Assessing the potential of integrating public transportation with e-hailing services: a randomized trial in a ridesharing platform in São Paulo, Brazil

How to cite

Biderman, Ciro. 2023. "Assessing the potential of integrating public transportation with e-hailing services: a randomized trial in a ridesharing platform in São Paulo, Brazil." AEA RCT Registry. AEARCTR-0012627. https://doi.org/10.1257/rct.12627-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Ciro Biderman*

* Primary investigator

Registry ID
AEARCTR-0012627
DOI
https://doi.org/10.1257/rct.12627-1.0
First registered
Last updated
Status
Completed
Trial dates
2018-10-01 – 2019-08-30
Countries
Brazil (São Paulo)
JEL codes
R40, R48
Registry record
http://www.socialscienceregistry.org/trials/12627

Abstract

This study investigates whether providing passengers with a pecuniary incentive to use both public transportation and ridesharing in their daily commute increases the demand for e-hailing services as a means to complete first- or last-mile trips. To do so, we designed and implemented an experiment in partnership with one of São Paulo's largest e-hailing platforms in which some of the service's active users were randomly assigned eligibility to two daily discounts for app trips starting or ending in a train/subway station. Discounts were available over ten weekdays, in two randomly assigned regimes: two daily 20% discounts (limited to 10 BRL per trip), and two daily 50% discounts (also limited to 10 BRL). Using the random assignment and the user's history in the platform, we assess whether increased discounts indeed increase modal integration.

Intervention

Our interventions consist in providing daily coupons to a subset of the users in the ridesharing platform. These coupons allow users to have a discount in car trips which start or end in a train/subway station. We considered both 20% and 50% discount coupons, both of which were limited to 10 BRL off per trip.

Experimental design

We randomized three experimental regimes for a subset of the users in the ridesharing platform for which we collected baseline information on commuting habits and sociodemographic traits (importantly, no information on a prospective experiment was disclosed when collecting baseline data, with only a discount coupon being offered upon completion of the survey). One group was randomly assigned to the control arm, and was not informed about the experiment. A second group was randomly assigned to the 20% discount regime. This group was eligible to two daily 20% discounts, limited to 10BRL per trip, on app rides starting or ending in a train/subway station. A third group was randomly assigned to the 50% discount regime, which consisted of two daily 50% discounts, also limited to 10 BRL per trip, on app rides integrating with train/metro. Both discount regimes lasted for ten weekdays. Communication on the availability of discounts proceeded as follows. On November 26, 2018 (a Monday), all users selected into active treatment received SMS and push notifications informing them about the daily discount coupons to integrate their app rides with the public transportation system until the 30th (same week’s Friday). They also received daily reminders for the rest of the business week. On the next Monday (3rd of December), all users in these groups received similar notifications informing them that the discounts were also available for that whole week (until December 7, Friday), again with daily reminders following.

Design details

Randomization method
Randomization was performed in a computer in the office of the ridesharing service.
Randomization unit
Randomization was conducted at the individual (user) level.
Clusters
4650 users.
Observations
4650 users.
Arms
1522 users in the control arm, 1603 users in the 20% discount regime, and 1525 users in the 50% discount regime.
Primary outcomes
number of app rides integrating with train/subway, number of total app rides, number of total app rides minus number of app rides integrating with train/subway

Cited by

No citation data available for this registration yet.

Keywords