Working Papers
“Colonialism, Institutional Trust and Out-Group Bias: Evidence from Morocco (Job Market Paper)” [Link to the paper] [Link to the AEA interview]
This study shows that Morocco’s exposure to different institutional and administrative practices under French and Spanish protectorates in the early 20th century has long-run effects on out-group bias and trust in institutions. The French colonial administration pursued a discriminatory policy implemented through education and by separating legal systems between ethnicities. Using spatial regression discontinuity design, I find that people on the former French side of the border have lower trust in the president and parliament compared to those on the Spanish side where this policy was not present. They also have worse perceptions of how the government is handling economic issues. Furthermore, they show more negative attitudes toward people of other religions and ethnicities. I show that these results are not driven by wealth, education, or economic development. The findings are robust to alternative specifications, narrower bandwidths, excluding coastal observations, and shifting the border.
"Mining, Local Wealth and the Resource Curse in Sub-Saharan Africa: The Role of Pre-Colonial Political Centralization" [Link to the draft]
The causes of the local resource curse in Sub-Saharan Africa remain poorly understood. This study examines pre-colonial political centralization as a factor explaining differences in local wealth around mining areas. Using a cross-sectional difference-in-differences approach, I show that the effect of mining on local wealth in Africa differs systematically with pre-colonial political centralization, with greater wealth gains concentrated in historically centralized areas. The effect is driven by proximity to gold and diamond mines and operates through both the extensive and intensive margins. When exploring the underlying mechanisms, I find local accountability of traditional chiefs to be a likely channel driving these results. I do not find evidence that increases in manual or service employment, cash earnings, or educational attainment near mines explain the observed wealth effects. The findings are robust to alternative distance cutoffs and standard error clustering levels.
“Labor Fixity and the Shielding Effect of Occupational licensing” (with Bobby Chung and Peter Q. Blair)
Licensed workers are less likely to experience layoffs than their unlicensed counterparts during economic downturns. Using historical data on occupational licensing laws in the United States, we document robust license shielding across three major recessions since the 1980s. We interpret these patterns through the lens of labor fixity -- the extent to which regulatory frictions limit firms’ ability to adjust labor input. Consistent with this mechanism, we find that license shielding operates through reduced layoffs rather than voluntary quits. Moreover, the shielding effect is stronger with longer-standing statutes and in industries more severely affected by recessionary shocks.
"Teacher Salary, Hiring, and Retention: Evidence from a Minimum Pay Shock" (with Bobby Chung) [link to the draft]
Utilizing administrative employment records, we examine the impact of Arkansas’s 2020 minimum teacher pay policy on district-level teacher recruitment and retention. To control for spatially heterogeneous pandemic trends, we combine a difference-in-differences framework with border-pair matching along the Arkansas–Texas border. Our empirical analysis documents a robust first-stage rise in teacher salary. Nevertheless, due to the brief post-intervention period, initial impacts on teacher hiring and separation are inconclusive. Because labor market responses unfold gradually, future studies tracking long-term data will be crucial.
Work in Progress:
“Effect of Universal Background Check Laws on Criminal Activity”
There are more than 400 million guns circulating in private ownership in United States which is one of main reasons for gun violence prevention programs and laws being an integral part of public safety. Using synthetic control method, this paper studies the effect of universal background check laws that extend the requirement for background checks to private sales on different crime outcomes. I employ both classic and bias-corrected synthetic control method, as well as synthetic control model with ”staggered” adoption that allows multiple treatment units at different time periods to study the relationship. Using FBI’s Uniform Crime Report and National Incident Based Reporting System’s crime data, I find a decrease in firearm homicides and aggravated assault, and the effect is significant. I find an increase in armed robbery rates which is also significant. Other than that, in order to look at potential mechanisms, I look at FBI data on background checks denials and observe an increase in background check application rate and no change in denial rates after laws were passed.