The over-representation of Indigenous Australians in prison continues to be a serious problem, even a decade after the recommendations of the Royal Commission into Aboriginal Deaths in Custody were handed down. The greatest leverage for reducing Indigenous imprisonment rates appears to lie in reducing the rate at which Indigenous persons appear in court rather than in reducing the rate at which convicted offenders are sentenced to imprisonment. This would mean not only diverting Indigenous defendants away from court, but reducing the rate at which Indigenous persons are arrested, through using alternatives to arrest, reducing the rate at which they offend or re-offend and addressing inappropriate differential treatment of Indigenous persons by the criminal justice system.
A unique opportunity to analyse the processes underlying Indigenous arrest is provided by the 1994 National Aboriginal and Torres Strait Islander Survey (NATSIS) data, with its unprecedented range of socioeconomic and cultural data. This report documents the factors associated with Indigenous arrest, rather than directly analysing the nature of offence (re-offence) or differential treatment by the police.
Data and method
Criminological research points to several factors that are likely to explain the Indigenous arrest rate:
- Torres Strait Islanders, a distinct ethnic group within Indigenous Australia with relatively low arrest rates;
- the adequacy and availability of police services;
- economic predictors of arrest, such as labour force status and education, which attempt to capture the labour market options of individuals and the capacity to communicate with various actors in the criminal justice system;
- alcohol consumption;
- the extent of verbal and physical abuse in Indigenous community;
- health conditions;
- socioeconomic conditions within families—particularly the removal of children from their parents (sometimes known in the Indigenous context as the ‘stolen generation’);
- housing need (included as an alternative measure of income poverty which tends to be poorly measured for Indigenous Australians); and
- positive and negative peer group influences.
The NATSIS had valid information on these factors for 10,235 respondents aged 13 years and over (e.g., the poor quality of information available for prisoners meant that such data was not used in the main analysis). This data is used to estimate the probability of arrest for various groups of people to assess the relative importance of the factors underlying Indigenous arrest. The results are merely broadly indicative of the factors associated with arrest because of residual concerns about the direction of causality and measurement error.
The major factors underlying the high rates of Indigenous arrest include sex, labour force status, alcohol consumption, whether a person had been physically attacked or verbally threatened, various age factors, and the cluster of education variables:
- males are 13.1 percentage points more likely to be arrested than females;
- having a job (especially in non-Community Development Employment Projects (CDEP) scheme employment) appears to lower arrest rates by: reducing the time available for illegal activities, reducing immediate financial disadvantage and improving the ‘social capital’ of workers who are in a better position to engage with the mainstream non-CDEP economy. CDEP scheme participants are less likely to be arrested than Indigenous unemployed (8.1, and 13.1 percentage points);
- alcohol consumption is one of the largest single factors underlying overall Indigenous arrest rates (12.8 percentage points);
- having been physically attacked or verbally threatened increases arrest by a similar amount to the alcohol consumption (10.9 percentage points);
- the probability of arrest peaks among 18 to 24 year-olds and then declines, being lowest among Indigenous people aged 45 years and over; and
- with a few notable exceptions, education variables behave as predicted by criminological theory with the arrest rates declining as the level of schooling increases.
Other factors examined also had a significant impact on Indigenous arrest rates:
- Torres Strait Islanders are much less likely to have been arrested (7.7 percentage points);
- family environment also had a significant effect on arrest. Among these, living with non-Indigenous people and being taken from one’s natural family are significantly associated with Indigenous arrest rates (-2.5 and 5.1 percentage points, respectively);
- the housing variables increased Indigenous arrest by between 2.5 and 3.0 percentage points;
- long-term health conditions are also associated with a higher rate of arrest (2.4 percentage points);
- the adequacy and availability of police services also had a significant impact, albeit relatively small compared to that in other studies;
- not only do negative peer influences increase Indigenous arrest but positive role models may also reduce arrests; and
- while dependents exerted no influence on overall arrest, the responsibilities (and the tighter time constraints) associated with parenthood and guardianship affected certain categories of female arrest (i.e., arrests for drinking-related charges and theft).
The overall results were robust, with the basic findings not changing substantially when the analysis was conducted separately for minors (under 18 year-olds), for each sex, or after prisoners were included in the analysis. The apparently small differences between the processes underlying male and female arrests points to cross-cultural or Indigenous-specific factors being more important than gender related issues.
The top six factors underlying the various categories of arrests (drinking-related, assaults, theft and outstanding warrants) are basically the same as those identified above. However, alcohol consumption and being a victim of physical attack or verbal threat are particularly important factors underlying arrests on drinking-related and assault charges. This would seem to confirm the suspicion that there is a cycle of violence and abuse in Indigenous communities which is probably related to alcohol consumption.
Ensuring that Indigenous citizens stay out of the criminal justice system should be a priority for governments who are concerned about Indigenous wellbeing. Unfortunately, there are a limited number of policy instruments among the factors identified. For example, sex and age are not factors that will be responsive to policy intervention. However, it is important to take them into account in designing appropriate policy as attitudes and circumstances vary dramatically across demographic groups.
The policy implications are complicated by the fact that ‘feedback mechanisms’ have been identified whe