Daniel Vignon ,

PhD

New York City's 2019 ride-hailing surcharge cut overall taxi and ride-share trips by 11 percent in Manhattan but failed to reduce traffic congestion, a key goal of the policy, according to a new NYU Tandon School of Engineering study published in Transportation Research Part A.

“While this surcharge differs from the MTA's proposed congestion pricing plan, the study's findings can contribute to the current discourse,” said Daniel Vignon — assistant professor of Civil and Urban Engineering (CUE) and member of C2SMARTER, a U.S. Department of Transportation Tier 1 University Transportation Center — who led the research with CUE PhD student Yanchao Li. “Indeed, the research reveals how pricing policies can disproportionately affect different communities and emphasizes that accessible transit alternatives play a crucial role in shaping how such policies impact travel behavior.”

Using a Difference-in-Differences framework — a statistical method that compares changes in outcomes between locations subject to a policy and those that are not — researchers isolated the impact of the $2.50 to $2.75 fee imposed below 96th Street by analyzing patterns both inside and outside the surcharge zone, while also comparing the same areas before and after the policy took effect.

"We were not necessarily surprised by the findings," explained Vignon. "The City claims that Uber, Lyft and taxis increase congestion, but we would say that they are not the major contributors," noting that research from other cities has also found ride-hailing services don't significantly contribute to traffic congestion. “In general, most cities experience a reduction in travel speed between 2% to 8% following the entry of Uber/Lyft.”

While traffic speeds remained virtually unchanged after the surcharge, Lyft experienced a 17 percent decrease in trips, and Uber and yellow cabs saw drops of 9 percent and 8 percent respectively, the research showed.

The policy's impact varied based on available transportation alternatives. Areas without subway or Citi Bike access saw only a 1.6 percent reduction in rides, while neighborhoods with both options experienced a 7.4 percent decrease. Areas with Citi Bike alone showed a 6.8 percent reduction.

The study revealed a complex relationship between income and transit access. Higher-income neighborhoods, despite typically having better transit options, showed minimal reduction in ride-hailing use. In contrast, lower-income areas saw sharp declines even though they often had fewer transit alternatives.

"When policymakers plan for any type of congestion pricing, it's critical they account for the alternative transportation options available at a granular level. A policy that works well in one neighborhood may impose a very high cost in areas where people live with far fewer resources and choices," said Vignon, noting that the street-hailing industry saw an 8 percent decrease in revenue after implementation. “It seems that this policy resulted in a net welfare loss for the city, at least in the shorter term, when considering all factors, such as abandoned rides and the decrease in driver revenues. In the longer term, to determine whether the policy is a net positive, one would have to account for how the collected fees are spent.”

This study is part of Vignon’s body of work examining how regulatory policies affect transportation systems. His research interests span ride-hailing regulations, autonomous vehicles, and infrastructure investment, analyzing how agencies can improve system performance while considering that users and transportation service providers will adapt their behavior based on policy changes.

The research also contributes to the portfolio of C2SMARTER, a consortium of seven universities led by NYU Tandon that is pursuing an ambitious research, education, training, and technology transfer program agenda to address the U.S. DOT priority area of Congestion Reduction. It received its most recent Tier 1 UTC designation in early 2023, providing it $15 million in funding for five years and extending its first such designation that came in 2016. 

In this study, Vignon and Li analyzed over 300,000 ride-hailing records from NYC's Taxi and Limousine Commission, along with nearly 1 million traffic speed measurements from Uber Movement, incorporating data from Citi Bike, subway locations, household income statistics, and weather patterns.

Yanchao Li, Daniel Vignon, Do ride-hailing congestion fees in NYC work?, Transportation Research Part A: Policy and Practice,
Volume 190, 2024, 104274,
ISSN 0965-8564 https://doi.org/10.1016/j.tra.2024.104274.