The report from the San Francisco County Transportation Authority crunched data from November-December 2010 and the same two months in 2016 to get snapshots of how traffic changed over those six years.
“For sure, congestion has been getting worse,” said Joe Castiglione, Transportation Authority deputy director for technology, data and analysis. “The question we tried to answer is, do (Uber and Lyft) affect congestion in San Francisco and by how much?”
The answer, the report said, is “yes” and “by about 50 percent.”
In 2010 — not long after the Great Recession ended — fewer people were driving, traffic flowed more freely, Uber was new and Lyft didn’t exist. In 2016, after the economy had recovered, traffic was much worse, the city had many more jobs and residents, and both ride-hailing services had become potent forces.
The report attributes about half the increase in congestion over the six years to ride hailing, with the rest due to 110,000 new jobs and 70,000 new residents. Ride-hailing’s impact is due both to the number of cars on the streets and behaviors such as stopping in traffic lanes to pick up and drop off passengers, thus delaying vehicles behind them, the Transportation Authority said.
Both Uber and Lyft contested the report’s conclusions and methodology and said they’d like to work with the city on addressing congestion. Both, for instance, have recently backed the concept of congestion pricing, in which cities impose a surcharge on cars during peak traffic hours. Both also pointed to their carpooling options as reducing traffic.
“While we appreciate efforts to better understand the causes of congestion, this study fails to consider critical factors like the spike in tourism or the growth of freight deliveries, both of which have exploded since the study’s baseline date of 2010,” Uber spokesman Davis White said in a statement. “It also completely overlooks the role new alternative forms of transportation like our Jump bike product can play in reducing the need for personal car travel.”
“Congestion is a complex issue, and Lyft is committed to being a part of the solution,” Lyft spokeswoman Lauren Alexander said in a statement. “Since day one, Lyft has focused on creating more efficient, affordable transportation options that take cars off the road, increase occupancy of cars on the road, and reduce transportation barriers.”
Ride-hail impacts varied by neighborhood and time of day, the report said. Uber and Lyft’s nighttime effect was particularly dramatic, it found, with speeds in the evening — between 6:30 p.m. and 3 a.m. — declining by more than 4 mph — 69 percent of which was due to ride-hailing.
Traffic slowdowns were most pronounced in District Six, which includes South of Market, the Mission, Mission Bay and Treasure Island. The city’s west and south sides, which have much less Uber and Lyft activity and less population growth, also had less congestion growth.
Certain choke points saw disproportionate impact. For instance, delays on Bryant Street between Fifth and Sixth streets (leading to the Bay Bridge on-ramp) doubled in the 2010-16 time frame, with 80 percent of that increase due to ride-hailing, the Transportation Authority said.
Looking just at 2016 congestion, rather than growth since 2010, the report said ride-hailing accounts for a quarter of trip delays citywide — meaning the additional time needed to complete a trip — and 36 percent of delays in the downtown core. A full 5 percent of all vehicle trips in the city in 2016 were ride-hailing cars, it said.
The Transportation Authority report, which Castiglione described as “grounded in data and rigor,” drew its ride-hailing data from a previous Transportation Authority study, in which researchers from Northeastern University used automated software to collect detailed information directly from the companies’ apps about where and when trips occur. That study found that Uber and Lyft drivers rack up more than half a million miles a day on city streets, including both the rides they give, their time en route to pick up passengers, and “deadhead” time when they circle as they await fares.
Overall traffic slowdowns were measured with data from INRIX, which draws it from real-time GPS monitoring sources and highway performance monitoring systems. The University of Kentucky participated in the data crunching.
Both Uber and Lyft keep their data under close wraps. Although they share trip information with their regulator, the California Public Utilities Commission, they’ve persuaded that entity to shield it as trade secrets. San Francisco went to court to compel the companies to share data, and Lyft has done so since February, but it is still kept under lock and key for confidentiality reasons. Uber is fighting the city’s subpoena for data on its rides and drivers.
More data may be forthcoming soon, however. Uber and Lyft last month said they’d provide ride information, including curbside pickup/drop-off counts, to the new SharedStreets initiative, a public-private collaboration that allows for the exchange of data.
In addition, more data may emerge through a future San Francisco tax on ride-hailed trips, which Uber and Lyft have agreed to because it’s preferable to a stiffer tax the city had threatened. The proposed tax is scheduled for the November 2019 ballot and would need a two-thirds margin to pass, although since it faces no organized opposition, experts think it’s likely to succeed. It would take effect in January 2020.
The Transportation Authority report underscores the tax’s importance to help mitigate ride-hailing’s impact, according to District Three Supervisor Aaron Peskin, who chairs the Transportation Authority and negotiated the tax agreement with the companies. The money, which his office estimates at $30 million annually, would be administered by the Transportation Authority for transit uses.
The report used three commonly accepted measures of congestion: how much extra time vehicles needed to complete certain trips, which it found grew by 40,000 hours a day over the six years; how many extra vehicle miles were traveled on streets (630,000); and how much slower cars had to travel on given routes (average speeds fell 3.1 miles per hour).
©2018 the San Francisco Chronicle Distributed by Tribune Content Agency, LLC.
