Youth crime has long been associated with residing in disadvantaged or high crime neighbourhoods. Case and Katz (1991) and Damm and Dustmann (2014) find a strong relationship between neighbourhood criminal activity and the likelihood that individual youths are arrested for a crime. Kling and Votruba (2009) find that participation in the Gautreax Assisted Housing Program was associated with large and robust declines in the mortality of young black males. Ludwig et al. (2001), Kling et al. (2005) and Ludwig and Kling (2007) examine the effects of the Moving to Opportunity Program and find an association between assigned neighbourhood and youth crime among males.
Many studies point to social multipliers or social interactions as playing a role in the neighbourhood concentration of crime. Glaeser et al. (1996) directly test for social multipliers demonstrating that the variation in crime across cities was much larger than the variation across cities in underlying city aggregates that might explain crime rates. Bayer et al. (2009) find that exposure to others in correctional institutions contributes to future crimes, presumably through the transmission of information between youth. Jacobson (2004) and Donohue and Levitt (2001) find evidence that the size of the youth cohort, or the number of potential peers, is very important for explaining crime rates, and Damm and Dustmann (2014) link neighbourhood effects on crime to the share of criminals, again the number of potential criminal peers, as opposed to the amount of criminal activity in the neighbourhood. Finally, Patacchini and Zenou (2009) examine spillovers through friendship networks concluding that conformism to peer norms is important in the commission of crimes by youth.
Other recent studies document peer effects in crime that arise within schools. In terms of direct effects, Lochner and Moretti (2004) using changes in compulsory school laws and Jacob and Lefgren (2003) using details from school calendars show that the imposition of additional time in school reduces criminal activity. Using variation in school quality due to school choice lotteries, Deming (2011) finds that admission to a student’s first or second choice school reduces the likelihood of committing a crime and so concludes that both school quality and school peers impact adult crime. Further, Billings et al. (2014) examine the effects of redistricting following the end of school desegregation based busing in a large southern school district. One of their strongest results is that school assignment has a large impact on young adult crime. Their findings are especially strong for African-American youth that were assigned to highly segregated schools.
In new work, we examine how middle and high school assignment mediates the effect of neighbourhood on crime (Billings et al. 2016). We exploit the redistricting of Charlotte-Mecklenburg School District following the end of court-ordered busing in order to compare the arrest outcomes of students who live on either side of newly drawn attendance zone boundaries. The first set of results examines the impact of neighbourhood and school peer concentration on young adult crime. Using a range of spatial definitions of neighbourhoods and peers and a large set of controls for individual and neighbourhood attributes, the main conclusions are that same school peer concentration matters for adult crime, but only for narrow definitions of neighbourhoods and peers. We find strong causal evidence of large effects of having a higher number of similar youths assigned to both the same school and same grade with a standard deviation increase in same school-age-gender-race peers generating a 25% increase in arrest rates. The estimated effects are not present when the potential peers live further than one kilometre apart and are weaker when the peers are of different race or gender.
In the most novel aspect of our paper, we examine unique data on whether youths were arrested together for the same crime in order to gain insights into the mechanisms behind the neighbourhood peer concentration effects discussed above. Starting with a sample of youths who live near the newly drawn boundaries and are ever arrested (offenders), we create a sample of pairs of offenders by matching each offender in the boundary sample with all students arrested in Mecklenburg county within three years of age. A simple plot of the probability of a pair of students committing a crime together indicates that the probability declines rapidly with residential distance between the students, with probabilities near zero for distances further than one kilometre, but only when the pair of students are assigned to the same school (see Figure 1). For pairs of students assigned to different schools, the probability of committing a crime together is always very small. Being assigned to the same school and grade results in a six-fold increase in the baseline frequency of being arrested together.
Figure 1 Probability of partnership by residential distance
Our results strongly support the idea that social relationships between students at school contribute to larger neighbourhood effects on crime. The estimated partnership effects are strongest when both offenders reside near each other and are assigned to the same grade, are both minorities and are both male. Analyses on friendship networks show that friendships within school are much more likely to occur within grades and along racial lines (Weinberg 2007, Fletche et al. 2013). Further, the four dimensions above – proximity, grade, race, and gender – are the same criteria that result in the largest neighbourhood effects for potential peers when those peers are assigned to the same school. We also find that our estimated partnership effects are larger if the individuals have resided in the neighbourhood longer and if the individuals attended the same elementary school, consistent with larger effects when the individuals have had more time together to form a social relationship.
While a growing literature has documented the important role of neighbourhood and schools in determining the risk of youth crime, our ability to separate out and understand the role of neighbourhoods and schools in crime has been limited by the fact that school and neighbourhood are usually selected together. Most neighbourhoods are shaped by attendance zones that assign children to local schools, and urban crime takes place in city neighbourhoods often committed by youth who attend the city schools.
Papers like Deming (2011) and Billings et al. (2014) would seem to imply that school attendance itself is a major determinant of risk of youth criminal activity, especially among African-American youth, because in both of those papers the neighbourhood was held constant while school attendance changed. For example, peers in school may contribute to the level of disruption in classrooms or set norms of behaviour affecting human capital accumulation and the associated return to crime (Ross 2011). However, the findings in our paper suggest that the results of Deming (2011) and Billings et al. (2014) may have in part been driven by lottery selection and school re-assignment disrupting social relationships between those students and their neighbourhood peers.
References
Bayer, P., R. Hjalmarsson, and D. Pozen (2009). "Building Criminal Capital behind Bars: Peer Effects in Juvenile Corrections". The Quarterly Journal of Economics, 124, 105–147.
Billings, S. B., D., D. J., and J. Rockoff (2014). "School Segregation, Educational Attainment, and Crime: Evidence from the End of Busing in Charlotte-Mecklenburg". The Quarterly Journal of Economics, 129(1), 435-476.
Billings, S. B., Deming, D. J., and S. L. Ross (2016). "Partners in Crime: Schools, Neighborhoods and the Formation of Criminal Networks". NBER Working Paper #21962.
Case, A. C. and L. F. Katz (1991). "The Company You Keep: The Effects of Family and Neighborhood on Disadvantaged Youths". NBER Working Paper #3705.
Damm, A. P., and C. Dustmann (2014). "Does growing up in a high crime neighborhood affect youth criminal behavior?" American Economic Review, 104(6), 1806-1832.
Deming, D. J. (2011). "Better Schools, Less Crime?" The Quarterly Journal of Economics, 126(4), 2063-2115.
Donohue, J. J. and S. D. Levitt (2001). "The Impact of Legalized Abortion on Crime". The Quarterly Journal of Economics, 116(2), 379-420.
Fletcher, J. M., S. L. Ross and Y. Zhang (2013). "The Determinants and Consequences of Friendship Composition". NBER Working Paper #19215.
Glaeser, E. L., B. Sacerdote, and J. A. Scheinkman (1996). "Crime and Social Interactions". The Quarterly Journal of Economics, 111, 507–548.
Jacob, B. A. and L. Lefgren (2003). "Are Idle Hands the Devil's Workshop? Incapacitation, Concentration, and Juvenile Crime". American Economic Review, 93(5), 1560-1577.
Kling, J. R., J. Ludwig and L. F. Katz (2005). "Neighborhood Effects on Crime for Female and Male Youth: Evidence from a Randomized Housing Voucher Experiment". The Quarterly Journal of Economics, 120(1), 87-130.
Kling, J. and M. Votruba (2009). "Effects of Neighborhood Characteristics on the Mortality of Black Male Youth: Evidence from Gautreaux". Social Science and Medicine, 68(5), 814-823.
Lochner, L. and E. Moretti (2004). "The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports". American Economic Review, 94(1), 155-189.
Ludwig, J., G. J. Duncan and P. Hirschfield, Paul (2001). "Urban Poverty and Juvenile Crime: Evidence from a Randomized Housing-Mobility Experiment". The Quarterly Journal of Economics, 116, 655-80.
Ludwig, J. and J. R. Kling (2007). "Is Crime Contagious?" Journal of Law and Economics, 50, 491–518.
Patacchini, E. and Y. Zenou (2009. Juvenile Delinquency and Conformism. Journal of Law, Economics, and Organization, 28, 1-31.
Ross, S. L. 2011. "Social interactions within cities: Neighborhood environments and peer relationships". In The Handbook of Urban Economics and Planning (Eds. N. Brooks, K. Donaghy, G. Knapp). Oxford University Press.
Weinberg, B. A. (2007). "Social Interactions with Endogenous Associations". NBER Working Paper #13038.
