A Few Bad Apples? Racial Bias in Policing
(with Steve Mello)
We estimate the degree to which individual police officers practice racial discrimination. Traffic police regularly discount the charged speed on drivers' tickets to avoid a discrete jump in the fine schedule. This behavior leads to an excess mass in the distribution of charged speeds just below the jump. Using a bunching estimation design and data from the Florida Highway Patrol, we show that minorities are less likely to receive this break than white drivers. We disaggregate to the individual police officer level and find significant heterogeneity across officers in their degree of discrimination, with 40% of officers explaining the entirety of the aggregate discrimination. Our measure of discrimination is easy to calculate and can be used by police departments as part of an early warning system. Using a simple personnel policy that reassigns officers across locations based on their lenience, departments can effectively reduce the aggregate disparity in treatment.
Media Coverage: The Conversable Economist, The Economist, Miami New Times, Marginal Revolution, Mother Jones, Quartz, Vox
The Effects of School Construction on Student and District Outcomes: Evidence from a State-Funded Program in Ohio
I study an ongoing state-subsidized program of rebuilding and renovating Ohio’s K-12 public schools and investigate the effect of improved facility quality on student and school district outcomes. The completion of a project increases public school enrollment and district property values. Test scores do not measurably improve upon completion and suffer significant reductions during construction. The implied willingness to pay for a project is lower than total costs but greater than the cost borne by district residents. While the program led to a narrowing in expenditures across district wealth, I find little evidence that it reduced disparities in student outcomes.
Does the Punishment Fit the Crime? Speeding Fines and Recidivism
(with Steve Mello)
We estimate the causal effect of harsher speeding punishments on future driving behavior of cited drivers. To account for the fact that punishments are not randomly assigned, we leverage variation in ticket-writing practices across highway patrol officers in Florida. The fine associated with a ticket written for 10-14 MPH over the speed limit is, on average, $75 higher than that with a ticket for 9 MPH over the limit. Over 30% of tickets are written for exactly 9 MPH above the limit, while less than 3% are written for 10 MPH over, suggesting that officers manipulate the ticketed speed, and by extension, the fine faced by the driver. Officers vary considerably in their propensity to write tickets for the lower fine amount, and we instrument the punishment faced by a ticketed driver with the stopping officer’s average lenience towards other drivers. Our estimates suggest that, compared with those receiving a higher fine, drivers receiving the lenient fine are over 25% more likely to receive an additional speeding ticket in the following year. We also find that drivers receiving the lenient fine are about 14% more likely to be involved in a car accident in the following year, although this result is more sensitive.
Works in Progress
Police Unions and Officer Misconduct
I study whether police unions affect the salaries, employment, and misconduct of officers. For this project, I digitized tallies of over 500 Florida police union elections for the years 1974 (when unionization was legalized) onwards, collected salaries for all officers enrolled in the state pension system, and acquired a record of state-level investigations into officer misconduct. Preliminary evidence suggests that, while unionization leads to no change in department size, earnings of officers increase after unionization. Contrary to popular opinion, I find that unionization decreases the incidence of police misconduct in the year after an election and has no long-run effect on misconduct.
Statistical Discrimination and MBA Grade Non-Disclosure
(with Conrad Miller and Chris Moser)
We examine the rollout of grade non-disclosure at top-tier MBA programs, whereby students can no longer reveal their grades to recruiting employers during their final year of school. We use this policy change to study how employers change their inference about an applicant’s ability when they are less able to observe their school performance. In particular, we test for the presence of statistical discrimination, where employers rely on demographic characteristics, such as race and gender, to determine an applicant’s productivity. If this type of discrimination is occurring, we expect that racial and gender differences should become exacerbated upon the implementation of grade non-disclosure. We have received approval from two business schools to use their data and are in the process of linking their records of demographics, grades, and post-graduation wages.