Abstract

۱ | INTRODUCTION

۲ | THE EARNED INCOME TAX CREDIT AND ITS EFFECTS

۳ | DATA AND MODEL

۴ | RESULTS

۵ | DISCUSSION

ACKNOWLEDGMENTS

REFERENCES

Abstract

The Earned Income Tax Credit (EITC) transferred more than $67 billion to taxpayers in 2016. We estimate changes in spending that occur following EITC disbursement. We make three key advances in identifying the effect: using estimated EITC dollar benefits to differentiate the impact of low from high benefits; estimating the impact by liquid assets; and, employing a triple-difference model around the 1993 expansion and a separate triple-difference model over the number of children. We find that the EITC acts as supplementary SNAP benefits for those with small EITC benefits, while those with larger benefits increase spending on durables, particularly automobiles.

۱ | INTRODUCTION

The Earned Income Tax Credit (EITC) is one of the largest anti‐poverty programs for working age adults with children, transferring more than $67 billion per year (IRS, 2017). In 2018, the EITC along with the Child Tax Credit lifted 5.6 million individuals out of poverty, including over 3 million children, and reduced the severity of poverty for another 16.5 million people, including 6.1 million children (Center on Budget and Policy Priorities, 2019). The EITC transfers result in several beneficial impacts for recipients based on both the direct impact of the transfers but also through increasing earnings and employment. The greatest impacts tend to be for households with earnings up to the threshold phaseout part of the schedule ($18,340 dollars of earnings per year for a two dependent child household), which are more likely to be female‐headed single parent households, as well as for households without a member previously in the labor force (Dickert et al., 1995; Eissa & Hoynes, 2006). There is some evidence that likelihood of EITC receipt increases in economic downturns—enabling it to act as a social safety net (Bitler et al., 2017). Other benefits of the EITC include positive impacts on children’s test scores (Chetty et al., 2011; Dahl & Lochner, 2008).