Matthew Unrath

Research

Publications

Measuring Take-up of the California Earned Income Tax Credit with State Administrative Data

Online Appendix

with John Iselin and Taylor MackayJournal of Public Economics, Vol. 227, November 2023.

Policy report
Media Coverage: Sac Bee, Capital Public Radio, KPBS, LA Times

The Earned Income Tax Credit (EITC) is the largest cash-based means-tested transfer program in the United States. In 2021, 31 million households received $64 billion from the federal EITC. Twenty-eight states also offer eligible taxpayers a supplement to the federal program. An estimated one-fifth of eligible households fail to claim the federal credit, but little is known about take-up of these state programs. We use administrative data from California on the population of Supplemental Nutrition Assistance Program (SNAP) recipients linked to state tax records to estimate the number of households who are eligible for California’s supplement to the federal EITC (CalEITC) but do not claim it. We find that over 400,000 households who received SNAP benefits and who were eligible for the state EITC in 2017 did not receive the credit. This includes approximately 40,000 eligible households who claimed the federal EITC but not the state credit; nearly 98,000 eligible households who filed a state tax return but did not claim the state or federal credit; and roughly 270,000 eligible households who did not file a state tax return. The corresponding take-up rate for the CalEITC among eligible SNAP-enrolled households was 54%. Altogether, these households left a total of $71 million in state EITC funds on the table. If received, these credits would have increased incomes among these households by 2.7% and increased total state EITC outlays by 20%.

Can Nudges Increase Take-up of the Earned Income Tax Credit?: Evidence from Multiple Field Experiments

with Elizabeth Linos, Allen Prohofsky, Aparna Ramesh, and Jesse RothsteinAmerican Economic Journal: Economic Policy, Vol. 14, No. 4, November 2022.

Policy reportMedia Coverage: NPR, Vox's The Weeds

The Earned Income Tax Credit (EITC) distributes more than $60 billion to over 20 million low-income families annually. Nevertheless, an estimated one-fifth of eligible households do not claim it. We ran six pre-registered, large-scale field experiments, involving over one million subjects, to test whether “nudges” could increase EITC take-up. Despite varying the content, design, messenger, and mode of our messages, we find no evidence that they affected households’ likelihood of filing a tax return or claiming the credit. We conclude that even the most behaviorally informed low-touch outreach efforts cannot overcome the barriers faced by low-income households who do not file returns.

Measuring the Labor Market at the Onset of the COVID-19 Crisis

with Alexander W. Bartik, Marianne Bertrand, Feng Lin, and Jesse RothsteinBrookings Papers on Economic Activity, Summer 2020.
Media Coverage: NYT, NYT, WSJ, Slate, WaPo

We use traditional and non-traditional data sources to measure the collapse and subsequent partial recovery of the U.S. labor market in Spring 2020. Using daily data on hourly workers in small businesses, we show that the collapse was extremely sudden -- nearly all of the decline in hours of work occurred between March 14 and March 28. Both traditional and non-traditional data show that, in contrast to past recessions, this recession was driven by low-wage services, particularly the retail and leisure and hospitality sectors. A large share of the job loss in small businesses reflected firms that closed entirely. Nevertheless, the vast majority of laid off workers expected, at least early in the crisis, to be recalled, and indeed many of the businesses have reopened and rehired their former employees. There was a reallocation component to the firm closures, with elevated risk of closure at firms that were already unhealthy, and more reopening of the healthier firms. At the worker-level, more disadvantaged workers (less educated, non-white) were more likely to be laid off and less likely to be rehired. Worker expectations were strongly predictive of rehiring probabilities. Turning to policies, shelter-in-place orders drove some job losses but only a small share: many of the losses had already occurred when the orders went into effect. Last, we find that states that received more small business loans from the Paycheck Protection Program and states with more generous unemployment insurance benefits had milder declines and faster recoveries. We find no evidence so far in support of the view that high UI replacement rates drove job losses or slowed rehiring substantially.

Working Papers

Work Requirements and Child Tax Benefits

with Jacob Goldin, Tatiana Homonoff, Neel Lal, Ithai Lurie, and Katherine Michelmore
[Subsumes Employment Responses to Earned Income Tax Benefits: Evidence from California's Young Child Tax Credit]

Many U.S. safety-net programs condition benefit eligibility on work. Eliminating work requirements would better target benefits to the neediest families but attenuates pro-work incentives. Using administrative records, we study how expanding a California child tax credit to non-workers affected maternal labor supply. We rely on quasi-random birth-timing and a novel method for using placebo analyses to maximize estimator precision. Eliminating the work requirement caused very few mothers to exit the labor force; our 95% confidence interval excludes reductions over one-third of one percent. Our results suggest expanding tax benefits to the lowest-income families need not meaningfully reduce workforce participation.

National Experimental Well-Being Statistics

with Adam Bee, Joshua Mitchell, Nikolas Mittag, Jonathan Rothbaum, Carl Sanders, and Lawrence Schmidt

Version 2

Accurately measuring household income and poverty is essential to understanding the nation’s overall economic well-being. Many studies show that measurement error stemming from unit nonresponse, item non-response and misreporting biases key official statistics such as mean or median income and the official poverty rate. The direction of bias differs between these sources of measurement error. Unit and item nonresponse have been found to bias income up and poverty down (Rothbaum et al., 2021; Rothbaum and Bee, 2022; Bollinger et al., 2018; Hokayem, Raghunathan and Rothbaum, 2022), while misreporting can bias income down and poverty up (Bee and Mitchell, 2017; Meyer et al., 2021b; Larrimore, Mortenson and Splinter, 2020). Since these error components are typically studied in isolation, their overall impact on the accuracy of survey estimates remains unclear.This paper summarizes the National Experimental Well-being Statistics (NEWS) Project, which integrates this research and address each of these sources of bias simultaneously in order to produce more accurate estimates of household income and poverty. The NEWS project makes three unique contributions. First, we address as many sources of measurement error as we can simultaneously – including unit and item nonresponse and underreporting in surveys as well as the various challenges in administrative data such as measurement error, conceptual misalignment, and incomplete coverage. Second, we bring together all of the available survey and administrative data, which allows to address many of the shortcomings of individual data sources. Third, we propose a model to combine survey and administrative earnings data given measurement error in both sources, replacing ad hoc assumptions that have been used in prior work.

Targeting, Screening, and Retention: Evidence from the Supplemental Nutrition Assistance Program in California

Online Appendix
Policy report
Media Coverage: The Daily Californian, Capital Public Radio, KPBS, The AtlanticPolicy Mentions: White House Burden Reduction Initiative

Many households eligible for the Supplemental Nutrition Assistance Program (SNAP) do not enroll. Using enrollment histories for all SNAP participants in California between 2005 and 2023, this paper documents how procedures used to verify eligibility lower retention and contribute to incomplete take-up. Program exits largely coincide with reporting schedules, and the majority of cases that leave appear income eligible in the months before and after their exit. I also show that these reporting requirements most deter enrollment among relatively more advantaged recipients. Cases with higher earnings, lower benefit amounts, children, and lower levels of predicted food insecurity are more likely to exit in reporting months. The paper quantifies this trade-off between take-up and targeting which characterizes the screening process. Using enrollment effects from a reform that widened the reporting interval in California, the paper concludes that reducing the frequency of these verifications is an efficient way to improve participation, despite worse targeting, because of how costly these ordeals are to administer.

Beating the Clock: Using Year-end Changes to Identify Intensive Margin Labor Supply Responses to Taxation

Identifying the effect of tax policy on the labor supply of individuals who would work regardless has been a longstanding empirical challenge. This paper proposes a new strategy for identifying workers’ intensive-margin labor supply elasticity using within-year variation in anticipated year-end tax rates. I modify the standard non-linear budget set approach to include uncertainty about future employment. With uncertainty, households must forecast their annual income in order to anticipate the average and marginal tax rates that apply to their earnings. Workers only learn their true tax rates as the year progresses and they realize their employment history. Using survey and administrative data, I conclude that low-income households’ labor supply responds more to expected tax rates at the end of the year, when certainty about annual income is greatest. I use the excess sensitivity to tax incentives near the end of the year, relative to other periods, to estimate an intensive margin labor supply elasticity between .08 and .18. This response is identified largely from non-linearity in the Earned Income Tax Credit (EITC) schedule and implies a larger intensive margin response to this program than previous estimates.

Filling in the Blanks: Improving SNAP Retention through Pre-Population?

with Pamela Herd, Donald Moynihan, and Gwen RinoDraft available upon request.

Policymakers have shown an increasing willingness to reduce administrative burdens that are impeding access to critical social safety net programs. One commonly proposed tool, which was highlighted in a recent Presidential Executive Order, involves pre-filling forms for applicants with existing administrative data. We test the effectiveness of “pre-filled forms” as a technical instrument to reduce compliance costs and administrative burdens. We argue that the nature of this intervention, and its effects, depends greatly on how it is implemented, contingent on factors such as policy design. We study the effect of pre-populating eligibility renewal forms when beneficiaries clients' recertify their eligibility in the Supplemental Nutrition Assistance Program (SNAP). As many as one-third of beneficiaries lose coverage during recertification, often due to their failure to navigate administrative procedures. Using a field experiment involving over 20,000 cases, we find that being offered the chance to use a webform pre-filled with information from their initial application increases the chance that clients submit the form by 11 percent. Our analysis shows that, in our case, the effects of pre-filling are concentrated at the initial stages of the application, where applicants authenticate their status by providing personal information and case number. The results point to the promise of reducing churn off programs at the re-certification stage by making administrative data that the client previously provided easily accessible to them, and paying particular attention to early stages of the process.

Married... With Children?: Assessing the Alignment Between Survey Households and Tax Units

Draft available upon request.

To produce official income and Supplemental Poverty Measure (SPM) statistics, the Census Bureau uses an in-house tax model to estimate households’ tax liabilities. This paper assesses the accuracy of the first stage of this tax model: the construction of likely tax units from the person-level Current Population Survey Annual Social and Economic Supplement (CPS ASEC) file. I compare the composition of those tax units to actual 1040 rosters. I document non-trivial misalignment between the two, especially with respect to claiming of dependents. I also identify the impact of dependent misassignment on Census’s estimates of after-tax income statistics.

Select Works in Progress

The Long-Run Effects of Growing Up in a High-Crime Neighborhoodwith Nick Gebbia and Jonathan Rothbaum

Understanding Existing Insurance from Job Loss in the United States

with Sreeraahul Kancherla and Carl McPherson

Page updated

Google Sites

Report abuse