Three Essays on the Economics of Crime

Abstract

This dissertation consists of three essays on economics of crime. The first chapter investigates the extent to which variations in temperature affect crime rates in the southeastern states within a typical empirical crime model framework, where crime rates are specified as a function of standard controls. County-level panel data for the period from 2010 to 2014 for the southeastern states are collected from the Federal Bureau of Investigation (FBI), the Centers for Disease Control and Prevention (CDC), and the Behavioral Risk Factor Surveillance System. This empirical analysis uses 2SLS because police presence is endogenous in the crime rate equations, and police presence is instrumented by the adjusted gross income level of each county. The results show that in the region, rise in average maximum and minimum daily temperature is associated with higher robbery, property, and grand larceny. Additionally, higher alcohol consumption has an impact on the increase of violence, robbery, assault, grand larceny, and motor vehicle theft. Chapter 2 examines the relationship between mental health, opioid prescribing rate and the incidence of crime in the United States. The data from 2012 to 2015 for 3142 counties are employed to estimate the impacts of the mental health on the incidence of crime. An endogeneity issue of the number of law enforcement personnel and crime is observed. To break down the endogeneity problem between crime and the number of law enforcement personnel, an instrumental variable “annual average police wage” is employed. Results from the fixed-effect approach indicate that an increase in the number of law enforcement personnel decreases the probability of the number of robberies in urban counties. Additionally, an increase in the number of mental health providers lowers the number of burglaries in the urban counties.

iii

Chapter 3 examines whether the supplemental nutrition assistance program (SNAP) implementations had an effect on criminal activities. To address this problem, the present study utilizes two main variations: (i) changes in waiver of the time limit for areas within the states, and (ii) increase in SNAP benefits. A county level panel data from 2009 to 2015 for 3134 counties is employed to investigate the relationship between SNAP benefits and crime. The findings show that SNAP benefits contribute a significant reduction in the criminal activities in both rural and urban counties. The estimation results indicate that changes in waivers to work related time limit is one of the significant factors which has impact on the criminal activities. Additionally, Income motivated crimes such as property, robbery, and burglary are more likely to be affected by the changes in individuals’ welfare and income level changes.