U.S. crime rates peaked in the early 1990s; by 2004, crime rates were more than 30 percent lower than in 1991. Most research studies the sources of these declines and little is known regarding whether these declines will be reflected in the crime patterns of the next generation. Thus, the primary goal of this project is to estimate the correlation of criminal activity across generations and decompose the mechanisms driving these correlations. This project also aims to identify how interventions, such as incarceration, mediate this relationship. This project takes advantage of a unique Swedish data set that spans more than 40 years and that contains administrative crime records for two generations. There are many mechanisms that could generate intergenerational criminal correlations. Common socioeconomic factors may lead a father and son both to have an elevated criminal propensity. Biology could also play a role. For instance, a father may pass on to his son low intellectual ability or a limited ability to regulate impulsivity that leaves both the father and son at a higher risk for antisocial behavior. Third, it is also possible that behavioral mechanisms play an important role. The father as role model is a prime example of learned behavior.

This project first uses traditional regression techniques to explain the father-child crime relationship, controlling for: (i) common background characteristics, (ii) ability, (iii) other sources of household heterogeneity, such as mental illness, and (iv) neighborhood. Such a regression, however, does not yield evidence that a father?s criminality causes that of his child. Thus, the project also provides more direct evidence of the inherited traits and father as a role model channels through three experiments. First, the investigators calculate sibling crime correlations for the sample of twins as well as for monozygotic and dizygotic twins. Second, they compare the father child relationship for adopted and non-adopted children. Third, they assess whether this relationship varies with the timing of the fathers? criminality. The primary challenge to identifying the role of interventions is unobservable heterogeneity. For instance, in studying the effect of father incarceration on child criminality, there could still be omitted characteristics that are both correlated with whether a father is incarcerated and whether the child commits a crime, even with a vast set of controls. Thus, choosing an appropriate comparison group is essential; the project will consider fathers convicted of crimes but not incarcerated as well as fathers incarcerated before the child is born.

This research fills gaps in both the crime literature and the labor literature on intergenerational mobility. First, while the crime literature has considered many determinants of crime, little is known about the role of paternal criminality and incarceration. Second, if one equates certain criminal activities with the illegitimate labor market, then previous research, which focuses on intergenerational mobility in legitimate earnings, has omitted to study a segment of the labor market that is particularly important to children of low income parents.

Broader Impact: The existence of intergenerational criminal correlations potentially has substantive policy implications. If common background factors, such as socioeconomic status, are the driving force, then policies aimed at reducing poverty also reduce crime, and do so for more than one generation. On the other hand, if inherited traits play an important role, then there may be less scope for policies aimed directly at the children of criminals, unless one believes that there are strong interaction effects between nature and nurture. However, if the observed intergenerational criminal correlations are mainly due to behavioral responses, such as criminal parents being role models for their children, then a new avenue for fighting crime appears feasible. Deterring crime today has a larger impact on total crime than previously recognized, since it also produces a second generation effect. Identification of such a behavioral mechanism would be among the most important contributions of this project.

National Science Foundation (NSF)
Division of Social and Economic Sciences (SES)
Standard Grant (Standard)
Application #
Program Officer
Nancy A. Lutz
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Maryland College Park
College Park
United States
Zip Code