Working Papers

Estimating the Average Treatment-on-the-Treated Effects of DACA
Abstract (Click here to expand)

I examine the labor market response of undocumented youth that participated in the Deferred Action for Childhood Arrivals (DACA). DACA provides temporary work authorization and deferral from deportation to eligible undocumented youth. I use data from the U.S. Citizenship and Immigration Services to construct a probabilistic measure for the unobserved DACA participation. Using ACS data, I estimate a two-sample model of the effect of participating in the DACA program. I also estimate the spillover effects of DACA on eligible but non-participating undocumented youth. I find that DACA significantly improved labor market outcomes of DACA recipients, with magnitude of the treatment-on-the-treated effects at least twice as large as the intent-to-treat estimates obtained from using only the observed eligibility indicator. I also find an increase in school attendance among DACA recipients. Evidence of a negative spillover effect on eligible non-participants is documented with a decrease in labor force participation and school attendance.

Credible Interval Estimates of the Size and Legal Composition of the US Foreign-Born Population
(w/ Christopher Bollinger) 
Abstract (Click here to expand)

Government agencies and academic researchers commonly report the size and legal composition of the foreign-born population as point estimates. These survey estimates are impacted by both sampling and nonsampling error. This paper considers nonsampling error from item nonresponse in estimates from the American Community Survey. Our approach allows us to form credible interval estimates without assuming values from missing data that consider all uncertainty related to item nonresponse. Without assuming the distribution of citizenship status among non-respondents, the size of the foreign-born population in the US is estimated to be between 40.4 and 59.4 million in 2019 compared to the Census estimate of 44.9 million. When taking into account item nonresponse from all questions used in the legal status imputation procedure, the size of the undocumented population falls between 7.3 and 23.3 million compared to the widely accepted estimates of 11 million undocumented immigrants.

​The Effectiveness of Employment-Based Tax Credits Under Labor Market Frictions
(w/ Hyein Kang)
Abstract (Click here to expand)

One of the central goals of employment-based tax credits is to incentivize work. This paper considers the effectiveness of these tax credits when individuals face labor market frictions. We analyze the labor market response of unauthorized single mothers to the 2008-09 Child Tax Credit expansion in an event study framework. Due to their lack of legal status, unauthorized single mothers face severe labor market frictions. Using data from the American Community Survey, we find large increases in the labor force participation and employment rate of single unauthorized mothers after the expansion relative to unauthorized childless women. The initial increase in employment is partially driven by employment into the informal labor market (proxied by self-employment). The increase in labor market participation coincides with a sharp decrease in school attendance among unauthorized single mothers relative to the control group.

Estimating the Probability of non-U.S. Residence Among People in U.S. Administrative Records
Abstract (Click here to expand)

In an era of declining survey response rates, administrative records of U.S. residents can be used to improve the coverage and accuracy of U.S. population statistics. When using administrative records to identify U.S. residents, analysts face a tradeoff between over- and under-inclusion. Emigration from the U.S. is not consistently captured in administrative data, so improving coverage of current residents while excluding former residents is a particular challenge. This report provides an overview of the development of a method for estimating the probability that a person observed in administrative data around a reference date is residing outside the U.S. The analysis uses the 2020 extract of the Census Bureau’s Demographic Frame, which compiles person-place records from more than 20 administrative data sources, to identify people likely to have been U.S. residents on the reference date for the 2020 Census. The Demographic Frame is augmented with signs of living outside the U.S., such as foreign addresses, from the contributing sources. Using absence from the 2020 Census as a proxy for absence from the U.S., I estimate the probability of emigration as a function of these signs. Finally, I compare population estimates adjusted for migration to unadjusted estimates.

Work in Progress

On the Quality of Address Data in Administrative Records and Its Impact on Data Linkage Rates

Income Tax Credits and Immigrant Families" (w/ Hyein Kang)

Imputing Treatment: Misclassification Bias in the Effects of DACA

Immigrant’s Legal Status and English Skill Acquisition