In light of this week’s Wall Street Journal article, The Ride-Hail Utopia That Got Stuck in Traffic, I feel the need to share a previous article I wrote. The below commentary was originally published in the Chicago Policy Review (University of Chicago) on December 26, 2019.
In the last decade, transportation network companies (TNCs) such as Uber and Lyft have significantly disrupted urban mobility. Commonly known as “ride-sharing,” TNCs’ app-based services provide millions of customers an alternative to traditional transportation modes such as public transit.
A subset of these services is “ride-splitting,” such as UberPool, which tends to be cheaper than individual ride-sharing trips. With ride-splitting, drivers make several stops to pick up customers going in the same general direction, similar to public transit. But unlike public transit, ride-splitting offers an on-demand and, in many cases, door-to-door service. Because of this option, UberPool customers may be willing to pay more for ride-splitting than they would for public transit. Despite the alleged benefits of ride-splitting, however, its advantages over public transit may be negligible.
A recent study from Joseph Schwieterman and C. Scott Smith takes a closer look at the matter. Using Chicago as their testing grounds due to the city’s relatively high population density and Chicago Transit Authority’s (CTA) broad public transit network, the authors conduct a “paired-observation” approach comparing UberPool with CTA services.
With this approach, they collect data from a sample of 50 paired trips (100 trips total) with one rider using UberPool and the other using CTA buses and trains. Paired riders depart from the same location and travel to the same destination. All trips depart from the city’s north side, and riders are assigned to destinations in either the downtown area or the city’s outer neighborhoods. These trips offer data on numerous variables including price, wait times, CTA seating availability, and number of UberPool stops.
For downtown trips, the authors find a near 50-50 split between modes in terms of fastest speed.
The authors find that travel time differences depended on destination. For downtown trips, the authors find a near 50-50 split between modes in terms of fastest speed. On trips to outer neighborhoods, UberPool is almost always faster. The average travel time for UberPool riders is 35 minutes and 52 seconds, while CTA riders average 48 minutes and 29 seconds.
Differences in travel times are explained in part by Chicago’s hub-and-spoke system of rail networks (all lines point to downtown), in which crosstown transit trips tend to require one or more transfers. CTA riders are also slowed down by the time it takes to walk to their destinations after getting off the train or bus.
The average cost per hour saved with UberPool is $34.10 (and significantly higher for downtown trips at $77.60).
Since UberPool is more expensive than CTA, the authors calculate the additional cost that riders pay to save time. The average cost per hour saved with UberPool is $34.10 (and significantly higher for downtown trips at $77.60). This is expensive for everyday riders and above the U.S. Department of Transportation’s recommended value of $14.05 in time savings per hour.
Schwieterman and Smith also conduct a multiple regression analysis to predict CTA and UberPool travel times. Certain variables, such as CTA transfers and UberPool stops, predict longer travel times for those modes, while proximity to transit predicts a faster ride via CTA than UberPool. In addition, the authors suggest that UberPool riders may have concerns about the lack of predictability of travel time and price, such as during times of surge pricing. Schwieterman and Smith say they would need a larger sample to examine weekend and late-night hours, the latter being when TNCs tend to implement surge pricing and when some transit service is lacking.
Speed and convenience are key factors in daily transportation choices, and this study may help local governments analyze travel behaviors and allocate municipal resources accordingly. Policymakers, however, should be mindful of how TNCs interact with urban transportation systems. Although TNCs are often faster, they may also contribute to more traffic congestion and reduced transit ridership, negatively impacting transit revenue, maintenance, and service. Legislators could cap TNC vehicle count, as some do with taxis, or levy fees on TNCs that pay into a transportation fund to mitigate congestion. More extensive research is needed to better assess transportation usage and determine whether TNCs are complementary to urban transport networks (as Uber claims) or competitive to existing systems.