Evaluating the Lump Sum Tax Refund and Household Spending

How to cite

Bertrand, Marianne; Aurelie; Jones, Damon. 2017. "Evaluating the Lump Sum Tax Refund and Household Spending." AEA RCT Registry. AEARCTR-0002440. https://doi.org/10.1257/rct.2440-2.0

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

Marianne Bertrand* , Aurelie , Damon Jones

* Primary investigator

Registry ID
AEARCTR-0002440
DOI
https://doi.org/10.1257/rct.2440-2.0
First registered
Last updated
Status
In development
Trial dates
2017-02-01 – 2022-01-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/2440

Abstract

The Earned Income Tax Credit (EITC) is the largest means-tested cash transfer program in the United States, and accounts for a significant share of income for those who receive it. As opposed to most other federal and state transfer programs, the credit is disbursed only once a year, resulting in uneven cash flow for recipients that may undermine their economic security. Interestingly however, past programs that have offered a distributed payment schedule have suffered from low take-up. The purpose of this research is to explore the determinants of demand for, and the effects of, a novel periodic EITC payment program. In cooperation with the Center for Economic Progress, researchers will first conduct pilot activities to learn more about demand for periodic payment and how best to implement it. We hypothesize that offering periodic payment outside of tax season (when the loss of a lump sum may loom large relative to year-long stability) and providing framing for this offer could increase demand for periodic payments. Based on findings from the pilot, the full-scale study will use a randomized controlled trial to test the effects of a periodic EITC payment on participant financial security, purchasing power, and labor-force behavior.

Experimental design

This study is currently in the pilot phase. The work that is being done now will determine the feasibility of conducting a future randomized controlled trial and will be divided across two phases.

For Phase 1, we plan to conduct four quarterly surveys using a panel of at least 1,500 low-income tax filers who are eligible for the EITC. These surveys will elicit respondents’ preferences for a periodic payment model, measuring demand across differing program structures and framing conditions, in addition to demand over time. Ultimately, survey responses will indicate whether demand for a periodic payment is high enough under certain conditions (structure, framing, and/or time of year) to make feasible a full-scale study. Moreover, if we find that there is indeed demand for the program, the survey responses will help us determine when and how to target the intervention.

For the survey, respondents will be randomized to receive either a question about their preferences for an advanced periodic payment versus a lump sum or a question about their preferences for a deferred periodic payment versus a lump sum. Additionally, respondents will be randomly assigned to receive different framings of the same question and compare results across framings. We hypothesize, for instance, that framing the lump sum as an interest-free loan to the government may lower preference for it. Similarly, framing periodic payments as seasonal mental accounts (e.g., “back-to-school money,” “holiday season money”) may increase demand for them. We will also ask respondents about their financial planning decisions related to their refund. Through these questions we aim to learn whether EITC recipients intend to smooth their spending of the credit, but overestimate their ability to do so.

Based upon sufficient demand from the surveys—defined as 50 percent or more of respondents preferring a periodic payment model to the current lump sum at one or more times in the year—we will engage in Phase 2 of the pilot.This second stage will include a small operational pilot (comprising of approximately 50 - 75 people) to determine the logistics of administering the periodic payments for a full-scale study. A key activity here will be recruiting a financial service provider to manage the distribution of the periodic payments.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual
Clusters
N/A
Observations
1500
Arms
750
Primary outcomes
Debt-to-income ratio, number accounts delinquent, personal savings, monthly spending, hours worked

Cited by

No citation data available for this registration yet.

Keywords