Research
Working Papers
Less Policing, More Violence: Evidence from Police Strikes in Brazil, April 2023. [pdf soon]
The paper explores Military Police strikes in two of the most populated Brazilian states to identify the causal effect of the lack of preventive policing on crime. Using a difference-in-differences design, I compare violent and non-violent crimes in Bahia and Ceará, which experienced a 33% and 70% short-term decrease in the number of officers available for patrolling, with similar states in the Brazilian Northeast region before and during police walkouts. In this setting, deterrence had an essential role in violent behavior: the evidence regarding the police-crime elasticity for violent crimes is as strong as -2.5, and a back-of-envelope calculation reveals that police strikes could cost up to U$ 8.68 million daily. The large police-crime elasticities associated with de-policing suggest that the offenders' perceived probability of being caught has asymmetric effects on their expected utility of committing a crime. Also, motor vehicle theft and robbery analysis suggest that criminals preferred violent methods to achieve their goals when the probability of being caught committing a crime was very low. The present study sheds light on what can happen without preventive policing on the streets and, specifically, shows the implications of police strikes on public safety.
Is Crime Displacement Inevitable? Evidence from Police Crackdowns in Fortaleza, Brazil, January 2024. With José Carvalho [pdf soon]
We evaluate one of the most common policing strategies in Brazil: the allocation of blitzes. This place-based intervention has well-defined policing assignments, and 3,423 interventions were precisely recorded in Fortaleza-CE between 2012 and 2013. We leverage the high spatiotemporal data resolution to make comparisons of small intervention areas at the same period of day and day of the week while controlling for common daily trends and show that an average police crackdown causes a 35% decrease in violent crime occurrences. As somewhat expected, there are diminishing returns of public safety to hours spent by the police in a single area. Although crime increases by 6% immediately after the end of a blitz, we observe lasting deterrent effects after 2-3 days. The residual deterrence cancels the crime relocation, and the intervention does not generate significant temporal displacement. Besides, we do not find crime spatial displacement to blocks up to 1.5 km from a blitz. This type of policing tactic generates deterrence by being highly visible in a street segment for a short period (30 minutes to 8 hours) and quasi-random in space-time - this intermittent design produces uncertainty that might be crucial to minimizing temporal and spatial displacement of crime.
Work in Progress
Neighborhood Intervention, Crime, and School Achievement: Evidence from Football Fields Construction in Fortaleza, Brazil, April 2023.
In July 2014, Fortaleza started a citywide large-scale urban renewal project focused on disadvantaged neighborhoods with high shares of young adults. The program builds football fields, invests in citizenship formation through football lessons, and improves the nearby infrastructure with a playground, street lighting, and sidewalk. This paper uses a difference-in-differences design to provide the causal effects of this neighborhood intervention on violent crime and students' performance. I find that football fields cause, on average, a 2/3 decrease in homicide rates with no evidence of crime moving around the corner or clock. Estimates are driven by young males with a criminal history, which suggests that renovated areas experience fewer gang fights after the policy's implementation. Voluntary incapacitation, more state presence and eyes on the streets are relevant mechanisms that explain the decrease in violence. Also, fifth-graders who study close to a football arena improved their Math scores by 2.5%, on average. Considering the neighborhood's amenity value and public safety improvements, social benefit exceeds costs within two years. Urban policies that blend person- and place-based interventions might be cost-effective alternatives to Police Services to reduce homicides.
Other
Are There Reasons for a New Fiscal Adjustment in Brazil?, December 2014. In School of Finance Administration (ESAF) - XIX National Treasure Award. [pdf], Portuguese only.
The Social Structure of Crime and the Spatial Spillover of the Police: The Case of Fortaleza, Brazil”, July 2016. Dissertation Paper, Universidade Federal do Ceará. [pdf under request], Portuguese only.