Laura Bliss is a staff writer at CityLab, covering transportation and technology. She also authors MapLab, a biweekly newsletter about maps (subscribe here). Her work has appeared in the New York Times, The Atlantic, Los Angeles magazine, and beyond.
Ride-hailing services drive down bus and rail ridership in urban markets, a new University of Kentucky paper claims.
When Uber and Lyft arrived on the urban scene a decade ago, they claimed to pair well with buses and trains. By shuttling riders to train stations, guaranteeing late-night returns, and plugging in as paratransit, on-demand transportation could encourage travelers to abandon private cars and use public modes, they said. A lot of people believed them.
But now, transit systems are in crisis. Some 31 of 35 major metropolitan areas in the United States lost passengers in 2017, including the cities with largest ridership bases. There are numerous factors at play, but a small mountain of research singles out the rise of Uber and Lyft. A new paper by University of Kentucky scholars piles on: For every year after ride-hailing companies enter an urban market, rail ridership can be expected to fall by 1.3 percent, and bus ridership by 1.7 percent, it shows.
“We are starting to piece together multiple parts to the story,” said Gregory Erhardt, a University of Kentucky civil engineering professor and the lead author of the study. “For a long time it’s been about ride-hailing complementing transit in different ways. That is true to a degree. But it’s a question of whether it’s happening enough.”
This research picks up where some other recent attempts to uncover the cause of transit declines have left off. Last year, a study from McGill University pointed to service cuts across North American transit agencies as the primary driver for ridership downturns between 2000 and 2015. The methodology was sound, Erhardt believed, but the data it relied on didn’t capture the most recent trends. Similarly, another 2018 paper from UCLA pointed to an uptick in car ownership by lower-income immigrants as the explanation in Los Angeles. But again, as Erhardt points out in the paper, it only looked as far as 2015.
That year was important: It marks the take-off point for ride-hailing’s most dramatic growth and transit’s biggest stumble in many cities. Between 2015 and 2018, Uber and Lyft trips ballooned from 60,000 to nearly 600,000 in New York City, the largest U.S. ride-hailing market and one of the few that requires for-hire transportation services to report trip data. Meanwhile, subway and bus ridership there fell by about 580,000 boardings per day. Such a magnitude of change also indicated to Erdhardt that the effects of ride-hailing might accelerate the more the services grew.
With two graduate students, Erdhardt pulled publicly available ridership and system data for 22 U.S. metropolitan areas from 2002 to April 2018 (the freshest information available) from the National Transit Database. They then conducted a regression analysis to determine which of a list of possible variables could best explain the declines. Alongside the boom in ride-hailing (from the start of market entry), they studied the effects of other likely culprits, such as fare changes, population, employment rates, the share of car-free households, the presence of bikesharing, and gas prices. All of these datapoints came from open, public sources, including the Census Bureau.
Many of these factors had some relationship to ridership trends—some positive, some negative. The introduction of Uber and Lyft fell in the latter category, Erdhardt found. In the 22 markets served by ride-hailing, ride-hailing companies were shown to further drag down boardings as they grew over time.
That effect looks slight from year to year. But added it up, it’s substantial. “After 8 years, [ride-hailing services] would be associated with a 12.7 percent decrease in bus ridership,” the study states.
The working paper was presented at the Transportation Research Board annual meeting in Washington, D.C. last week and is still under review for publication. But it captured a lot of attention at the conference and on Twitter. “In case you are still under the illusion that ride-hailing is somehow beneficial for transit ridership, here’s additional evidence to the contrary,” tweeted Yonah Freemark, a transportation analyst and consultant. Streetsblog reporter Angie Schmitt declared: “Two companies single handedly wiped out transit ridership gains across the U.S. over the past 3+ years. The environmental and social effects of this are just staggering.”
“Single-handedly” may not be quite right, since the paper also finds that cheap gas prices and higher fares have a negative effect on ridership. And this analysis does not include other, more local factors that could be pushing riders off of buses and trains, such as the maintenance and reliability issues plaguing New York City and Washington, D.C. in particular. Nor does it look at safety perceptions among riders, a growing concern on Bay Area transit.
“Perhaps the presence of innovative modes makes it a more compelling or sexier question to ask, but I think there are lots of reasons people aren’t taking these modes as much as before,” said Susan Shaheen, the co-director of the Transportation Sustainability Research Center at UC Berkeley and an expert in shared mobility. “The question of causality is perplexing and hard, and it’s not necessarily one thing.”
Plus, the study is highly aggregated—it lumps together trend-lines of 22 cities with very different shapes and compositions. What’s hurting bus ridership in Houston may be quite different from the factors in Portland, and ride-hailing may not be equally responsible from place to place.
Still, Shaheen and other researchers praised the study’s straightforward and unbiased methodology. It adds good-faith, empirical support to the narrative that ride-hailing and transit “are competing for the same pie in many cities,” said Joseph Schwieterman, the director of the Chaddick Institute for Metropolitan Development at DePaul University.
And the analysis uses publicly available information that any other researcher can use to replicate it. Ride-hailing companies are famously protective of their trip data, which presents a challenge to researchers trying to untangle how new and old modes are interacting—and to reporters trying to decipher science from spin.
“The data is transparent and methods themselves are very transparent,” said Joe Castiglione, the deputy director for technology, data, and analysis at the San Francisco County Transportation Authority, which recently found that Uber and Lyft have significantly worsened congestion in San Francisco. “I think one of the things that is important is that we provide decision-makers with the most empirically based information we can."
Those decision-makers need to pay attention: Cities are facing worsening congestion and CO2 emissions as more private cars fill the roads and vehicle-miles rise. That doesn’t mean Uber and Lyft aren’t fulfilling their promise of “complementing” public transit in certain cases: Riders who do not own vehicles, for example, and who weren’t previously served by taxis have been shown to benefit from these services. But overall, the relationship seems to be more competitive than collaborative. It’s becoming clearer that one side is losing.