• Continuous Treatment Difference-in-Differences with Unknown Controls: A Data-Driven Approach

• w/ Thomas Helgerman

• This paper studies difference-in-differences research designs where all observations receive a continuous treatment (or dose) in response to an aggregate policy, so there is no group that is ex post unexposed. This contrasts with the recent literature examining difference-in-differences estimators, which typically requires a subset of observations go untreated. When treatment takes effect only after some cutoff value, the Minimum Effective Dose (MED), we can target the Average Treatment Effect on the Effectively Treated (ATET) estimand under non-parametric assumptions on the dose response function. In a hold-out sample, first estimate the MED with results from the pharmacology literature. This cutoff defines the estimating equation of the ATET in a second step using remaining observations. The Split Sample estimator is asymptotically conservative in that it never erroneously categorizes treated units as untreated in the limit, even if the cutoff is on the boundary of the parameter space. This also implies it is attenuated for the ATET. We test the MED existence assumption by permuting dose values below the estimated cutoff using observations in the second step. Finally, we use the Smoothed Bootstrap estimator of the standard error of the ATET estimate to properly capture uncertainty over the unknown cutoff. Simulations show the estimator performs well in finite samples.

• COVID-19 and Stay-At-Home Orders: Identifying Event Study Designs with Imperfect Testing 

• w/ Jaedo Choi, Thomas Helgerman, Nishaad Rao, and Taeuk Seo

• in CEPR Covid Economics working paper series

• This paper estimates the dynamic effect of Stay-At-Home (SAH) orders on the transmission of COVID-19 in the United States. Identification in this setting is challenging due to differences between real and reported case data given the imperfect testing environment, as well as the clearly non-random adoption of being treated by SAH. We extend a Susceptible-Infected-Recovered (SIR) model from Epidemiology to account for endogenous testing at the county level and exploit this additional structure to recover identification. With the inclusion of model-derived sufficient statistics and fixed effects, SAH orders have a large and sustained negative effect on the growth of cases under plausible assumptions about the progression of testing. Point estimates range from a 44% to 54% reduction in the growth rate of cases one month after a SAH order. We conclude with a discussion on extending the methodology to later phases of the pandemic.

• Emigration Increases Schooling at Home: Evidence from Romania

• Remittances are a large source of income in developing countries but come at the cost of losing workers to foreign labor markets. Fears of Brain Drain abound when emigrants are college-educated but may be assuaged by Brain Gains at home. These compensating effects are often motivated by an increased college wage premium, but also arise when poor people with constrained access to credit use remittances to invest in schooling. Dominance of the premium channel can make short-run brain gains transitory, especially when mobility becomes easier, while the remittance channel increases the likelihood they persist. To infer their relative contributions to reduced-form estimates, I show that remittances are more dominant whenever Brain Gains are accompanied by a shrinking schooling gap between constrained and unconstrained. I study Romania since 1990, where over 20% of the population (six million people) emigrated during a period of rapid economic growth but persistent gaps between rich urban, and poor rural areas. In 2002, Schengen visa requirements were waived creating heterogeneous opportunities to emigrate that I capture with a continuous measure of foreign migrant presence. Difference-in-differences estimates show increases in college enrollment and graduation flows in response to the shock, but no resulting increase in future stocks of higher educated. Urban-rural schooling gaps also do not shrink, implying the premium channel generated most of the short-run brain gains, which disappeared with Romania’s subsequent European integration.

Work in Progress

• Immigration and Voting Trends: Disentangling Economic and Cultural Effects

• Multinational Production and the Propagation of Immigration Shocks (w/ Luis Espinoza)

Ross School of Business – Primary Instructor

University of Michigan

• Principles of Applied Microeconomics (4 semesters)

Economics Department – Graduate Student Instructor

University of Michigan

• Econometrics for Masters Students (1 semester)

• Microeconomics for Masters Students (1 semester)

• Intermediate Econometrics (1 semester)

• Intermediate Macroeconomics (6 semesters)

• Intermediate Microeconomics (1 semester)

• Introduction to Statistics and Econometrics II (3 semesters)

• Principles of Microeconomics (2 semesters)

Economics Department – Primary Instructor

Alma College

• Intermediate Microeconomics (1 semester)

Economics Department – Primary Instructor

University of Detroit, Mercy

• Masters Microeconomics (1 semester)

• Seminar in Fiscal and Monetary Policy (1 semester)

• Financial Economics II (1 semester)

• Money and Capital Markets (1 semester)