Final Report Summary - HC & CRIME (Human capital and criminal decision making in youths)
This research project aimed to understand the impact of education on criminal participation in youth. It considered different ways to disentangle the problem of factors which jointly influence decision-making in both these activities. The project also made use of individual data from the Netherlands and cohort level information from the United Kingdom. The work carried out during this was written up as four working papers. The methodology and findings from these research output are summarised in what follows:
'Social bonding, early school leaving, and delinquency' (with Tanja Traag and Rolf Van Der Velden)
In this paper, we investigate how successful social bonding theory is at predicting juvenile delinquency and school dropout behaviour. We adopt a simple dynamic approach which assumes that past involvement in risky behaviour reduces individual restraints for future participation in risky behaviour. We use a ten-year education panel following Dutch adolescents who participated in a survey in their first year of high school in 1999. This information was matched to annual information on police arrests based on registry data. Our results show that school performance (as measured by test scores) is the key social bond element preventing young people from engaging in risk behaviour. We also find that involvement in past risky behaviour increases the likelihood of future missteps and that the protective influence of school performance is mitigated.
'The crime reducing effect of education' (with Stephen Machin and Suncica Vujic)
In this paper, we study the crime reducing potential of education, presenting causal statistical estimates based upon a law that changed the compulsory school leaving age in England and Wales. We frame the analysis in a regression-discontinuity setting and uncover significant decreases in property crime from reductions in the proportion of people with no educational qualifications and increases in the age of leaving school that resulted from the change in the law. The findings show that improving education can yield significant social benefits and can be a key policy tool in the drive to reduce crime.
'Education and youth crime: individual measures of the causal relationship' (with Tanja Traag and Rolf van der Velden)
We measure the causal relationship between individual educational attainment and criminal participation. We consider the endogeneity problems which plague this relationship. We solve these by instrumenting obtaining a qualification by differences in timing of birth of students and debate the validity of the instrument. We use a unique database matching a large survey of Dutch youths to administrative data on arrests and educational attainment. We observe that relatively older students are more likely to leave secondary school without a valid degree. We find that secondary school qualification reduces the probability of arrest after leaving school by about a third.
'Youth crime and education expansion' (with Stephen Machin and Suncica Vujic)
We present new evidence on the causal impact of education on crime, by considering a large expansion of the United-Kingdom's post-compulsory education system that occurred in the late 1980s and early 1990s. The education expansion raised education levels across the whole education distribution and, in particular for our analysis, at the bottom end enabling us to develop an instrumental variable strategy to study the crime-education relationship. At the same time as the education expansion, youth crime fell, revealing a significant cross-cohort relationship between crime and education. The causal crime reducing effect of education is estimated to be negative and significant, and considerably bigger in (absolute) magnitude than ordinary least squares estimates. The education boost also significantly impacted other productivity related economic variables (qualification attainment and wages), demonstrating that the incapacitation effect of additional time spent in school is not the sole driver of the results.
The overall findings from this research is therefore that increasing the educational attainment of young people has a very strong causal reduction impact on their probability of committing crimes.
'Social bonding, early school leaving, and delinquency' (with Tanja Traag and Rolf Van Der Velden)
In this paper, we investigate how successful social bonding theory is at predicting juvenile delinquency and school dropout behaviour. We adopt a simple dynamic approach which assumes that past involvement in risky behaviour reduces individual restraints for future participation in risky behaviour. We use a ten-year education panel following Dutch adolescents who participated in a survey in their first year of high school in 1999. This information was matched to annual information on police arrests based on registry data. Our results show that school performance (as measured by test scores) is the key social bond element preventing young people from engaging in risk behaviour. We also find that involvement in past risky behaviour increases the likelihood of future missteps and that the protective influence of school performance is mitigated.
'The crime reducing effect of education' (with Stephen Machin and Suncica Vujic)
In this paper, we study the crime reducing potential of education, presenting causal statistical estimates based upon a law that changed the compulsory school leaving age in England and Wales. We frame the analysis in a regression-discontinuity setting and uncover significant decreases in property crime from reductions in the proportion of people with no educational qualifications and increases in the age of leaving school that resulted from the change in the law. The findings show that improving education can yield significant social benefits and can be a key policy tool in the drive to reduce crime.
'Education and youth crime: individual measures of the causal relationship' (with Tanja Traag and Rolf van der Velden)
We measure the causal relationship between individual educational attainment and criminal participation. We consider the endogeneity problems which plague this relationship. We solve these by instrumenting obtaining a qualification by differences in timing of birth of students and debate the validity of the instrument. We use a unique database matching a large survey of Dutch youths to administrative data on arrests and educational attainment. We observe that relatively older students are more likely to leave secondary school without a valid degree. We find that secondary school qualification reduces the probability of arrest after leaving school by about a third.
'Youth crime and education expansion' (with Stephen Machin and Suncica Vujic)
We present new evidence on the causal impact of education on crime, by considering a large expansion of the United-Kingdom's post-compulsory education system that occurred in the late 1980s and early 1990s. The education expansion raised education levels across the whole education distribution and, in particular for our analysis, at the bottom end enabling us to develop an instrumental variable strategy to study the crime-education relationship. At the same time as the education expansion, youth crime fell, revealing a significant cross-cohort relationship between crime and education. The causal crime reducing effect of education is estimated to be negative and significant, and considerably bigger in (absolute) magnitude than ordinary least squares estimates. The education boost also significantly impacted other productivity related economic variables (qualification attainment and wages), demonstrating that the incapacitation effect of additional time spent in school is not the sole driver of the results.
The overall findings from this research is therefore that increasing the educational attainment of young people has a very strong causal reduction impact on their probability of committing crimes.