Individual Submission Summary
Share...

Direct link:

Light-touch Changes to Prevent Unfinished Applications: Three Randomized Evaluations on SNAP Income Reporting

Thursday, November 5, 8:30 to 10:00am, Property: Boston Marriott Copley Place, Floor: 4th Floor, Room: Salon H

Abstract

Introduction
An estimated 7 to 8 million Americans are eligible for SNAP, but do not receive benefits. Administrative burdens contribute to underenrollment: beneficiaries must overcome barriers such as learning about their eligibility and complying with program requirements. Social and behavioral scientists and public policy scholars have extensively studied ‘light-touch’ interventions to encourage potential beneficiaries to apply for safety net programs, including SNAP. Yet, less work has explored whether ‘light-touch’ interventions introduced during the application process can lessen burdens  in ways that could increase program take-up. 

Purpose
We help fill this evidence gap by presenting findings from three randomized field trials of light-touch interventions encountered by potential beneficiaries  after starting SNAP applications. One study (N=32,879) examines whether giving single-income households different sets of income reporting options impacts application submission  rates and the amount of income reported. In the same state, a second study (N=20,549) examines whether providing multiple-income households  with a more structured way to report their income impacts these same outcomes. In a second state, a third study (N=33,547) builds evidence on alleviating burdens after households submit applications, but before they provide verification documents. This evaluation examines whether being sent a text campaign encouraging the use of a redesigned document uploader impacts SNAP enrollment compared to being sent basic text reminders.

Data and Research Method
Across all three evaluations, the sample includes online SNAP applicants who applied as part of the regular enrollment process. Random assignment allows for the identification of causal impacts. Administrative data collected during the SNAP application and enrollment process are used to measure outcomes. 

Findings
The light-touch interventions did not induce meaningful impacts on benefit access as measured by application submissions and enrollment. In Study 1, submission rates among single-income households were not statistically different from one another (p = 0.54) when applicants were given a pay-stub reporting option (66.5%), an annual salary reporting option (66.2%), or both options (65.9%). In Study 2, submission rates for multiple-income households given an unstructured reporting option (58.9%) were not significantly different (p = 0.14) when given a structured reporting option (57.9%). In Study 3, 21.1% of households in the basic text group were denied benefits due to incomplete information and only 43.3% successfully enrolled. Sending texts encouraging use of the uploader reduced denials due to incomplete information by 1.1 percentage points (p=0.01), but did not increase enrollment (p=0.23).

Implications
While these light-touch interventions did not move the needle on access to benefits, the findings highlight a need for reducing administrative burdens after households start SNAP applications. At most two-thirds of households succeeded in submitting applications. Even after overcoming that application submission barrier, less than half successfully enrolled. Future research could explore whether light-touch interventions improve accuracy and reduce administrative barriers in unmeasured ways or whether turning to higher-touch interventions is necessary to reduce burdens and improve access.

Author