Job Loss and Crime

• In Event: Courts, Crime, and Consequences: Using Administrative Data to Estimate Causal Effects of Court Interactions

Abstract

We use administrative microdata from the U.S. Census, which links employment, mortality, and criminal records, to examine the effects of job loss on mortality and crime for both primary job losers and their family members. To identify causal effects, we use mass layoff events as exogenous unemployment shocks. We compare laid-off workers to similar workers matched on a rich set of observable characteristics, including labor market attachment, wages, and prior involvement with the criminal justice system. We also analyze effect heterogeneity, most notably by gender, age, and prior criminal history. 

For reviewers: I have obtained preliminary results. However, due to confidentiality rules of the Census, I cannot disclose them at this time.

Overall, this project seeks to answer the following questions:

1. Does job loss increase individuals’ likelihoods of committing crimes? If so, what kind of crimes (felony/misdemeanor, property/violence, etc.)?

2. What are the heterogenous effects of job loss on crimes for people with different genders, races, family compositions, ages, job-attachments, and previous criminal justice records?

3. Are the effects different between individuals losing jobs due to firm-level shocks and individuals losing jobs during industry- and economy-wide shocks, where they face different aggregate economic conditions?

4. Are there spillover effects of job loss on the on criminal behavior of partners of individuals who lose their jobs?

Author

• Duan Zhang, University of Missouri-Columbia
  Presenting Author