Nudging Take-up of the Earned Income Tax Credit in California

How to cite

Linos, Elizabeth; Ramesh, Aparna; Rothstein, Jesse; Unrath, Matt. 2020. "Nudging Take-up of the Earned Income Tax Credit in California." AEA RCT Registry. AEARCTR-0005734. https://doi.org/10.1257/rct.5734-2.0

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

Elizabeth Linos* , Aparna Ramesh , Jesse Rothstein , Matt Unrath

* Primary investigator

Registry ID
AEARCTR-0005734
DOI
https://doi.org/10.1257/rct.5734-2.0
First registered
Last updated
Status
Completed
Trial dates
2018-02-26 – 2020-01-21
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/5734

Abstract

The federal Earned Income Tax Credit (EITC) is the largest means-tested cash transfer program in the US, distributing more than $60 billion to low-income households annually. Nevertheless, an estimated one-fifth of households eligible for the EITC do not claim it. Drawing on previous evidence suggesting that low-cost behavioral interventions can address some of the administrative burdens that limit take-up, we ran six pre-registered field experiments testing the effect of outreach aimed at increasing claiming of the California and federal EITC. These studies were conducted in collaboration with the California tax and social service agencies and with a large NGO that conducts statewide outreach, and collectively involved more than one million subjects. We varied the content, design, messenger, and mode of messages within and across these studies. We find no evidence that any of our nudges affected households’ likelihood of filing a tax return or claiming the credit. While our messages spurred high levels of engagement with the outreach material, we conclude that even the most behaviorally informed outreach efforts cannot overcome the barriers to EITC claiming faced by low-income households who do not file returns.

Experimental design

See analysis plans for more information:
Study 1: https://osf.io/ct58w
Studies 2, 3, and 4: https://osf.io/z8ebc
Study 5: https://osf.io/msh7t
Study 6: https://osf.io/p2q4y

Design details

Randomization method
by computer
Randomization unit
households
Clusters
Study 1: 639,244 households Study 2: 96,370 households Study 3: 1,084,018 households Study 4: 204,285 households Study 5: 38,093 households Study 6: 47,104 households
Observations
Study 1: 648,890 individuals Study 2: 114,192 individuals Study 3: 1,236,174 individuals Study 4: 206,022 individuals Study 5: 109,046 households Study 6: 121,085 individuals
Arms
Study 1: 200,000 control; remainder treatment Study 2: 10,000 to each treatment arm; remainder control Study 3: 20% to each treatment arm; 20% control Study 4: 15,000 to each treatment group; remainder control Study 5: 50% treatment; 50% control Study 6: 25% to each treatment arm; 25% control
Primary outcomes
- Indicator for the presence of a matched return - Indicator for the presence of any EITC claim

Materials

Cited by

No citation data available for this registration yet.

Keywords