Consequences of violent assaults for young victims and their parents🍋

Measuring the effects of youth crime and spillovers to parents

Falling victim to crime is a frightening event, with serious negative consequences for victims’ health, labour market participation, and human capital accumulation.¹ Most existing studies on the impacts of victimisation focus on adult victims, but the group most likely to experience victimisation is youth.

This post previews findings from one of my PhD papers: The effect of violent assaults for youth victims and their parents.²

Fiordland coastline. Dave Heatley

Who falls victim to crime?

Aotearoa has some of the highest rates of victimisation among developed countries. 21.5% of New Zealanders report having recently been a victim of crime, which is much higher than the OECD average of 15.5%.³

Victimisation disproportionately happens to young people. The graph below shows the victimisation-age profile for all victims of reported crime in New Zealand from 2014-23. Reported victims are mostly young people, with the age-distribution peaking at age 24.⁴

Age distribution for all reported victims

A traumatic experience in one’s formative years can have long-term negative impacts on health, wellbeing, and economic outcomes.⁵ So, it is important to understand how victimisation affects young people, and whether these effects persist.

Estimating the causal effects of victimisation

Victimisation is not a random event, so we cannot simply compare the outcomes of victims and non-victims to understand the causal effects of victimisation.

In my research project, we overcame this challenge by focusing only on (reported) victims and using differences in the timing of assault to create comparable treatment and control groups. Our treatment group are young people victimised in the present and the comparison group are young people victimised sometime in the future.

These two groups look alike, on average, with the only difference being the timing of the assault. Therefore, the yet-to-be-victimised comparison group informs us about what would have happened to the young victims had they not been assaulted.

The sample is the full population of young victims (ages 0-30) who experienced a non-family-violence assault that was reported to the NZ Police.⁶ Of course, this only captures a small sample of victims and does not speak to the effects of assault on victims who did not go to the Police.

Using a stacked difference-in-differences model, we explored how assault victimisation impacts health, work, and schooling behaviours in the first 12 months after the assault.⁷

Health effects of assault victimisation

Young people are hurt in many ways when they are assaulted. For physical health outcomes, we found immediate, significant increases in hospitalisations and pain-relief prescriptions due to the assault.

Victims also increased their use of mental health services as well as anti-anxiety, anti-depression, and anti-psychotic medications in the aftermath of victimisation.

Flow-on effects to school and working behaviour

These health effects play a key role in shaping how victims participate in society. For primary- and secondary-school aged victims (ages 5-18), being assaulted increases the likelihood of needing a schooling intervention, which can be for learning difficulties, behavioural issues, or excess truancy. These effects are larger for victims whose offender attended the same school, suggesting that the relationship between victims and offenders influence how victims are affected by assaults.

Working-aged victims (ages 18-30) reduce the time spent at work after the assault. The blue markers in the following graph illustrate youth’s labour earnings losses caused by the assault.⁸ Over the 12 months following an assault, these losses average about 5% of pre-assault earnings, with larger effects seen for victims whose offender was a colleague.

The orange markers show that victims’ benefit and ACC earnings increase in the first few months after being assaulted, but this tails off over time. Coupling the growing labour earnings losses with decreasing benefit gains results in a net loss in total income, with no signs of recovery one year after the event.

… do the effects stop there?

It is not unreasonable to think that when a violent assault happens to a young person, their parents will be affected. This could be due to the emotional toll of seeing their child in pain, or because they need to provide extra care during the child’s recovery.

This research is the first to estimate the spillover effects of youth victimisation to parents.⁹ We found that the victimisation of a young person has an immediate impact on parents’ mental health. During the month of the assault, we see significant uptake in parents’ use of mental health services and anti-anxiety medication. Parents also take time away from work during the 12 months after their child was assaulted. This reduction in labour earnings is larger for parents whose child was more severely hurt.

What does this mean for policy in NZ?

Our youth victimisation project presents two stylised facts for NZ policymakers.

• The labour earnings losses caused by the victimisation of young people are over 400% greater than the take-up of benefit and ACC earnings within the first 12 months after victimisation.

• The cost of victimisation goes well beyond the direct effects on the victims: parents make up one-third of the total annual labour earnings losses caused by the assaults of young victims.

Together, these results tell us that New Zealand’s current social welfare system plays only a small role in compensating lost income for those affected by youth victimisation. Whether all labour earnings losses should be fully compensated, including the losses for parents of victims, is an open question.

By Livvy Mitchell

Tree fuschia, Resolution Island, Fiordland National Park. Dave Heatley

1. See, e.g., Bindler & Ketel (2022).

2. I presented this paper at the 2024 NZAE Conference. Two of my PhD supervisors, Peer Skov and Mikkel Mertz, are co-authors. We hope to publish a full working paper of this project in early 2025.

3. OECD (2009).

4. This figure shows the age distribution of all reported victims in NZ from 2014 to 2023. The y-axis shows the share of the sample, and the x-axis shows the age at the time of the victimisation. The age-victimisation curve peaks at age 24.

5. Conti et al., (2021).

6. These are physical assaults only; sexual assaults are excluded.

7. We used anonymised linked administrative data from StatsNZ’s Integrated Data Infrastructure for this analysis. StatsNZ’s disclaimer applies:

These results are not official statistics. They have been created for research purposes from the Integrated Data Infrastructure (IDI), which is carefully managed by Stats New Zealand. For more information about the IDI please visit https://www.stats.govt.nz/integrated-data/.

8. This graph shows the main estimates of the effect of assault victimisation on labour market outcomes. The sample is 68,907 working-aged people aged 18-30 whose police-reported assault victimisation occurred between 2014 and 2022. We estimate a difference-in-differences model where the treatment group are those assaulted in the present (month t) and the comparison group are those assaulted in the future (month t+13). We plot the point estimates for the effect on labour earnings (blue diamonds) and benefit earnings (orange circles), together with the 95% confidence intervals (vertical bars).

9. It is well-established that the costs of crime extend beyond the direct impact for victims. Our estimates of the spillover effects of youth victimisation to parents measures just one element of the indirect cost of crime to society. The total cost of crime includes impacts for a wider range networks and effects along different margins, including health, safety, and wellbeing. See, e.g., Cornaglia et al. (2014), Dustmann and Fasani (2016), and Johnston et al. (2018).