Governments and non-profits devote substantial resources to increasing take-up of the Earned Income Tax Credit (EITC) through educational outreach. We study a different approach -- policies that encourage tax filing -- to determine whether such policies can be an effective way to increase take-up of tax-administered social benefits. Specifically, we study the effect of IRS letters about free tax preparation on filing and EITC take-up using a large field experiment.
External Link(s)
Citation
Goldin, Jacob et al. 2021. "Tax Filing and Take-Up: Experimental Evidence on Tax Preparation Outreach and EITC Participation." AEA RCT Registry. January 20. https://doi.org/10.1257/rct.7071-1.0.
In this intervention, the IRS sent letters to randomly assigned prior non-filers detailing free tax preparation methods available to low- and middle-income taxpayers.
Intervention Start Date
2019-03-16
Intervention End Date
2020-12-31
Primary Outcomes (end points)
We measure the effect of receiving a letter on filing, tax preparation method, EITC claims, and whether the taxpayer claimed a refund in the current and subsequent tax year.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Our sample consists of the universe of taxpayers who did not file a tax return in the year prior to the intervention and appeared income-eligible for free tax preparation assistance. Sample members were randomly assigned to receive letters from the IRS describing either the Free File program, the Volunteer Income Tax Assistance (VITA) program, or both.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
Randomization was conducted at the individual level.
Was the treatment clustered?
No
Sample size: planned number of clusters
n/a
Sample size: planned number of observations
Total observations: 1.8 million. Treatment group members: 56,000.
Sample size (or number of clusters) by treatment arms
Five treatment letter variants with roughly 11,200 individuals per variant. Remainder of the sample assigned to the control group (no letter).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)