Nudging Towards Stability: Addressing Churn in the Supplemental Nutrition Assistance Program (SNAP)

How to cite

Gilmore, Dallas; Lee, Monica; Loeb, Susanna. 2025. "Nudging Towards Stability: Addressing Churn in the Supplemental Nutrition Assistance Program (SNAP)." AEA RCT Registry. AEARCTR-0017396. https://doi.org/10.1257/rct.17396-1.0

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

Dallas Gilmore* , Monica Lee , Susanna Loeb

* Primary investigator

Registry ID
AEARCTR-0017396
DOI
https://doi.org/10.1257/rct.17396-1.0
First registered
Last updated
Status
Completed
Trial dates
2024-03-04 – 2025-02-28
Countries
United States of America (Massachusetts )
Registry record
http://www.socialscienceregistry.org/trials/17396

Abstract

We study whether simple behavioral nudges can reduce churn in SNAP—when participants exit the program and re-enroll shortly after. In a 2023–2024 randomized controlled trial involving 119,882 Massachusetts households facing a recertification or interim report deadline, participants received either no messages or one of two text-message reminders designed to reduce complexity and mitigate inattention. The interventions increased submission rates by 2.5 percentage points (3 percent), improved completion rates by 3 percentage points (5 percent), and reduced churn by 0.6 to 0.8 percentage points (10–14 percent), with the simpler reminder performing slightly better. The findings show that low-cost, light-touch messages improve continuity in means-tested programs.

Experimental design

This study evaluates whether text-message reminders can reduce churn among SNAP participants who are due for a recertification or interim report. Eligible households that have provided a phone number are randomly assigned to one of three groups: a control group receiving standard DTA communications; Treatment 1, which receives text reminders containing informational links and a phone number for assistance; and Treatment 2, which receives reminder-only messages. The first text is sent 44 days before each household’s deadline. If a household has not submitted required paperwork, a second text with the same content is sent 25 days before the deadline. This design allows us to estimate the effects of reminders on submission, completion, and churn, and to compare reminders with and without embedded informational supports.

Design details

Randomization method
Randomization was conducted by the Massachusetts Department of Transitional Assistance using their internal computer system. Eligible SNAP households were individually and randomly assigned to either the control group or one of the two treatment groups. Assignment was automated and not influenced by researchers.
Randomization unit
The randomization unit is the individual SNAP case (head of household).
Clusters
0
Observations
144,276 heads of household
Arms
57734 heads of household control 43294 heads of household treatment 1 43248 heads of household treatment 2
Primary outcomes
1. Churn: An indicator for whether a SNAP household exits the program at its recertification or interim report deadline and then re-enrolls within 120 days. 2. Submission Rate: An indicator for whether the household submits the required recertification or interim report form by the deadline. 3. Submission Timing: Number of days between the text message and the household’s submission of required paperwork. 4. Completion Rate: An indicator for whether the household fully completes the recertification or interim report process and remains continuously active without disruption. 5. Completion Timing: Number of days between the text message and the household’s completion of the recertification or interim report process.

Cited by

No citation data available for this registration yet.

Keywords