ESSAYS ON THE MICROECONOMETRICS OF LABOR MARKETS AND CRIMINAL BEHAVIOR
Open Access
- Author:
- Carvalho, Jose
- Graduate Program:
- Economics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- October 07, 2002
- Committee Members:
- Coenraad Arnout P Pinkse, Committee Member
Jaeyong Lee, Committee Member
Susumu Imai, Committee Member
Joseph Terza, Committee Member
Nezih Gunner, Committee Member
Herman J Bierens, Committee Chair/Co-Chair - Keywords:
- Recidivism
Unemployment
Program Evaluation
Competing Risks
Survival Models
Econometrics - Abstract:
- The subject of the present thesis is to apply microeconometric techniques to labor economics and criminal behavior. There are two independent chapters. The aim of chapter one is to build and estimate an econometric model of recidivism for the United States using data set of released prisoners from state prisons in 1983. The data set is unique in the sense that it contains information on recidivism for both federal and state crimes in contrast of past studies restricted only to state samples with few or no information about federal crimes. The paper extends the literature on the application of econometrics to criminal behavior by treating the process of recidivism as a competing risks model where dependence arises from unobserved heterogeneity. The effects of regressors are estimated in a model that considers the non-monotonic hazards feature of criminal recidivism data sets. An accelerated time model is compared to the traditional Cox proportional model. The former appears to perform better, and is shown to capture some features of recidivism behavior reasonably well. Chapter two evaluates the effect of a reemployment program for ex-convicts on their recidivism behavior. The econometric model has to deal with the following institutional aspects of the program. First, upon release, each ex-inmate looks for a job and this duration is represented by T_{s}. Second, at the same release-time the ex-criminal could commit a crime and the time of recidivism is represented by T_{c}. Third, the difference between controls and treatments samples only occurs conditional on finding a job. Hence, the random search duration, T_{s}, represents the timing of receiving the treatment. Finally, the available duration data is grouped. Our paper extends the literature on econometric program evaluation by building and estimating a model that considers the conditional feature of the treatment. We use a sample from the 1985 Employment Services for Ex-Offenders Program and find that the program helps reduce criminal activity.