Research on Hugo Sant'Anna

Mass Incarceration Reforms: The Alabama Experience

hsantanna@uab.edu (Hugo Sant'Anna) — Fri, 11 Sep 2026 00:00:00 -0400

Alabama’s prison system has been among the most scrutinized in the United States, operating under active federal litigation over unconstitutional conditions while simultaneously enacting some of the most significant sentencing reforms in its history. Between 2013 and 2016, Alabama implemented a sequence of graduated criminal justice reforms: the transition from voluntary to presumptive sentencing guidelines, and the creation of a new Class D felony classification that established a presumption against incarceration for low-level drug and property offenses. Using facility-level data from the Alabama Department of Corrections and a state-year panel from the Vera Institute of Justice, we evaluate the effect of these reforms on both the level and racial composition of Alabama’s prison population. We employ a two-way fixed effects event-study design with a comparison group of nine states that remained untreated throughout the sample window. We find that Alabama’s total custodial prison population declined by approximately 13.5 percent between 2014 and 2019. More strikingly, the black share of new prison admissions, or the flow into prison, declined by approximately 19.6 percent relative to comparison states in the post-reform period, while the black share of the stock of the standing prison population declined by approximately 7 percent. The gap between the racial composition of new admissions and the existing prison stock widened by approximately 13 log points post-reform confirming a sharp and immediate flow-margin response.

Difference-in-Differences with “Bad Controls”

hsantanna@uab.edu (Hugo Sant'Anna) — Tue, 04 Aug 2026 00:00:00 -0400

This paper considers difference-in-differences identification strategies when the parallel trends assumption holds after conditioning on covariates that may themselves be affected by the treatment (often referred to as “bad controls”). We show that common approaches such as simply dropping bad controls are often ill-advised and develop two alternative approaches that allow bad controls to function as genuine controls despite being affected by treatment. First, we derive explicit conditions that rationalize conditioning only on pre-treatment values of the bad control, leading naturally to the Callaway and Sant’Anna (2021) estimator with pre-treatment values as covariates. Second, under a covariate unconfoundedness condition, we develop imputation and double/debiased machine learning estimators that recover the average treatment effect on the treated. We extend these results to staggered treatment adoption, provide pre-tests for the identifying assumptions, and apply the methods to study the effects of job displacement on earnings.

Labor Market Effects of the Venezuelan Refugee Crisis in Brazil

hsantanna@uab.edu (Hugo Sant'Anna) — Thu, 01 Jan 2026 00:00:00 -0500

We use administrative panel data on the universe of Brazilian formal workers to investigate the labor market effects of the Venezuelan crisis in Brazil, focusing on the border state of Roraima. The results using difference-in-differences show that the monthly wages of Brazilians in Roraima increased by around 2 percent, mostly driven by those working in sectors and occupations with no refugee involvement. The study finds negligible job displacement for Brazilians but finds evidence of native workers moving to occupations without immigrants. We also find that immigrants in the informal market offset the substitution effects in the formal market.

AI Exposure and Entrepreneurship

hsantanna@uab.edu (Hugo Sant'Anna) — Tue, 01 Sep 2026 00:00:00 -0400

This paper asks whether artificial intelligence is changing the prevalence of growth-oriented entrepreneurship. In a commuting-zone-by-sector panel of U.S. workers, we compare sectors with differing AI exposure within the same commuting zone and year, absorbing every shock common to a place. A one-standard-deviation increase in AI exposure lowers the incorporated self-employment share by 0.401 percentage points (10.8% of baseline) in 2019–2024, while unincorporated self-employment shows no response. The decline first emerges in 2019, four years after the deep-learning era began, and is concentrated in the managerial and professional occupations from which incorporated business owners are drawn.

The End of Free Movement and International Migration

hsantanna@uab.edu (Hugo Sant'Anna) — Mon, 01 Dec 2025 00:00:00 -0500

This paper examines how the end of free movement following Brexit in January 2021 reshaped non-transient international migration flows involving the UK. Using a novel dataset of monthly bilateral migration flows from 2019 to 2022 constructed from Facebook user location data covering 181 countries with high spatial and temporal resolution, we estimate a gravity model framework to assess post-Brexit changes in migration patterns. We find that the end of free movement significantly reduced migration between the UK and EU countries, while migration from non-EU countries to the UK increased. Migration from the UK to non-EU countries also declined. These findings illustrate that Brexit reconfigured the geography and composition of international migration flows.

Immigration Enforcement and Local Business Dynamics

hsantanna@uab.edu (Hugo Sant'Anna) — Fri, 01 May 2026 00:00:00 -0400

This paper examines the effects of interior immigration enforcement on local business dynamics. Exploiting the federally determined rollout of the Secure Communities (SC) program across U.S. counties, we use county-sector-year Census administrative data to show that enforcement reduces establishment entry and job creation, with no significant effect on exit or job destruction. We provide evidence consistent with a labor supply mechanism: SC activation reduces the immigrant population, raises wages, and generates larger effects in immigrant-intensive and high-turnover sectors. The decline in job creation is driven by continuing establishments adjusting along the intensive margin rather than exiting.

Does Employer-Based Immigration Enforcement Reduce the Undocumented Immigrant Population? Re-examining the Effects of LAWA

hsantanna@uab.edu (Hugo Sant'Anna) — Tue, 12 May 2026 00:00:00 -0400

This paper replicates Bohn, Lofstrom, and Raphael (2014), who use the synthetic control method to estimate the effects of Arizona’s Legal Arizona Workers Act (LAWA) on the state’s likely unauthorized immigrant population. Our narrow replication confirms their main finding: LAWA led to a 1.5–2 percentage point decline in the noncitizen Hispanic share. In a wide replication, we apply the augmented synthetic control method and synthetic difference-in-differences and extend the sample through 2015. The core finding is robust to these alternatives, though the effect attenuates after 2011. Migration data suggest responses along both outflow and inflow margins.

Down the River: Labor Market Effects of Brazil's Worst Environmental Disaster

hsantanna@uab.edu (Hugo Sant'Anna) — Sun, 01 Mar 2026 00:00:00 -0500

Employer concentration shapes who bears the cost of an environmental shock. We show this using Brazil’s 2015 Fundão dam collapse, a sudden disaster that contaminated 650 kilometers of the Doce River with mining waste. With matched employer-employee data and a difference-in-differences design across watershed boundaries, formal wages fell 1.8 percent while job separations, job-to-job transitions, and migration held flat. Workers absorbed the shock through lower pay, not reallocation. The wage decline falls almost entirely on workers in concentrated labor markets, scales with baseline concentration, persists outside mining, and is sharpest where contamination and employer power coincide. Estimating labor supply to the firm directly, we find an elasticity near 0.55, far below competitive levels, implying pay near a third of marginal product. The foregone wages, from US$158 million over three years to US$1.16 billion in perpetuity, appear in no settlement, including the US$31.7 billion Definitive Settlement.

Gender Differences in Comparative Advantage Matches: Evidence from Linked Employer-Employee Data

hsantanna@uab.edu (Hugo Sant'Anna) — Wed, 01 Jan 2025 00:00:00 -0500

In this paper, I introduce a novel decomposition method based on Gaussian mixtures and k-Means clustering, applied to a large Brazilian administrative dataset, to analyze the gender wage gap through the lens of worker–firm interactions shaped by comparative advantage. These interactions generate wage levels in logs that exceed the simple sum of worker and firm components, making them challenging for traditional linear models to capture effectively. I find that these “complementarity effects” account for approximately 17% of the gender wage gap. Larger firms, high human capital, STEM degrees, and managerial roles are closely related to it. For instance, among managerial occupations, the match effect goes as high as one-third of the total gap. I also find women are less likely to be employed by firms offering higher returns to both human capital and firm-specific premiums, resulting in a significantly larger firm contribution to the gender wage gap than previously estimated. Combined, these factors explain nearly half of the overall gender wage gap, suggesting the importance of understanding firm–worker matches in addressing gender-based pay disparities.