered to do in-app messaging to notify customers of the pilot, but Pierce Transit wanted to refrain from doing so in the event that the transit agency would not be able to adequately respond to a large influx of new customers. Project Wins and Areas of Further Development Project Wins • Pierce Transit received a lot of support from the public, press, and various transit agencies in the region. Positive comments praise the transit agency for being “innovative.” • The collaborative outreach process between Pierce Transit, Lyft, Sound Transit, and Pierce College Puyallup helped garner excitement and support for the pilot. Areas of Further Development • Customer-service representatives required a more involved training process than anticipated. There was some initial hesitation and inaccuracies with information given to the public. Pierce Transit staff conducted a secondary training and issued periodic reminders to ensure that representatives were properly trained to field calls regarding the pilot.

Case Studies 51 Lessons Learned • Uber and Pierce Transit were unable to reach an agreement due to concerns around the indemnification of guest customers. • Pierce Transit staff questioned whether the pilot would have benefited and been implemented sooner with a more narrow focus by starting with one zone as opposed to four. • Monthly calls with other MOD Sandbox Grant recipients have revealed that other programs have been similarly delayed due to data-sharing concerns and contract negotiation.

52 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Pinellas Suncoast Transit Authority (PSTA) Direct Connect, TD Late Shift, P4-MOD – Pinellas County, FL Partnership Development and Implementation Why Enter a Partnership? A one-cent sales tax referendum to fund transit operations in Pinellas County failed to pass in 2014. The PSTA had been relying on reserves to fund basic operations, and as a result of the transit referendum’s failure, the transit agency was considering discontinuation of three underperform- ing local bus routes and consolidation of two others. In its initial “Direct Connect” program, PSTA sought partnerships with TNCs to provide more cost-effective first/last mile transit connections to areas that could poten- tially lose fixed-route service. A subsequent partnership, “TD Late Shift,” aimed to improve point-to-point mobility for low-income service workers who need transportation in late evening hours after PSTA’s fixed-route ser- vice stops operating. A third partnership in development (Public-Private- Partnership for Paratransit Mobility on Demand, or “P4-MOD”) aims to provide same-day, on-demand mobility for paratransit riders, at lower cost to PSTA, via a modernized central dispatch system. Partnership Development In its original zone-based “Direct Connect” partnership with Uber and United Taxi, PSTA paid up to $5 toward first-/last-mile Uber and United Taxi rides to and from selected bus stops or transit stations in zones within Pinellas Park and the East Lake area. The program began operating in February 2016 and expanded countywide in January 2017. In April 2018, PSTA removed the zones and added additional locations. PSTA subsequently began a second partnership in August 2016, called “TD Late Shift,” to provide unemployed or low-income residents up to 25 discounted Uber, taxi, or wheelchair transport rides to and from work per month when PSTA fixed-route service is unavailable. In its Public-Private-Partnership for Paratransit Mobility on Demand (P4-MOD) project, PSTA aims to improve mobility of paratransit cus- tomers and operate paratransit trips more cost effectively than its current Demand-Response Transportation (DART) program. According to the transit agency’s MOD grant application, PSTA currently spends $22.50 per ride on its current DART service, which utilizes nearly 10% of the transit agency’s operating budget. PSTA seeks to achieve these goals through cen- tralized dispatch technology that matches riders with TNCs, taxis, or wheelchair vans depending on the rider’s needs, estimated arrival time, and cost. Basic Parameters For Direct Connect, riders who hail Uber, United Taxi, or Wheelchair Transport rides with origins or destinations within 800 feet of one of the program’s 24 eligible transit stops/ stations receive a $5 discount. On Uber, the $5 discount is given when a person enters the code “UBER2PSTA” in the payment section of the Uber app. Initially at the program’s launch, a 50% discount up to $3 was available. The discount was later increased to $5. Riders who hail a ride from Wheelchair Transport receive a $25 discount. Case Study Highlights Vendor/Partner(s) Direct Connect: Uber, United Taxi, Wheelchair Transport TD Late Shift: Uber, United Taxi, Care Ride P4-MOD: Lyft, United Taxi, CareRide, Goin, Wheelchair Transport Type of Arrangement Formal Key Themes Multi-vendor, ADA and Title VI provisions, means-based eligibility Partnership Scope at-a- Glance Direct Connect: first/last mile connections for general population TD Late Shift: late night commute for low-income shift workers P4-MOD: same-day, on-demand rides for paratransit riders Duration Direct Connect: February 2016–present TD Late Shift: August 2016–present Indicators of Success Increasing ridership Lower response times Number of rides per month Number of unique users PSTA tracks NTD metrics for rides done on Wheelchair Transport, Care Ride, and United Taxi. Funding Source(s) Direct Connect: PSTA operating cost savings from previous circulators TD Late Shift: Center for Transportation Disadvantaged P4-MOD: FTA MOD Budget As of September 2018 Allocated: The funding for Direct Connect and TD Late Shift vary between each fiscal year. First Mile/Last Mile Suburban Mobility Late Night Subsidized MOD Recipient

Case Studies 53 TD Late Shift participants must pre-enroll with PSTA to become eligible for the program. Participants pay $11 per month for a discounted PSTA transit pass (regular value $70) and an additional $9 per month for up to 25 Uber, United Taxi, or Care Ride rides per month. Late Shift riders must earn no more than 150% of the federal poverty level, and discounted rides are only valid from 10 p.m. to 6 a.m., when PSTA fixed-route transit is not in service. Riders can only use the service to go between registered home and work addresses. The P4-MOD project, funded by the FTA’s Mobility on Demand grant program, will allow enrolled DART riders to hail same-day rides on-demand in lieu of the current system, which requires 24-hour advance reservations. Lyft is slated to operate on-demand rides, while Goin is set to do software integration between the integrated software platform and each provider. Goin takes the place of initial vendor TransLoc, which backed out of the project after deciding to take its business model in a different direction. The University of South Florida’s Center for Urban Transportation Research (CUTR) will assist with evaluation of project data for PSTA, while the FTA’s MOD program will evaluate the project with assistance from Booz Allen Hamilton. Partnership Details Data Sharing • Uber provides PSTA only the total number of monthly Direct Connect trips. Beginning in April 2018, PSTA began to receive total number of rides to and from each Direct Connect stop location on a quarterly basis. • United Taxi and Wheelchair Transport Service provide PSTA with trip-level data, including passenger name, origin, destination, response time, and total cost for each trip. • For the Late Shift Program, PSTA can access data available through the Uber for Business platform. This includes rider name, time and date of the trip request, cost of the trip, gratuity, vehicle type, and the city where the trip occurred. Regulatory/Policy Considerations • PSTA cannot currently report the TNC trips to the NTD, as they do not provide the data to meet NTD reporting requirements and cannot demonstrate that trips can be pooled. PSTA can report United Taxi, CareRide, and Wheelchair Transport Service trips to NTD. • Based on local policy, participants of the TD Late Shift program must make no more than 150% of the federal poverty threshold. • To comply with Title VI, Direct Connect or TD Late Shift riders who do not have access to smartphones or wish to pay in cash may call a 24/7 telephone hotline to request a cab from United Taxi. Additionally, the use of taxicabs that provided drug/alcohol testing and back- ground checks allowed PSTA to comply with FTA guidance. • To provide equivalent service for people with disabilities, Direct Connect and TD Late Shift riders with disabilities can request discounted rides on-demand from Wheelchair Transport and Care Ride, respectively. Marketing/Outreach Methods • Uber sent staff in early 2016 to Direct Connect services areas to do in-person outreach, with a focus on current PSTA riders. • In early 2018, Uber launched an in-app marketing campaign of the Direct Connect pro- gram to prepare for its countywide expansion. Two PSTA staff were dedicated to public outreach. • Additional outreach took place on PSTA’s website, social media, and through distribution of brochures to riders at transfer centers. Direct Connect is also advertised at eligible bus stops, with signage stating that Direct Connect service is available from several partners.

54 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) • PSTA made a video of how to use Uber on smartphones to publicize the partnership and help train people on how to use the Uber app. • TD Late Shift was advertised to Transportation Disadvantaged Program participants when they went to transfer centers to get their monthly discounted bus pass. Project Evaluation • PSTA did not identify specific performance metrics for the program, citing the fact that their programs were some of the first to be initiated. • Direct Connect effectively encourages short taxi/TNC trips, because if the trip is under one mile they have to pay little or nothing out of pocket at the current rate of PSTA subsidy. The goal is to connect with the PSTA fixed-route network. • Uber’s rates keep increasing, especially due to surge pricing. While Direct Connect’s initial subsidy was 50% of the ride up to $3, this subsidy was later increased to cover a larger portion of Uber’s minimum fare. • Care Ride, an initial provider of wheelchair-accessible rides under Direct Connect, pulled out of the partnership because it worried that providing on-demand rides would compromise its core line of business, providing scheduled paratransit rides. Project Wins and Areas of Further Development Project Wins • TD Late Shift is a highly popular program with 4,000 participants in Pinellas County. PSTA surveys Late Shift riders and has learned that participants have been able to get new jobs and work more shifts. In addition, people who previously walked or rode bikes late at night to get home from their jobs told PSTA they now feel safer. • Direct Connect was the first large-scale transit agency/TNC partnership, and PSTA received a great deal of positive media attention as a result. • Direct Connect has given PSTA about $40,000 per year in operating cost savings compared with the fixed-route circulators it replaced. Areas for Further Development • Now that Direct Connect has expanded countywide, PSTA needs to fully implement its public outreach plan to train existing riders on how to use the program to easily access the PSTA core and frequent local route network. PSTA also need to focus some outreach on attracting new riders to the fixed-route service as a result of having Direct Connect to get them there more directly. • Limited means of measuring first/last mile connections, as there is no means of determining whether riders are actually transferring from TNCs to fixed-route transit, as the program is intended. • Further app integration that allows for trip planning with multiple modes, including Direct Connect plus bus option. • Explore partnerships with other on-demand providers that will provide more data and provide shared ride service. Lessons Learned • It is difficult to get TNC’s to share the level of data needed to report to NTD and therefore count the trips toward ridership. PSTA tried to establish data-sharing agreements and per- formance metrics but the ultimate agreement included only limited data. Uber’s limited data sharing and NDA made measuring performance of Direct Connect and TD Late Shift difficult, and it has created problems for PSTA in being unable to report this ridership activity to NTD.

Case Studies 55 • Focusing on transportation disadvantaged communities (e.g., shift workers, seniors) is likely to have better outcomes than on discretionary riders, because these communities are more likely to use transit. • Multi-format marketing campaigns have significant effects on ridership; ridership in Direct Connect tripled after Uber staff performed in-person outreach during the program’s initial weeks. • Ensure options for unbanked riders, riders without smartphones, and riders with disabilities are in place before partnership launch.

56 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Sacramento Regional Transit (SacRT) RT Station Link – Sacramento, CA Partnership Development and Implementation Why Enter a Partnership? In 2016, SacRT and city officials sought to make travel easier to and from the new downtown sports arena, Golden 1 Center. The arena was scheduled to open in October, had limited parking supply, and was anticipated to increase congestion downtown. SacRT initiated the RT Station Link pilot to alleviate these concerns, encourage transit ridership, demonstrate innovation, and address a lack of late night service in the SacRT service area. Partnership Development SacRT issued a formal RFP in 2016 to solicit vendors to provide first mile/ last mile service to and from RT light rail stations. Lyft, Uber, and Yellow Cab Company were the only vendors to submit proposals and thus selected as partners. In the interest of increasing visibility, SacRT launched RT Station Link in conjunction with the grand opening of the Golden 1 Center. Basic Parameters Using geofencing technology, RT Station Link provided a maximum $5 dis- count for trips that started and ended within one mile of six designated light rail stations. The stations were chosen based on a number of factors, includ- ing presence of security personnel, geographic distribution throughout the system, and lack of major nearby destinations that customers might travel to instead of actually using light rail. The pilot lasted for six months and discounted up to 10 rides per customer. Partnership Details Data Sharing • One TNC partner provided trip manifests (i.e., date, time, origin, and destination), whereas the other two partners only provided total trips and a dollar-denominated invoice. Regulatory Considerations • SacRT partnered with the Yellow Cab Company to provide WAVs and serve customers who were unbanked and/or did not own smartphones. However, customers complained that WAV service was more expensive for the same trip. • Secretary Foxx’s letter, which highlighted key issues with TNC partnerships, such as ADA, Title VI, and labor union concerns, was an important factor in SacRT’s decision to end the program. Marketing/Outreach Methods • SacRT promoted the pilot with in-vehicle light rail advertisements (car cards) and a light rail vehicle wrap. • The TNC partners designed graphic material and handled most electronic promotions. The TNC partners were viewed as a better fit for these aspects of the marketing program. For example, Uber and Lyft had larger mailing lists and a more developed strategy that included regular e-blasts when the Golden 1 Center held events, and resending emails to recipients who did not open initial e-blasts. The TNCs also provided branded graphics for printed outreach materials, but were resistant to having both logos on the same promotional material. Case Study Highlights First Mile/Last Mile Subsidized Target Market Transit customers Service Area Context Suburban Vendor/Partner(s) Lyft, Uber, Yellow Cab Company Type of Arrangement Formal Key Themes Multi-vendor, timing of program launch Partnership Scope at-a-Glance SacRT is a regional transit agency that serves Sacramento. The RT Station Link provided discounted rides to/from six designated light rail stations. Duration October 2016 – March 2017 Indicators of Success Ridership Customer/public feedback Funding Source(s) Grant from the Sacramento Metropolitan Air Quality Management District and Sacramento Area Council of Governments Budget As of Mar. 2017 Allocated: $50,000 Expended: $4,554, plus promotional consideration

Case Studies 57 Project Wins and Areas of Further Development Project Wins • Customers were initially required to input a promotional code into the TNC platforms to apply the discount. However, SacRT opted to apply the discount automatically to attract more customers. Areas for Further Development • RT Station Link did not garner strong support from the public. This was in part due to general disinterest in the midst of other Golden 1 Center grand opening events as well as the lack of dedicated SacRT staff to oversee the extension of the pilot. Lessons Learned • Timing of partnership launch is key. Although the pilot launch with the Golden 1 Center grand opening was intended to attract more attention, this strategy was actually counter- productive. The local news cycle was dominated by general Golden 1 Center and Sacramento Kings news rather than the Station Link program. • Do not be overly restrictive with location, date, and time of day when developing basic parameters. SacRT originally intended to restrict the program to specific dates and times (on days of major Golden 1 Center events) but ultimately decided against date and time restrictions due to the belief that doing so would complicate and confuse the customer experience.

58 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) San Mateo County Transit District (SamTrans) TNC Partnership Pilot – San Mateo County, CA Partnership Development and Implementation Why Enter a Partnership? SamTrans has experienced a recent decline in ridership despite a robust local economy, while transit agency operating costs continue to increase. SamTrans is continuing to explore strategies with TNCs to enhance first mile/last mile connections to transit stations as well as potentially replace certain low productivity routes to achieve operating cost savings. Partnership Development SamTrans began its exploration of potential partnerships in early 2017, when it completed a feasibility study that evaluated potential service models for TNC partnerships in hard-to-serve areas of San Mateo County. The transit agency considered several potential service models, and any partnership with TNCs it ultimately implements is likely to focus on enhancing first mile/last mile con- nections to major transit hubs, like Caltrain stations or SFO Airport through geofencing or temporally based TNC trip subsidies. Partnerships to provide service to seniors and paratransit customers are also possible. SamTrans has engaged in discussions with Uber and Lyft but has not established any formal or informal partnership. The TNC pilot project is currently postponed indefi- nitely. There is more transit agency interest in a potential microtransit pilot. Basic Parameters No partnership or partner has been formally identified, and the parameters are not yet specified. However, the transit agency’s feasibility study identified several potential ser- vice zones, including: • Replacing Routes 251 and 256 in Foster City with a subsidized TNC service • Supplementing Routes 251 and 256 in Foster City with a subsidized TNC service • Replacing Route 280 in East Palo Alto with a subsidized TNC service • Supplementing Route 280 in East Palo Alto with a subsidized TNC service • Replacing Route 17 in Half Moon Bay with a “Coastside” subsidized TNC service In each of these service zones except Half Moon Bay, the partnership would be designed to enhance first/last mile connections to Caltrain stations using subsidized TNC service based on geofenced service areas (e.g., Foster City boundaries and Hillsdale Caltrain station complex). In the Half Moon Bay scenario, the partnership would replace low productivity transit routes in the hard-to-serve, lower density, and topographically isolated coastal communities of Half Moon Bay, Miramar, and Montara. SamTrans initially considered replacing low-performing routes with subsidized TNC ser- vices. However, SamTrans received negative feedback on this strategy from its Operations division. The transit agency has since changed its approach to focus on supplementing, rather than replacing, low-performing routes. Partnership Details Data Sharing • In informal discussions, TNCs agreed to provide SamTrans with total ridership and cost- per-rider data. Other high-level performance metrics have not yet been finalized, but SamTrans is likely to request data to support reporting to the NTD. Case Study Highlights First Mile/Last Mile Subsidized Target Market Transit riders, commuters Service Area Context Suburban Vendor/Partner(s) TBD Type of Arrangement Formal Key Themes First/last mile connections to transit, replacing or supplementing low productivity fixed-route transit Partnership Scope at- a-Glance SamTrans is in preliminary discussions with TNCs and has completed a feasibility study, but no partnership has been established. Duration TBD Indicators of Success Specific metrics would need to be developed once the partnership is finalized. They may include ridership shifts from paratransit and improving first/last mile connectivity to transit stations. Funding Source(s) SamTrans operating budget Budget As of September 2018 Allocated: None to date. Expended: N/A

Case Studies 59 Regulatory Considerations • Any replacement of existing fixed-route transit service would require a Title VI analysis. SamTrans has not yet established forms of access and/or payment in a potential partnership for unbanked customers or those without smartphones. • Additionally, hours of service, background checks and drug testing of driver-partners, and ADA compliance issues with a potential TNC partnership have not yet been resolved. Marketing/Outreach Methods • No marketing has taken place yet because no partnership has been established. Only informal outreach with Uber and Lyft is underway. Project Wins and Areas of Further Development Project Wins • None specified. Areas for Further Development • SamTrans will need to develop and refine the following components if the TNC pilot program proceeds: – Procurement and compliance with the ADA and Title VI – Data collection and reporting – Driver safety checks/compliance – Performance metrics – Goal-setting – Revenue and cost projections – Agreements with organized labor Lessons Learned • Developing internal consensus around a strong vision is critical. One thing that SamTrans lacked as an organization was a project champion who truly believed in the concept to “carry the torch” and win people over to their way of thinking. There was skepticism within the transit agency about the value of the pilot, and that led to its postponement. • Adopt a strategic approach to pilot programs. Be clear about the problem the pilot will solve and what the impacts will be. • Work with operations and organized labor well in advance of any pilot implementation, and ensure they have buy-in on the pilot program. SamTrans ended up compromising signifi- cantly on certain key aspects of the program, such as the question of whether to replace or supplement low-performing routes with subsidized TNC services. • A key obstacle the transit agency encountered was challenging (though informal) negotiations with the TNCs on key operations matters such as vehicle fleets and data-sharing. Microtransit vendors have been much easier to work with, as they have stronger motivations to create part- nerships with public transit agencies. Negotiating data-sharing with them has been a much smoother process compared with the TNCs. • Unreliable access to WAVs will continue to be a challenge with the TNCs.

60 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Southeastern Pennsylvania Transportation Authority (SEPTA) Philadelphia, PA Partnership Development and Implementation Why Enter a Partnership? SEPTA has a large and robust transit network that pre-dated many of its regional communities. As such, the transit agency views TNCs as a potential complement to its system. SEPTA had multiple motivations for initiating this partnership, but primary motivations were to expand capacity and access to stations with limited parking and serve sub-markets for which first mile/last mile connections are par- ticularly important (e.g., Airport Line service). Within the 11 pilot stations, SEPTA was interested in testing several different use cases, including airport access; as mitigation for construction occupying parking spaces at rail stations; and filling a gap where stations have limited connecting bus service. The 11 stations were chosen to enable testing of these use cases as well as to achieve geographic equity of the benefit. Partnership Development Prior to initiating a partnership, SEPTA worked to improve first mile/last mile access through other means, such as installing sheltered bike racks throughout the system. The partnership initiation was motivated by first mile/last mile access in suburban communities, and was informal—there was no RFP or unsolicited proposals process, but the TNC’s non-disclosure agreement template was used. In the negotiations, SEPTA decided that it did not believe that Uber could meet federal require- ments for accessibility ADA and equity (Civil Rights Act); therefore, no public money could be used to fund the program, which would be treated as a customer amenity as opposed to a core service. From the outset, it was decided that the promotion would be entirely funded by Uber and be active for three months. Basic Parameters The partnership between SEPTA and Uber involved no exchange of funds between the transit agency and the TNC partner. SEPTA purchased ad space on its own system to market the promotion; Uber funded a 40% discount for rides to or from the 11 stations, up to $10 per trip. Rides did not have to be shared. The 11 eligible stations were Glenside, Jenkintown-Wyncote, Elkins Park, Melrose Park, Warminster, Doylestown, Lansdale, Woodbourne, Swarthmore, Wayne, and Exton. The discount was available to anyone traveling to or from these stations. Partnership Details Data Sharing • No formal data-sharing agreement was set. • SEPTA did not develop an evaluation framework for the partnership, but ultimately is inter- ested in understanding whether TNCs can be used as a first mile/last mile solution that meets the needs of SEPTA, its customers, and TNCs. They want to test some of the early evidence of TNC’s complementary nature with transit in suburban areas. • Uber provided a presentation at the conclusion of the pilot; otherwise, no specific data was received. SEPTA asked for, but did not receive, trip-level data on time of day and day of week, Case Study Highlights Suburban Mobility First Mile/Last Mile Marketing Target Market People driving and parking at SEPTA rail stations Service Area Context Suburban Vendor/Partner(s) UberX Type of Arrangement Informal Key Themes Marketing partnership, limited data sharing, no exchange of funds Partnership Scope at-a-Glance A short pilot to address first mile/last mile connectivity to suburban rail stations through a marketing partnership (no exchange of funds) Duration May 27, 2016 – Sept 5, 2016 Indicators of Success Net new SEPTA ridership Positive publicity Funding Source(s) N/A; no exchange of funds Budget Allocated: N/A Expended: N/A

Case Studies 61 as well as aggregated data on top destinations, unique customers versus total trips, and new customers going to or from the SEPTA station for the first time. • The partnership was not viewed as a core public transit service, nor did SEPTA use federal funding; as such, it did not seek data for NTD reporting purposes. Regulatory Considerations • SEPTA decided to target suburban communities because, at the time, the City of Philadelphia was in the process of negotiating TNC regulations (differently from the other counties in SEPTA’s service area). • SEPTA took the approach of co-promotion, rather than ride subsidies, to reduce the complex- ity of the program and avoid scrutiny by regulators. Marketing/Outreach Methods • SEPTA set a marketing agreement with Intersection, its marketing partner. Uber was not included in the agreement, but did review and approve creative materials used to promote the program on SEPTA property. • Marketing involved presence on the transit agency’s website, social media, local news media, advertisements in stations, and promotions through Uber’s channels. • There was no initial outreach conducted. Project Wins and Areas of Further Development Project Wins • The partnership was received positively by the general public. • The partnership helped SEPTA customers experience an alternative access option that it hopes will provide a growth opportunity for ridership. • Even at stations that were not selected for the partnership, an increased awareness of this new first mile/last mile option was observed on social and local media. Areas for Further Development • SEPTA is interested in continuing to address first mile/last mile and suburban mobility issues through partnerships, but would like better access to data as a key part of a future pilot. • The project manager would have liked to be able to present the results to the transit agency board, demonstrating transfers to SEPTA service from Uber, and set a goal of increasing total trips by a certain percentage. Lessons Learned • Due to the limited data shared, SEPTA was not able to answer the questions it was interested in testing through a partnership with a TNC.

62 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Southwest Ohio Regional Transit Authority (SORTA/Metro) Cincinnati, Ohio Partnership Development and Implementation Why Enter a Partnership? SORTA was aware that many peer transit agencies were preparing to enter into partnerships with TNCs as a means to improve the customer experience and increase ridership, and the transit agency was eager to be innovative. In a region experiencing declining transit ridership, SORTA had expected to leverage discounted TNC rides as a means of improving ridership. Partnership Development SORTA’s ridership coordinator initiated the partnership with Uber on an ad-hoc basis. There was no formal RFP process, and no other TNCs were con- tacted. Uber submitted a contract to SORTA, and SORTA signed the contract without adding any customized parameters. No specific partnership goals were embedded into the signed agreement. This was an entirely marketing-oriented partnership, though there had been a larger vision. SORTA staff had hoped the partnership would expand participation in the region’s Guaranteed Ride Home program, increase transit ridership, and supplement transit during late night periods, though no parameters were set to achieve these objectives. Basic Parameters Uber agreed to offer 20% discount codes on rides for first-time customers— a promotion that is widely available to new customers in many cities in which it operates—in return for SORTA placing Uber’s advertising on tran- sit vehicles and at bus stops. No discounts were available for SORTA custom- ers who already had Uber accounts. The ride discount codes could be used for any Uber ride, regardless of origin, destination, or time of day. For Guaranteed Ride Home-related rides, Uber covered 20% of the fares and SORTA leveraged CMAQ funds to cover the remaining 80%. SORTA did not budget any transit agency funding from this project. Partnership Details Data Sharing The partnership contract did not specify any data-sharing agreement. SORTA requested trip origin and destination and time of day information to verify proper use of GRH funding. The number of new accounts activated using the discount code also was requested, but SORTA did not receive any of these data. Regulatory Considerations • No special considerations were given to compliance with Title VI, ADA, or labor union con- cerns. Uber driver-partners only had to follow rules established by Uber. The transit agency did not receive any complaints related to these regulations, although some customers did complain they could not receive the Uber discounts because they were already Uber account holders. • CMAQ funds were used to reimburse 80% of the trip—after the initial 20% Uber discount was applied—for trips that qualified for the region’s Guaranteed Ride Home program. Customers covered the remaining 80% of the fare for non-Guaranteed Ride Home trips. Case Study Highlights Guaranteed Ride Home Marketing Suburban Mobility Target Market Transit customers Service Area Context Urban Vendor/Partner(s) Uber Type of Arrangement Formal Key Themes Marketing/outreach, transit advertising Partnership Scope at-a-Glance SORTA (Metro) is the public transit agency for the Cincinnati metropolitan area, operating mostly in Hamilton County, Ohio, and Kenton County, Kentucky. In 2016, Metro partnered with Uber to provide free advertising on transit vehicles and the agency’s website in exchange for offering discounted rides to first-time users. Duration March 2016 – March 2017 Indicators of Success Rides taken in CMAQ’s Guaranteed Ride Home program, trip origins and destinations by time of day Funding Source(s) CMAQ (for Guaranteed Ride Home only) Budget N/A, no exchange of funds outside of GRH

Case Studies 63 Marketing/Outreach Methods • Contractually, Uber was allowed to review all press releases that were issued by SORTA related to the partnership. • SORTA engaged with local press, social media, and vehicle and station/stop advertising to publicize the partnership. Project Wins and Areas of Further Development Project Wins • SORTA viewed this as a valuable learning experience. Areas for Further Development • SORTA staff recognize the importance of establishing clear objectives before entering into future partnerships. These objectives are likely to include increasing transit ridership, supple- menting late night service gaps, increasing GRH participation, and commute trip reduction from major employers, among others. Lessons Learned • SORTA’s assets and resources are valuable, and as a result the transit agency should not limit itself to a single TNC through an ad-hoc partnership. More formal, expansive procurement processes may be helpful in selecting future partners. • Do not enter into partnerships before establishing clear goals and objectives and an evalua- tion plan. • Establish key deliverables (e.g., data sharing), milestones, and performance metrics (e.g., reg- ulatory compliance) that potential partners must achieve.

64 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Solano Transportation Authority (STA) First Mile/Last Mile Pilot – Solano County, CA Partnership Development and Implementation Why Enter a Partnership? STA offers several programs that encourage Solano County employees to take transit, vanpool, carpool, bike, or walk to and from work. To encourage transit use for employees working at employment sites just a short distance from the Amtrak rail station with regional rail service, the transit agency looked at options to serve the last, short leg between the rail station and the employment sites. Partnership Development STA investigated contracting for peak hour shuttle service with a private provider to provide the first mile/last mile service, but found that option to be high cost and with only limited hours. Instead, STA reached out to see if a TNC might be interested and came to an agreement with Lyft. The TNC option was more cost effective and also more flexible—it could be used at any time by participating employees, not just the peak hour service that the shuttle option would offer. Basic Parameters STA provides a special code to participating employees at the selected employers. This code allows the employee to pay $2 for each trip to and from the rail station and the employee’s worksite. Geofencing technology ensures that only trips between the rail station and the employment sites are subsidized. If an employee uses Lyft for a trip from the rail station to the local shopping mall, for example, the full, unsubsidized fare applies. Participating employees use their smartphones for booking Lyft trips, with service provided in sedans—the normal TNC model. Lyft invoices STA monthly for the trips provided. Partnership Details Data Sharing • STA receives summary data from Lyft on the pilot users: the number of subsidized trips pro- vided, the number of unique individuals participating, and trip length, which is provided in ranges, for example, x trips from 0–2 miles, y trips from 2 to 4 miles, etc. • Monthly ridership is estimated to be 800 one-way trips based on 20 employees participating as of March 2018. Regulatory Considerations • The pilot is targeted to employees at selected employers—not the general public. Marketing/Outreach Methods • Considerable effort was made to promote the pilot to employers in the business park area. • Marketing efforts purposefully started by targeting only about half of the 70 businesses in the business park area. STA had no data with which to estimate demand for the pilot service and did not want to oversell the pilot. Case Study Highlights First Mile/Last Mile Subsidized Target Market Employees at selected worksites within 2 to 5 miles of a rail station Service Area Context Suburban Vendor/Partner(s) Lyft Type of Arrangement Formal Key Themes Transit use for commuting, first mile/last mile service for targeted employers Partnership Scope at-a-Glance STA subsidizes Lyft trips for employees working in the Solano Business Park area for the short trips to and from the Suisun City Amtrak station to encourage commuter use of the regional rail service. Duration May 2017–Present Indicators of Success Monthly ridership Funding Source(s) California’s Transportation Fund for Clean Air (TFCA) Budget As of Aug. 2018 Allocated: $100,000 Expended: Approx. $7,000

Case Studies 65 • STA’s employer outreach coordinator met with the selected employers and, in some cases, their employees to promote the pilot, finding that the larger employers expressed more interest than the smaller ones. • With hindsight and data showing fewer employees participating than the pilot could handle, STA realizes that marketing efforts could have initially promoted the pilot to all the businesses in the park area. Project Wins and Areas of Further Development Project Wins • Informal feedback from participating employees: They “love it.” • Some employees kept a “station car” parked at the rail station—an older extra car used by the employee for the sole purpose of driving back and forth between the rail station and the employee’s work site. Particularly for those employees, the first mile/last mile pilot with the TNC has been very welcomed. Area of Further Development • STA reported 20 employees were participating in the pilot as of March 2018; goal is 50 par- ticipating employees. • STA has found it is harder than anticipated to get employees to use transit. The enticement of the TNC pilot was useful for fewer employees than STA had hoped. “People are used to driving; trying to change [their mode] is hard.” • STA is continuing the pilot for another year, with plans to add additional employment loca- tions for the first mile/last mile service and incorporate transit centers in addition to the current rail station. Lessons Learned • Only one TNC followed through with genuine interest in a partnership with the transit agency. It is not clear if wider interest would have made a difference in design or imple- mentation of the pilot. • It is challenging to persuade commuters who have relied on their private vehicles for the commute trip to switch to public transit. This may be particularly so in a suburban area that lacks traffic congestion and high parking fees. Single-occupant vehicle commuters may need more than a $2 TNC trip subsidy for the first mile/last mile to their work site to switch modes.

66 Partnerships Between Transit Agencies and Transportation Network Companies (TNCs) Washington Metropolitan Area Transit Authority (WMATA) Abilities-Ride – An Alternative to MetroAccess – Washington, D.C. Region Partnership Development and Implementation Why Enter a Partnership? WMATA has planned and implemented a number of different pilots and programs to provide its ADA paratransit customers with more flexible and convenient options and thereby have the ADA paratransit service function as a safety net and not a primary option for these customers. The transit agency has concerns about the rising demand and cost for ADA paratransit. Partnership Development WMATA had experience with a same-day service for ADA cus- tomers in Washington, D.C., that shifted trips from ADA para transit to more cost-effective taxi service. Building on this experience, the transit agency planned a same-day program for the two Maryland counties in its service area. The transit agency conducted market research, reached out to other cities that had implemented a TNC-type service for ADA riders, wrote a business plan, consulted with the local disability community, issued an RFI and then an RFP: Abilities-Ride Subsidization of Transportation Services as an Alternative to MetroAccess. Through the planning and procurement process, the transit agency was very attentive to issues of vehicle accessibility and response time for trips for customers with mobility devices. The RFP noted that providers could charge WMATA an additional $10 for a trip with an accessible vehicle. The procurement process—using an evaluation committee—ulti- mately selected two taxi companies as the pilot’s providers. The pilot does not include any TNCs. One of these taxi companies had prior experience as a subcontractor for WMATA’s ADA service. Basic Parameters ADA eligible customers traveling in the two Maryland counties can book a trip with one of the two taxi companies, depending on the county in which they wish to travel. Customers can book a trip by telephone, online, or with the taxi company’s smartphone booking app. Customers can take up to four one-way trips each day, paying the first $5 per trip with WMATA paying the next $15. If the trip cost exceeds $20, the customer pays the excess. Trips must begin and end in the Maryland counties. Partnership Details Data Sharing • WMATA was very specific in its RFP for the pilot regarding the data to be submitted, stat- ing that the “Offeror shall confirm its capacity to submit to WMATA a monthly listing of all Case Study Highlights ADA Paratransit Subsidized Target Market ADA paratransit customers Service Area Context Suburban Vendor/Partner(s) Two local taxi companies Type of Arrangement Formal Key Themes Provide a same-day trip option for ADA eligible customers Though TNCs were encouraged to participate in the solicitation, none were selected as providers in the transit agency’s same-day service pilot for its ADA paratransit riders Partnership Scope at-a-Glance WMATA, the transit agency serving the greater Washington, D.C. region, has pursued several alternatives to its ADA paratransit service due to increasing demand and cost. The transit agency issued a comprehensive RFP for providers of same-day service in September 2016, and encouraged TNC participation. The procurement process ultimately chose two taxi companies as the providers, but not a TNC. The program subsidizes up to four one-way trips per day for ADA eligible customers traveling in two Maryland counties. The customer pays the first $5, WMATA subsidizes the next $15, and the customer pays any amount over $20. Duration September 2017–Present Indicators of Success Average Monthly Ridership. Approx. 4,100; (6% require WAVs); data as of Jan. 2018. Number of customers participating. Impact on-demand for ADA paratransit. Funding Source(s) WMATA’s ADA paratransit budget. Budget Allocated: No funds specifically allocated for As of Aug. 2018 pilot. Expended: Not shared publicly at present time.

Case Studies 67 trips provided the previous month in a sortable and formula enabled spreadsheet, preferably Microsoft Excel” that is to include, among other data elements, the following: – Customer’s MetroAccess ID number and first and last name – WAV requested trip – Pick-up time and pick-up location – Drop-off location and drop-off time – Trip miles – No show/late cancellation trip – Base fare – Additional trip fees – Total trip cost – WAV fare surcharge • WMATA reported that the two taxi companies are submitting all the required data. Regulation Considerations • WMATA purposefully included specifications in its procurement document to ensure the program would meet all ADA and Title VI requirements. The transit agency even paused its solicitation process when the FTA issued an official statement providing more specific guid- ance in December 2016. The statement was subsequently posted on the WMATA website as a type of RFP amendment. Marketing/Outreach Methods • Little marketing was done for the soft launch, but since then, marketing is more expansive, including direct contact to MetroAccess customers, brochures for new customers, and provi- sion of refrigerator magnets with key information on the program. • WMATA has provided a fare discount promotion for a limited time, which offered to pay the first $20 of the fare for specific trips that a customer regularly takes on MetroAccess. Project Wins and Areas of Further Development Project Wins • Customer feedback on the pilot has been mostly positive. There were a few issues at the start, as the service was a new offering for the taxi companies, and adjustments were made to address those. • A trip on the pilot provides cost savings to WMATA compared to ADA paratransit: $15 vs. $50. Areas of Further Development • WMATA is increasing customer outreach and marketing with ADA eligible customers to build ridership on the program and generate greater cost savings. Lessons Learned • While WMATA designed its plan and RFP for same-day service for ADA customers to allow for participation of TNCs as possible providers, TNCs are currently not providers in the pilot. Instead, two taxi companies are the providers. According to a news report, a TNC bid was not accepted because the company’s business model did not conform to ADA requirements. • The two taxi companies are meeting all WMATA requirements for the service including the detailed data requirements. • A same-day service for ADA eligible customers that includes the enhancements typically avail- able with TNCs but that also meets ADA and Title VI requirements can be successfully deployed with technology-sophisticated taxi companies that have an appropriate supply of WAVs.