Steven Raphael, UC, Berkeley
Improving Employment Opportunities for Former Prison Inmates: Challenges and Policy
Raphael analyzes the employment prospects of former prison inmates and reviews recent programmatic evaluations of re-entry programs that aim either to improve employment among the formerly incarcerated or to reduce recidivism through treatment interventions centered on employment. He first presents an empirical portrait of the U.S. prison population and prison releases, based on nationally representative surveys of these populations. In particular, he characterizes the personal traits of state and federal prison inmates, including their level of educational attainment and age, as well as the health and mental health issues that occur with a relatively high frequency among this population. Next he turns to the demand side of this particular segment of the U.S. labor market. Using a 2003 survey of California establishments, he characterizes employers’ preferences with regards to hiring convicted felons into non-managerial, non-professional jobs, and the degree to which employers check criminal history records, as well as the incidence of legal prohibitions against hiring convicted felons.He conducts multivariate analyses of the impact of checking criminal backgrounds on the likelihood of hiring workers of difference race/gender combinations, using legal prohibition against hiring felons as an instrument for checking.
Christopher Carpenter, UC, Irvine and NBER; and Carlos Dobkin, UC, Santa Cruz and NBER
Alcohol Regulation and Crime
Carpernter and Dobkin provide a critical review of economic research on the causal relationships between alcohol use and crime. They lay out several causal pathways through which alcohol regulation and alcohol consumption may affect crime, including: direct pharmacological effects on aggression, reaction time, and motor impairment; excuse motivations; venues and social interactions; and victimization risk. They divide theirr review broadly into four main types of alcohol regulations: price/tax restrictions, age-based availability restrictions, spatial availability restrictions, and temporal availability restrictions. They conclude that there is strong evidence from the tax and age-based restrictions of causal effects of alcohol availability on crime, and they discuss the implications for policy and practice.
Steven N. Durlauf, University of Wisconsin, Madison and NBER; and Daniel Nagin, Carnegie Mellon University
The Deterrent Effect of Imprisonment
Durlauf and Nagin review and critique the state of knowledge on the deterrent effects of imprisonment and policing. The evidence for substantial deterrent effects is overwhelming; at the same time, the strength of this evidence varies greatly when one considers specific classes of policies and marginal deterrent effects. There is relatively little evidence that increases in the severity of already lengthy prison terms in the United States yield strong deterrent effects. In fact, there are reasons to believe that the experience of imprisonment may be criminogenic. By contrast, there is substantial evidence that at current levels of severity, increases in the certainty of punishment -- generated by improvements in the efficacy of the police -- have large additional deterrent effects. Thus, one implication of our evidentiary conclusion is that crime prevention would be enhanced by shifting resources from imprisonment to policing. However, even this simple conclusion may be difficult to translate into a defensible operational plan beyond it strongly recommends against any further escalation of sentence length and leaves open many questions about the way resources should be used: more police, better logistics, more non-human capital, and so on. It also leaves open the question of the mechanism by which the resources would be transferred—corrections is largely a state and federal function whereas policing is largely a local function. Durlauf and Nagin make suggestions for generalizing the economic model of crime in a number of important directions; in particular, they address psychological and sociological aspects of criminal behavior whose integration into the standard economic crime model would enhance its explanatory power while preserving its core behavioral logic.
Richard Frank, Harvard University and NBER; and Thomas G.. McGuire, Harvard University
Mental Health Treatment and Criminal Justice Outcomes
John J. Donohue, Yale University and NBER; Benjamin Ewing, Yale University; and David Peloquin, Yale Law School
Rethinking America's Illegal Drug Policy
Seth Sanders, Duke University
Crime and the Family: Lessons from Teenage Childbearing
Sanders reviews the literature that links the "wantedness" of children, the age of a mother at the birth of her first child, and the probability that a child grows up without two parents to the child’s criminal outcomes as a young adult. He discusses the economic research that largely uses what is labeled “macro-level variation” -- that is, variation in policy at the state level that shifts the propensity of having wanted children, having children as a teen, and becoming a mother raising a child alone as the result of divorce. Sanders also reviews the literature in psychology and family therapy that uses variation at the individual level, which gives clearer statistical results at the potential cost of weaker statements of causality. This literature benefits from being decidedly more theoretical, which aids interpretation and also shows promise of allowing causality to be established using clinical trial methods. A central argument of Sanders's paper is that considering literature on teenage childbearing for girls affords insights into the origins of criminal activity, which almost always applies to boys. Three lessons emerge: 1)both theoretically and empirically, it is useful to think about teenage childbearing for girls as “female crime;” 2) the best established cohort explanation for the time-series pattern in crime is the link between having a mother who was a teen at her first birth and subsequent criminality of boys in early adulthood; and 3) the macro-level evidence is unlikely to be successful at sorting out various cohort explanations for the time-series pattern of crime because, just as in the case of teenage childbearing, the variation in potential explanations occurs approximately at the same time with limited spatial variation. Randomized trials of specific interventions may help a great deal in establishing and understanding this link.
Sara Heller, University of Chicago; Brian Jacob, University of Michigan and NBER; and Jens Ludwig, University of Chicago and NBER
Transfer Programs and Crime
Individual poverty and its geographic concentration have long been considered potential "root causes" of crime. Heller, Jacob, and Ludwig consider whether policies designed to alleviate individual poverty or the geographic concentration of poverty affect criminal behavior. They consider two pathways through which additional income might reduce crime: by changing the incentives that potential criminals face or by improving children's developmental environments in a way that changes their preferences about engaging in crime. In reviewing the best available evidence, they conclude that both individual- and neighborhood-level poverty do have a causal effect on crime. However, efforts to reduce crime by de-concentrating poverty are complicated by our limited understanding of how to reduce concentrated poverty on a large scale through public policy interventions, as well as the unknown effects of de-concentrating poverty on baseline and destination neighborhoods. While increasing the income of poor families is likely to decrease their criminal behavior, this strategy's effects may depend on the specific design of the income transfer program and, more generally, may be a less cost-effective approach than directly targeting the skills and development of children.
Peter Reuter, University of Maryland
If Drug Treatment Works so Well
Lance Lochner, University of Western Ontario and NBER
Education Policy and Crime
Patrick Hill, University of Illinois, Urbana-Champaign; Brent Roberts, University of Illinois; and Jonathan Guryan, University of Chicago and NBER
Decreasing Delinquency, Criminal Behavior, and Recidivism by Intervening on Psychological Factors other than Cognitive Ability: A Review of the Intervention Literature
One question that has plagued civilized nations for centuries is how best to address delinquent youth, in order to avoid recidivism. Approaches typically have centered on methods for improving adolescents' decisionmaking and reasoning skills. Implicit in these suggestions is the belief that youth make these decisions based solely, or at least mostly, through a deliberative process. However, two trends in psychology research have cast doubt upon this assumption. First, adolescents often act based on more automatic, non-deliberate forms of processing. Second, non-cognitive factors have been shown to predict important life outcomes as well, or often better than cognitive skills. Following this second trend, Hill, Roberts, Grogger, Guryan, and Sixkiller review the literature on those efforts to reduce recidivism and crime that intervene upon non-cognitive variables, such as social skills, health, and the family environment. They suggest that these approaches, often overlooked, can provide effective reductions in crime and delinquency, even when relatively short-term in duration. Future research should examine the relative cost effectiveness of these different approaches, as well as focus on methods for changing personality dispositions. Indeed, these researchrs believe that several of the effective interventions discussed can be characterized as promoting increases in conscientiousness or agreeableness among youth. Thus, interventions directly focused on these traits should prove particularly valuable in the effort to reduce youth delinquency and recidivism.
Philip J. Cook, Duke University and NBER; and John MacDonald, University ot Pennsylvania
Mobilizing private inputs for crime prevention
Cook and McDonald point out that crime control tends to be a one-sided topic. The focus is on the offenders. Criminal justice policy is concerned primarily with public policies intended to reform, deter, or control criminals and criminal behavior. What is missing from this approach is the possibility of reducing crime by limiting criminal opportunity. The non-consensual criminal transactions of theft and violence can be viewed as representing opportunities that are shaped by the behavior of potential victims and exploited by the criminal perpetrators. Private efforts to limit such opportunities prevent much crime. Those private efforts include everything from pedestrians avoiding dark alleys at night, to households installing burglar alarms, to banks and business improvement districts hiring security guards, to domestic partners moving out to end an abusive relationship. The fact that there were fewer motor vehicles stolen last year than in 1989 (despite the doubling of the vehicle population) has much to do with the remarkable improvements in locking and tracking devices. In addition to technological change, the increase in private security employment (which exceeds the number of police officers both in level and trend) is surely relevant.
Anne Piehl, Rutgers University and NBER; and Geoffrey F. Williams, Rutgers University
Institutional Requirements for Effective Imposition of Fines
A long theoretical literature in economics addresses the heavy reliance of the criminal justice system on very expensive forms of punishment, such as prison, when cheaper alternatives – such as fines and other sanctions – are available. Becker’s (1968) well-known result that the most efficient way to achieve deterrence is with a maximal fine has been analyzed or extended in a large number of papers, but the tradeoff among types of punishment has received somewhat less attention. Piehl and Williams analyze the role of fines as a criminal sanction, modeling heterogeneity in how people respond to various sanctions and threat of sanctions. Equilibrium is attained when those who never offend are happy to stay that way; those who offend once, in a weak moment, are glad it wasn't more than that, and wish it had never happened; and those who are undeterred wish they had better control of themselves. This paper concludes with a discussion of the policy elements necessary for attaining a stable separating equilibrium and contrasts these with existing institutional structures in the United States.