Aarian Marshall covers autonomous vehicles, transportation policy, and urban planning for WIRED.In many ways, the project feels like the future, an innovative embrace of Silicon Valley. The county nixed two poorly performing bus routes and used that money to give riders $3 breaks, and later $5 breaks, on Uber or taxi rides to specified public transit stops. By swiping to a special screen inside the app, Uber riders can travel to within 800 feet of 24 eligible bus stops and then transfer to a bus for free.“In our county, we really are trying to change the culture so that more people are using public transit rather than thinking about putting their car on the road,” says Janet Long, a Pinellas County commissioner and chair of the transit authority board. She and her fellow commissioners voted in May to extend the program, called Direct Connect, through 2021, for up to $300,000 a year.Pinellas County, which includes cities like St. Petersburg and Clearwater, “was a surprising first” for this kind of experiment, says Sharon Feigon, executive director of the nonprofit organization Shared Use Mobility Center. “It’s a smaller agency, a conservative community in a lot of ways, and they were taking all these risks.”Since 2016, the idea has been widely copied, with varying success. A program in exurban Toronto that went all in on subsidized Ubers in place of fixed-route buses was a victim of its own popularity, and was forced to raise prices and limit individuals’ use to hold down costs. Kansas City, Missouri, ended its app-based shared-van service after netting just four rides a day. Centennial, Colorado, nixed a partnership with Lyft after spending more money and serving fewer riders than its traditional call-a-ride service. Boston, meanwhile, just re-upped a three-year-old partnership with Uber and Lyft to supplement its call-a-ride service for residents with disabilities, despite some complaints from wheelchair-using riders that the program had left them stranded.In a new report, the Shared Use Mobility Center gives the Pinellas County program mixed reviews. (The center, which receives funding from Uber, worked with Uber and the transit authority on the report.) On one hand, public officials have worked diligently to improve the service, changing where it operates and boosting marketing when ridership foundered. They also showed creativity in partnering with private companies and adapting the strict guidelines and paperwork that typically come with public funding. (Transit authority lawyers decided, for example, to defer to Uber’s drug- and alcohol-testing process even though it did not meet federal standards, reasoning it provided taxis as an alternative. They also had to ensure that people without smartphones could access the service.) The transit authority “was willing to take risks and keep iterating. I think that was really commendable,” Feigon says.But viewed through a different lens, the Florida program raises questions about whether this kind of “first-mile, last-mile” collaboration between the public and private sectors is worth the time and money, and whether the tech can attract more people onto struggling transit services.
For one, not that many people are riding. In its first six months, Direct Connect supported fewer than two trips per day. After two years of experimentation, reworking, and national publicity, the numbers did climb, but only to 30 trips per day. By contrast, each stop on the two canceled bus routes served fewer than three riders a day. Average daily ridership for all of the transit authority’s routes is 32,419.Daniel Reck, a transportation doctoral student at the Swiss university ETH Zurich, is fascinated by Pinellas’ ride-hail collaboration. But when he got to Florida for a short research trip, he was puzzled by the program’s low ridership numbers. Reck’s research, which has not yet been peer-reviewed, suggests that Direct Connect is hampered by something that transportation folk call a “transfer penalty”—a measure of how many extra minutes commuters are willing to spend traveling to avoid a transfer. What’s it worth to avoid transferring to a bus that runs infrequently and often behind schedule?In Pinellas, Reck and a colleague found that the Direct Connect program saved riders about 15.7 minutes of travel time. But he questions whether the savings makes it worth taking two vehicles—the Uber and the bus—instead of just one. He says the transfer penalty is “a conceptual barrier” to using ride-hail services for the first and last mile. This has implications beyond ride-hail experiments, he points out. Many agencies are pondering how autonomous taxis or shuttles might supplement their networks, helping to move people from their homes to bus and train lines. But what if riders won’t get onboard?Uber, which has established at least 20 public transit projects like Direct Connect around the globe, acknowledges that it can be difficult to convince riders to transfer. “There’s always a transfer penalty, for any public transit service,” says Chris Pangilinan, who heads public transit partnerships for the company. But he says Uber sees a strong relationship between Uber rides and transit stops. When London launched a limited nighttime Tube service in 2016, for example, the company says the number of Uber journeys beginning within 200 meters of a Tube station during the nighttime service rose 22 percent.Here’s the especially tricky thing: Pinellas can’t track how many people are using its program to transfer to public transit, nor how many people have been converted to public transit users. The transit authority initially sought specific pick-up and drop-off data from Uber and the taxi companies, so it could understand how its new service was being used. Uber refused. “While we would have loved more data at the beginning of the pilot, we think it was important to get started with the partnership,” says Whitney Fox, a spokesperson for the transit authority.Uber later provided the transit authority with quarterly rankings of the transit stops most frequently used for the program. But Uber data is still marked as a trade secret, and public record requests must be run through Uber’s legal team. The taxi company that the agency works with, by contrast, provides information on pick-up and drop-off times and on rider identities and fares.There was also a data blooper. Uber says it accidentally allowed riders to request subsidized rides from places outside the agreed-upon service area for much of 2018. (Reck, the researcher, first spotted the discrepancy earlier this year.) The ride-hail company did not charge the transit authority for these extra rides, but they inflated the ridership numbers for a time.The mess-up points to another important question: Whether these kinds of projects can scale. As in the Toronto Uber experiment, public officials realized that “success”—that is, lots of people using Uber and taxis to get to transit—would have cost the agency more money than it was spending on its low-performing bus routes. In other words: The more people who use the Direct Connect, the less sustainable it is for the transit authority. That’s unfortunate for a program that seeks to convince more people to ride public transit.
And Pinellas County residents don’t want to spend more money on transit. The region is the nation’s 18th largest, but it ranks 180th in transit operating spending per capita. In 2014, voters defeated a referendum to redesign the county’s transit service, one of the worst in the US. That public tightfistedness has forced Pinellas to get creative—for better or worse.
- Inside Backpage.com’s vicious battle with the Feds
- Spiff up your real-world skills with old-timey YouTube
- Small sounds, big money: the commercialization of ASMR
- How to stop robocalls—or at least slow them down
- Why electric buses haven't taken over the world—yet
- 📱 Torn between the latest phones? Never fear—check out our iPhone buying guide and favorite Android phones
- 📩 Hungry for even more deep dives on your next favorite topic? Sign up for the Backchannel newsletter