The figures “are eye-popping,” said Joe Castiglione, deputy director for technology, data and analysis at the San Francisco County Transportation Authority. He co-authored the study with researchers from the University of Kentucky.
It shows that hours of vehicle delays increased by 62% throughout the city from 2010 to 2016, the period when ride-hailing services began proliferating on the streets. Traffic models that exclude Uber and Lyft cars show that hours of delay would have gone up 22% in their absence.
Extrapolating from those numbers, the study’s authors concluded that on-demand ride services — or transportation network companies, as they’re known in academic patois — are clogging roads and siphoning people from mass transit, going against the companies’ stated mission to wean people off of private cars. The authors laid out their findings in the scholarly journal Science Advances, providing fodder for policymakers seeking to regulate these companies.
Among the measures being considered in San Francisco are a proposal to tax Uber’s and Lyft’s net fares, as well as congestion pricing — a road-toll intervention that aims to unclog busy streets.
A similar study that the Transportation Authority published last year looked more broadly at swelling traffic from 2010 to 2016, and found that transportation network companies comprised about half of it, with the other half stemming from job and population growth. Wednesday’s study narrowly measured the correlation between ride-hailing services and increased congestion.
Uber and Lyft contested the data Transportation Authority officials released in October, saying that it didn’t account for the growth in tourism, freight or delivery services that increased with the economic recovery. Both companies support congestion pricing, and both say their on-demand services help bolster mass transit, claims that the researchers dispute.
“While studies disagree on causes for congestion, almost everyone agrees on the solution,” an Uber spokesperson said in a statement Wednesday. “We need tools that help ensure sustainable travel modes like public transportation are prioritized over single occupant vehicles. That’s one reason we believe in comprehensive congestion pricing, which would provide millions to invest in cities’ public transportation systems.”
To Castiglione, though, the report’s findings “are pretty clear.”
“Many factors contribute to congestion — including population growth,” he said. “But the addition of TNCs (such as Uber and Lyft) is greater than all of them.”
He cautioned that the story isn’t quite the same across the city. Although transportation network companies had a deep impact downtown and in North Beach, they barely made a blip in peripheral neighborhoods like the Outer Sunset.
While for-hire vehicles abound in urban areas throughout the globe, they’re especially popular in Uber’s birthplace, next to Silicon Valley. And maybe that’s not a bad thing, said Randy Rentschler, legislative director of the Metropolitan Transportation Commission.
“If Uber and Lyft are creating more traffic, maybe it’s because people want to be in the city now,” Rentschler said. “Maybe it’s a sign of economic vitality. One of the things that the Bay Area has a hard time struggling with is that traffic is not universally bad.”
Yet the problem with transportation network companies isn’t just volume. It’s also the drivers’ behavior, said Gregory Erhardt, an assistant professor of civil engineering at the University of Kentucky and co-author of the study.
“When you look at pickup and drop-off behavior, the drivers stop in turn lanes, travel lanes or bicycle lanes,” Erhardt said. Each time that happens in a major arterial, it blocks the flow of traffic for 140 seconds — more than two minutes of dead time, the researchers found.
Several other features of for-hire cars add to traffic misery in San Francisco. Most Uber and Lyft drivers — some 70% — come in from other cities, including a substantial labor force from as far away as the Central Valley. They spend 20 to 30% of the day trawling for passengers, mostly in downtown areas where public transit options are plentiful.
Nationally, buses and rail systems saw a precipitous decline over the past four years, because they’re competing for the same customers as the transportation network companies, Erhardt said. BART is fighting to keep night and weekend riders who have peeled off to Uber and Lyft, and Muni, while growing, is scrambling to improve service.
There is an optimal way to fit these companies into a complex transportation puzzle, if people use them to travel from a transit hub to a specific Point B that’s not served by mass transit. But a growing body of evidence suggests that’s not what’s happening.
“Between 43 and 61% of TNC trips substitute for transit, walk or bike travel or would not have been made at all, adding traffic to the road that otherwise would not have been there,” the report said.
Erhardt said it may be hard for other researchers to replicate those findings because Uber and Lyft keep such a tight lid on their trip data. Officials at the California Public Utilities Commission — the public agency that regulates transportation companies — are also reluctant to turn over numbers.
When Erhardt approached Uber for records two years ago, the company only offered to provide data on trips from rail stations, which show how Uber supports mass transit.
“That only tells the positive part of the story,” Castiglione said. He and Erhardt ultimately teamed up with computer scientists at Northeastern University to mine the data themselves.
The report came as Uber approaches its initial public offering of shares, scheduled for Friday. In San Francisco, Uber and Lyft drivers blocked off Market Street in protest of what they call unfair working conditions. It’s unclear how that action affected traffic.
©2019 the San Francisco Chronicle Distributed by Tribune Content Agency, LLC.
