New York City Mayor Bill de Blasio took aim at Uber this summer, trying (and failing) to set a cap on the number of its for-hire cars operating in the city. The ride-share service has drawn criticism on a number of issues — including its labor practices — but the mayor said his main concern was traffic congestion. The number of yellow and green taxis on New York’s streets is carefully controlled by the city. Would additional pickups from an uncapped Uber fleet lead to urban gridlock?
Based on year-over-year counts, these fears appear to be unfounded. Uber has not caused a net increase in pickups, at least not in Manhattan, according to a FiveThirtyEight analysis of pickup data. Instead, the ride-share service is replacing cabs in the center of the city and supplementing them in the outer boroughs.
In response to a Freedom of Information Law request we made in July, the New York City Taxi and Limousine Commission last month sent us Uber pickup data from January to June of this year. The TLC had previously sent us Uber pickup data for April 2014 to September 2014 in response to the same request. With the new data, we can now compare the April-to-June period in each year. Uber pickups in New York City rose sharply over this time span — no surprise, as the company was aggressively entering the market. Green cabs (which pick up passengers in the outer boroughs and in northern Manhattan) saw an uptick, while yellow cab pickups declined, according to ride data the TLC publishes online. All in all, total pickups for these three services (we don’t have data on location of pickups for other black-car companies for the 2015 period) increased from 48 million to 51 million.five Uber bases in the city: Unter, Hinter, Weiter, Schmecken and Danach. Cars from those bases handled virtually all Uber rides in the city at that time, according to a TLC spokesman. This year’s data also includes trips from newer Uber bases Grun, Dreist and Drinnen. There is a lag between when a base starts operating and when Uber is required to report its data, so our data set may be missing some Uber rides. In practice, new bases handle very few rides before they are required to report data, an Uber spokesman said, so any gaps are minor and don’t affect our overall findings.">1
Three million additional pickups sounds like a substantial increase, but the big question — as The Economist pointed out in August — is where those trips are coming from. It would be a problem if those new trips were originating in central and lower Manhattan, the region’s congested core. That’s where de Blasio focused his concerns about new traffic, and it’s the area that Uber was most aggressively targeting with driver incentives. On the other hand, new pickups would be less of a problem in the other boroughs and in northern Manhattan, where the city has tried to increase cab availability through the green Boro taxi program. It would be harder for de Blasio to push back against Uber if it were primarily adding pickups in these areas.
The maps below show the net change in total, taxi, and Uber pickups from April-June 2014 to April-June 2015. Zones marked in blue experienced a net decrease in pickups, while red zones saw a net increase.
Both taxis and Ubers increased their pickups in Brooklyn and Queens, particularly in gentrifying areas; the services are supplementing each other in these neighborhoods. Uber also gained in Staten Island and the Bronx, while taxi pickups were flat in those boroughs. In Manhattan’s core — the area outlined on the maps above, where green cabs aren’t allowed to make pickups2 — Uber added 3.82 million trips in April through June of this year, compared with the same period a year earlier. Taxis, in the same area, lost 3.83 million pickups. The total number of pickups was virtually unchanged: 39.37 million in 2014 versus 39.36 million in 2015.
|CHANGE IN APRIL-JUNE PICKUPS, 2014 TO 2015|
This shift was consistent throughout the core. There are 51 taxi zonespublished online with latitude/longitude coordinates.">3 in this part of the city, and Uber added pickups in every one — as we’d expect given its aggressive rollout. Taxis lost pickups in every zone, and these gains and losses were almost perfectly one-to-one:
One notable exception was the West Chelsea/Hudson Yards zone, home to the country’s largest private real estate development. Uber added 112,000 pickups in this zone, while taxi cabs lost “only” 63,000. In Tribeca, too, Uber pickups rose by far more than taxi rides fell, resulting in an additional 51,000 total pickups. Every other part of the core saw a decrease or a much smaller increase in net pickups — including the neighborhoods that collectively make up the Manhattan central business district, which is all of the island south of 60th Street. There, Uber gained 3.10 million pickups, while taxis lost 3.09 million.4
Throughout Manhattan, riders have shifted from taxis to Ubers millions of times, perhaps attracted to features Uber promotes as advantages: newer cars, no need to hail, driver ratings and no tipping. These features, though, don’t appear to have pulled Manhattanites from their private cars, subways, buses or bikes: The total number of taxi and Uber rides didn’t grow.
It’s worth noting that net pickups and “congestion” don’t precisely correspond to each other, as we don’t have data on dropoffs, on ride length, or on location of pickups by other services like Lyft, an Uber competitor, and established black-car companies. It’s also possible that many more empty cabs are cruising Manhattan streets than last year, although it would not be financially viable for drivers — who primarily rent their cabs — to keep this up for very long.the end of 2014.">5
“Pickups aren’t the whole picture, and they don’t shed light on most of the actual contributors to congestion,” Wiley Norvell, a de Blasio spokesman, said in an emailed statement. Those contributors, he said, include how many for-hire cars are on the road, how long their trips are, and how much they cruise between fares. He added that the city is conducting a “thorough analysis” of which factors are affecting congestion.
Josh Mohrer, Uber’s general manager for New York City, dismissed concerns that Uber is adding to congestion in the busiest parts of Manhattan. The average number of Uber cars in Manhattan’s central business district between 7 a.m. and 7 p.m. on weekdays is 2,000, Mohrer said in a telephone interview. “The congestion piece, I write off entirely,” he said.
Manhattan sees an enormous volume of pickups, so Uber’s 4 million additional trips only increased its share of rides in the borough from 4 percent to 13 percent. Uber’s market share, which last year was bigger outside Manhattan than in it, has continued to grow more in outer boroughs, as Uber has been happy to point out:
|UBER’S SHARE OF PICKUPS (APRIL-JUNE)|
|BOROUGH||DISTRIBUTION OF UBER + TAXI PICKUPS (2015)||2014||2015||CHANGE|
|New York City||–||4||15||+12|
These changes are dramatic — Brooklyn now has one Uber pickup for every two taxi pickups — but they obscure where the real battle is taking place. Over 70 percent of Uber pickups and 80 percent of taxi pickups still originate in Manhattan, the city’s busiest, wealthiest borough.Capital New York pointed out that most Uber rides continued to originate in Manhattan in its analysis last week, using data it got from TLC via its own FOIL request.">6 It appears that Ubers and taxis can coexist outside of Manhattan, at least for now. Within Manhattan, taxis might be in trouble.
CLARIFICATION (Oct. 13, 7:15 p.m.): A previous version of this article referred imprecisely to the degree of regulation of the Uber fleet in New York City. Although the number of Uber cars is not regulated, Uber’s drivers and cars are licensed and regulated by the city’s Taxi and Limousine Commission.
Dhrumil Mehta and Nate Silver contributed reporting.
The Uber data received through our Freedom of Information Law request is available on Github.