« Back to Results

Economics of Crime

Paper Session

Saturday, Jan. 6, 2018 10:15 AM - 12:15 PM

Pennsylvania Convention Center, 104-A
Hosted By: American Economic Association
  • Chair: Amanda Agan, Rutgers University

Early Releases and Recidivism

Stephan Meier
Columbia University
Armando Nicolas Meier
Columbia University and University of Basel
Jonathan Levav
Stanford University


Does an early release increase or decrease the probability that ex-convicts go back
to prison? We exploit variation in parole decisions due to grumpy judges to examine
this question. We use data from Israeli parole requests where the court appearance of a
prisoner throughout the day provides a natural experiment. We fi rst show that judges
more often deny parole requests of prisoners appearing later. We then use this variation
in instrumental variable estimations. Early releases reduce the propensity of prisoners to
go back to prison. Our robustness checks suggest that later and earlier cases are largely
comparable and that potential selection is unlikely to explain our results.

Losing Your Cool: Psychological Mechanisms in the Temperature-crime Relationship in Mexico

Gordon C. McCord
University of California-San Diego


We investigate the role of weather fluctuations on homicides in Mexico, where climate change will alter temperature and precipitation patterns and where violent crime has been a pressing issue in recent decades. Using the most detailed panel of homicides in any developing country, spanning 15 years and 2,345 municipalities, we explore the effect of weather on violence, looking at different mechanisms (economic structure, social exclusion and lack of access to electricity) through which weather shocks might play a role in human conflict. We are the first paper to add daily level analysis to this literature, allowing us to uniquely distinguish between slower-moving income channel effects of weather as opposed to same-day effects in line with human psychology literature. We find that hotter days have an immediate same-day effect on homicides, supporting hypotheses focusing on psychological effects of heat. However, the magnitude of the same-day effect is one-fourth the magnitude of the monthly effect, and even smaller compared to a 6-month effect, suggesting that slower-moving mechanisms - such as an income effect on crime - are more consequential than the psychological effect. In the case of rainfall, the effect on homicides is concentrated in municipalities with large proportions of the labor force working in agriculture, which suggests an economic mechanism linking rainfall to violence. We find no evidence that the temperature effect on homicides is attenuated by penetration of air conditioning. Results suggest that climate change will likely increase homicides through both psychological and income-related effects on criminal behavior.

The Effect of High School Peers on Juvenile Delinquency

Songman Kang
Hanyang University


In this paper, I examine the extent to which the presence of delinquent peers influences students' delinquency risk, using large administrative data from North Carolina public high schools. Fixed effect regression estimates, which exploit the year-to-year variation in the share of delinquents in a 9th grade cohort within a school, show that having more delinquent peers in school significantly increases a 9th grader's delinquency risk. Furthermore, I find evidence that 1) negative spillovers largely come from peers with similar demographic characteristics and 2) the magnitude of spillovers is positively correlated with the extent of academic tracking in high school. These findings suggest that students who are from similar demographic background and spend more classroom time together are more likely to develop close relationships and influence each other's behavior.

Dynamics in Gun Ownership and Crime - Evidence From the Aftermath of Sandy Hook

David Schindler
Tilburg University
Christoph Koenig
University of Bristol


Gun rights activists in the United States frequently argue that the right to bear arms, as guaranteed by the Second Amendment, can help deter crime. Advocates of gun control usually respond that firearm prevalence contributes positively to violent crime rates. In this paper, we provide quasi-experimental evidence that a positive and unexpected gun demand shock led to an increase in murder rates after the mass shooting at Sandy Hook Elementary School and the resulting gun control debate in December 2012. In states where purchases were delayed due to mandatory waiting periods and bureaucratic hurdles in issuing a gun permit, firearm sales exhibited weaker increases than in states without any such delays. Using additional data on gun purchasing intentions show that this finding is hard to reconcile with standard economic theory, but is in line with findings from behavioral economics. States that saw more gun sales then experienced significantly higher murder rates in the months following the demand shock, as murders increased by almost 15% over the course of a year. Our findings also show that women were relatively more likely to become murder victims than men as a result of the shock and that the circumstances point towards domestic violence.
JEL Classifications
  • K4 - Legal Procedure, the Legal System, and Illegal Behavior