The effect of public transportation fare subsidies on mobility and socioeconomic outcomes

How to cite

Chizeck, Seth; Mbonu, Oluchi. 2023. "The effect of public transportation fare subsidies on mobility and socioeconomic outcomes." AEA RCT Registry. AEARCTR-0011001. https://doi.org/10.1257/rct.11001-1.0

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

Seth Chizeck* , Oluchi Mbonu

* Primary investigator

Registry ID
AEARCTR-0011001
DOI
https://doi.org/10.1257/rct.11001-1.0
First registered
Last updated
Status
On going
Trial dates
2022-11-17 – 2025-12-31
Countries
Private
JEL codes
R48
Registry record
http://www.socialscienceregistry.org/trials/11001

Abstract

Public transportation fares are often unaffordable for low-income riders. We conduct an experiment that provides public transportation fare subsidies to 9,574 working-age residents of Allegheny County, Pennsylvania who receive Supplemental Nutrition Assistance Program (SNAP) benefits. Study participants were randomly assigned to receive either a 0%, 50%, or 100% discount on all Pittsburgh Regional Transit (PRT) bus and light rail trips for 12 months. Study outcomes will include public transportation use, transportation mode substitution, earnings and labor supply, health care utilization, self-reported measures of well-being, and spatial mobility as measured by cell phone GPS data.

Intervention

The experiment has three arms with equal allocation probabilities, randomized at the individual level:

1. The control group, which does not receive any discount on Pittsburgh Regional Transit (PRT) trips
2. The half-off treatment group, which receives a 50% discount on all PRT trips for 12 months
3. The free treatment group, which receives a 100% discount on all PRT trips for 12 months.

Participants are assigned a PRT farecard, called a ConnectCard, that is pre-programmed with the appropriate fare discount level. The ConnectCards for participants in the 0% and 50% subsidy groups contain $10 of preloaded fare value to encourage use of the card. Once this initial $10 balance runs out, participants in the 0% and 50% subsidy groups must load their own fare product onto the card in order to continue using it.

The 50% group ConnectCards automatically apply a 50% discount to any stored cash or timed pass that is loaded onto the card, with the exception of an annual pass. For context, a single PRT ride normally costs $2.75. The 50% subsidy group pays only $1.35 per ride. A 31-day unlimited ride pass normally costs $97.50, and costs $48.75 for the 50% subsidy group. The ConnectCards for the 100% subsidy treatment arm are programmed to allow unlimited free-fare taps on all PRT vehicles. Participants with these free-fare cards do not need to load any cash or timed pass onto the card.

Experimental design

The experiment has three arms with equal allocation probabilities, randomized at the individual level:

1. The control group, which does not receive any discount on PRT trips
2. The half-off treatment group, which receives a 50% discount on all PRT trips for 12 months
3. The free treatment group, which receives a 100% discount on all PRT trips for 12 months.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual
Clusters
9,574 individuals
Observations
9,574 individuals
Arms
3,158 individuals in control group, 3,251 individuals in 50% subsidy group, and 3,165 individuals in 100% subsidy group
Primary outcomes
We will estimate treatment effects on the following primary outcome measures: • Quarterly earnings: We will use Pennsylvania Unemployment Insurance records to measure participants’ total earnings in the third calendar quarter after the calendar quarter in which they were randomly assigned. This corresponds to the participant’s earnings at roughly eight to twelve months after their date of random assignment, depending on where their random assignment date fell in a calendar quarter. • Primary health care visits: We will measure participants’ total number of primary care visits taken in the first nine months after random assignment. We will measure this outcome using a combination of Medicaid claims and private health insurance claims.

Cited by

No citation data available for this registration yet.

Keywords