Researcher ORCID Identifier

0009-0005-9922-4251

Graduation Year

2026

Date of Submission

11-2025

Document Type

Campus Only Senior Thesis

Degree Name

Bachelor of Arts

Department

Economics-Accounting

Reader 1

Andrew Finley

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

© 2025 Melanie Chombo

Abstract

Take-up rates for the federal EITC have long remained incomplete. While eligible families are entitled to thousands of dollars in refundable credits each year, participation rates have conventionally remained between 75 and 86 percent. This is concerning because the EITC distributes over $60 billion annually to more than 25 million households. This thesis examines whether state EITC programs continue to influence federal EITC participation during 2009-2018, when e-filing adoption accelerated from approximately 60% to over 90%, fundamentally altering how taxpayers interact with the tax system. Neumark and Williams (2020) document that state programs modestly increased federal participation through 2008. However, whether these effects persist when most taxpayers access sophisticated filing assistance through commercial software remains an open question. Using county-level IRS data and a two-way fixed effects design, I leverage variation from six states that adopted EITC programs during this period, comparing changes in federal take-up in treatment counties against contemporaneous changes in control counties from fifteen states without programs. I estimate separate models for three income groups since state program effects may vary across the income distribution based on differential exposure to technology and information barriers. I find heterogeneous effects across the income distribution. For the lowest-income filers ($1,000-$10,000 AGI), refundable state EITC rate exhibits a positive association with federal take-up (coefficient = 0.0071, p < 0.05). A one standard deviation increase in refundable state generosity is associated with a 0.21 percentage point increase in federal take-up—approximately 2 additional claims per 1,000 filers. Non-refundable credits show no significant associations. These results demonstrate that state programs and technology-based filing assistance serve complementary rather than substitutable roles.

Download

This thesis is restricted to the Claremont Colleges current faculty, students, and staff.

Share

COinS