COVID-19, school attendance and youth crime

Summary

Background

Between 2019 and 2023, New South Wales (NSW) experienced an uptick in youth crime across several offence categories; a trend that was particularly pronounced in regional areas. 

This study examines whether changes in school attendance rates before and after the COVID‑19 pandemic were associated with changes in youth crime rates across NSW, after controlling for student and LGA characteristics. This analysis draws on linked Local Government Area (LGA)–level data on school attendance rates for Year 1–10 students and police legal actions involving individuals aged 10–16. The dataset covers 120 NSW LGAs from January 2013 to December 2023.

We separately estimated the relationship between school attendance and the rate of youth legal actions for different subsets of our data. This includes different crime types (i.e., all crime, violent crime and property crime), for observations in 2019 vs. 2023, and for LGAs located in Greater Sydney vs. Regional NSW. 

Key findings

Every LGA in NSW recorded a lower school attendance rate in 2023 compared with 2019, with an average decrease of 3p.p. (from 89% to 86%). Youth crime increased in LGAs with the largest drops in school attendance. However, the majority of LGAs (60%) with lower attendance rates experienced declines in youth crime.

Figure 1 shows our estimates of the association between decreases in school attendance and changes in rates of criminal incidents per 100,000 offenders aged 10-16 years between 2019 and 2023 by crime type, with corresponding 95% confidence intervals. Estimates are shown separately for LGAs in Greater Sydney and Regional NSW. 

In Greater Sydney, a 1p.p. decrease in school attendance was associated with increases in total, violent and property crime incident rates of 56.7, 15.2 and 2.2, respectively. In Regional NSW, a 1p.p. decrease in school attendance was associated with larger changes in rates of total, violent and property crime (by 125.0, 25.2 and 35.1, respectively). However, none of these estimates were statistically significant, indicating no association between post‑COVID changes in attendance and youth crime rates in Greater Sydney and Regional NSW. We reached the same conclusion when extending our analysis to all LGAs and expanding the observation period to 2013–2023.

Figure 1. Association between decreases in school attendance and changes in rates of youth crime between 2019-2023

Conclusion

Decreases in school attendance in the post-COVID period were not associated with significant changes in youth crime rates in NSW. This may be because declines in school attendance were relatively modest between 2019 and 2023 and/or that increases in youth crime were concentrated in a small number of LGAs rather than being widespread across NSW.