Download this Paper Open PDF in Browser

Measuring the Impact of an Unanticipated Suspension of Ride-Sourcing in
Austin, Texas

20 Pages Posted: 31 May 2017  

Robert Cornelius Hampshire

University of Michigan - Transportation Research Institute

Chris Simek

Texas A&M University - Texas Transportation Institute

Tayo Fabusuyi

Carnegie Mellon University - College of Engineering

Xuan Di

Columbia University

Xi Chen

University of Michigan at Dearborn

Date Written: May 31, 2017

Abstract

On May 7, 2016 residents of Austin, TX voted against Proposition 1, which would have allowed transportation networking companies (TNCs) to continue using their own background check systems. The defeat of the proposition prompted Uber and Lyft to suspend services in Austin indefinitely. The suspension provided for a natural experiment to measure the impact of the suspension on travel behavior. In examining the impact, we conducted an online survey that combines stated and revealed preference questions (N=1,840) of former Uber and/or Lyft users in Austin to explore the effect of the suspension on travel behavior.

Regression analyses, modeled to capture both the before and after travel behavioral pattern of the suspension, were used to test our hypothesis of the impact of the service suspension on travel behavior along three dimensions — mode choice, trip frequency, and vehicle ownership. Our analysis finds that 42 percent of respondents who had used Uber or Lyft to make a trip prior to the suspension reported transitioning to another TNC as the means by which similar trips were most often made after the suspension. A near equal proportion (41 percent) reported transitioning to a personal vehicle, while 3 percent transitioned to public transit. The analysis also suggests that, when looking at trips made for the same purpose pre and post suspension, individuals that transitioned from Uber or Lyft to a personal vehicle were more likely (23 percent more likely) to make more trips than individuals transitioning from Uber or Lyft to another TNC. Additionally, approximately 9 percent reported purchasing an additional vehicle in response to the service suspension. The vehicle acquisition trend was driven primarily by respondents who were inconvenienced by the service suspension — the odds of acquiring a car for an inconvenienced respondent was more than five times that of an individual who was not. These results suggest that TNCs may contribute to reduced car ownership and trip making.

Keywords: On-demand transportation, ride-sourcing, ride-hailing, transportation network companies, service suspension, travel behavior, vehicle ownership, mode shift

Suggested Citation

Hampshire, Robert Cornelius and Simek, Chris and Fabusuyi, Tayo and Di, Xuan and Chen, Xi, Measuring the Impact of an Unanticipated Suspension of Ride-Sourcing in Austin, Texas (May 31, 2017). Available at SSRN: https://ssrn.com/abstract=2977969 or http://dx.doi.org/10.2139/ssrn.2977969

Robert Cornelius Hampshire (Contact Author)

University of Michigan - Transportation Research Institute ( email )

500 S. State Street
Ann Arbor, MI 48109
United States

Chris Simek

Texas A&M University - Texas Transportation Institute ( email )

College Station, TX 77843-3135
United States

Tayo Fabusuyi

Carnegie Mellon University - College of Engineering ( email )

5000 Forbes Avenue
Pittsburgh, PA 15213
United States

Xuan Di

Columbia University ( email )

3022 Broadway
New York, NY 10027
United States

Xi Chen

University of Michigan at Dearborn ( email )

4901 Evergreen Road
Dearborn, MI 48128-1491
United States

Paper statistics

Downloads
408
Rank
58,609
Abstract Views
1,956