Published: August 13, 2018

Download: Life Course Centre Working Paper: 2018-13


Tony Beatton, Michael P. Kidd, and Matteo Sandi.

Non-Technical Summary:

Globally, millions of juveniles commit criminal offences every year, imposing considerable costs on society and drawing billions of dollars away from other productive uses. In Australia the police proceeded against almost 55,000 youth offenders from July 2015 to June 2016 (Australian Bureau of Statistics, 2017). As in many other countries, crime rates in Australia are observed to increase steadily until around age 18–20 and then decrease later in life. This paper examines the effect of school starting age on in-school disciplinary sanctions and youth crime. A growing economics literature has started to document that starting school at a younger age can place juveniles at a greater risk of committing crime. This paper contributes to this literature by focusing on indiscipline in school to develop an understanding of how behaviour in school may relate with the propensity to commit crime outside school.

This paper uses administrative records from the Department for Education of Queensland linked at the individual level with administrative records from the Queensland Police. The empirical findings suggest that younger pupils in cohort are more likely to commit criminal offences at all ages from the age of 18 to 24. Prior to reaching age 18, younger pupils in cohort do not appear more likely to commit criminal offences but they appear more likely to receive SDAs in school. Our findings also show that, compared to their older peers, younger pupils in cohort appear more likely to enrol in the final year of secondary school, and they appear equally likely to obtain a good certificate at the end of secondary school. Thus, the greater likelihood to commit crime after secondary school by younger pupils in cohort does not appear to derive from their poorer labour market prospects at the end of secondary school.

To test whether the observed pattern reflects the crime-reducing effect of school, this study exploits the introduction of the Earning or Learning education reform in 2006. This reform increased the minimum school leaving age from 16 to 17 in Queensland. This analysis shows that younger pupils in cohort who were incapacitated in school at age 16 due to the Earning or Learning (2006) reform committed fewer crime offences but, crucially, they received more SDAs in school at age 16. This finding is consistent with a view of crime and SDAs as similar acts and substitutes, and it supports the hypothesis that school incapacitates juveniles and, thus, it decreases their possibility to engage in crime. The increase in the propensity to receive SDAs reflects the fact that, to some extent, incapacitating these juveniles in school may move crime from outside to inside the school premises.


August 13, 2018