The Effect of Price and Travel Time on Shared Rides and VMT in the Context of On-demand Ridehailing 

Abstract
Transportation network companies (TNCs) offer several types of service, including private ridehailing (private trips) and dynamic ridesharing (shared trips). Policymakers responsible for transportation demand management have an interest in encouraging shared trips over private trips to promote more efficient use of the transportation network and reduce vehicle miles traveled. Doing so requires an understanding of 1) market segments (which trips are more likely to be shared) and 2) responses to differences in price and travel time (how these factors affect the choice to take a shared ride). To fill this need, this presentation analyzes a survey data of users of a major TNC to provide insights into the effectiveness of incentives to increase sharing among multiple market segments.

The first part of this presentation will describe the survey data – observed trip and user characteristics for 4,365 ridehailing trips in 15 cities – then present the final results of a new, but robust methodology for analysis of these users along valuable dimensions. By adding new detail to prior studies of TNC users, the presentation will show that sharing is more common among younger, lower-income riders who use transit more frequently and have less access to private vehicles. The presentation will also describe sharing rates according to trip type, finding more sharing for weekend trips, longer trips, and evening trips.

Beyond descriptive analysis of sharing, the second part of the presentation will focus on discrete choice models that evaluate how the choice to share would change in scenarios where respondents faced different price and travel time options. The presentation will argue that a $1 per mile increase in the relative price of private and shared rides increases the probability of sharing by over 8 percent. What’s more, the presentation will demonstrate that sharing discounts of up to 75 percent are not enough to convince one-third of TNC users to share rides, even when private and shared travel times are equal. Cross-tabulations will show that these price-insensitive riders are older, higher-income, and infrequent transit users. Regarding travel time, the presentation will show that a decrease in the relative travel time of 1 minute per mile increases the probability of sharing by over 33 percent. For both price and time, the presentation will explain how these estimates vary according to market segments that could be the target of time- or price-based policies: trips to/from airports, trips taken by lower-income riders, and trips starting or ending in dense office districts.

This presentation offers an important opportunity to planners seeking to increase vehicle occupancy and optimize use of the existing transportation network. The presentation will conclude by applying the results of this analysis to estimate the effect of time- and price-based policy scenarios on the use of shared rides and vehicle miles traveled. These findings will leave the audience with a greater understanding of user behavior in shared mobility and the ability to make informed decisions about incentives, regulations, and market signals as they promote equitable, efficient, and environmentally friendly mobility options.