The drivers of stratospheric rates of Indigenous incarceration

                                                            I have been studying how I may compare

This prison where I live unto the world.

Richard II, Act five, Scene one.

 

The Australian Institute of Criminology has just published a research report authored by academics Don Weatherburn, Michael Doyle, Tegan Weatherall and Joanna Wang titled Towards a theory of Indigenous contact with the criminal justice system (link here). In my view, this report represents the most important policy relevant research paper published in relation to Indigenous Australia in recent memory.

 

Over one third of the nation’s prisoners at any point in time are Indigenous. The ABS reports that at any one time, there are around 42, 000 people in prison in Australia, a rate of just over 200 per hundred thousand. Of these, some 14,400 prisoners are Indigenous, a rate of 2,549 per hundred thousand (link here). These data lead inexorably to a conclusion that Indigenous incarceration rates are a national disgrace.

 

Extremely high levels of Indigenous incarceration are directly and indirectly expensive for taxpayers, and impose extraordinary personal costs on those who are imprisoned and their families and on communities. These costs include diminished and constrained life opportunities; the economic and social ramifications of imprisonment; ongoing mental and physical health impairments to prisoners and their families; and ongoing intergenerational trauma. In addition, the opportunity costs (which we might conceptualise as the lost opportunities of high levels of incarceration) on both the Indigenous and mainstream Australian communities, while largely unmeasured and intangible, are bound to be substantial. In these circumstances, addressing over-incarceration of First Nations people should be a major policy priority for governments at all levels. 

 

Of course, under the National Agreement on Closing the Gap, the parties have determined that there is a target to reduce Indigenous incarceration rates. Using a baseline of incarceration levels in 2019, Target 10 aims to reduce the rate of Aboriginal and Torres Strait Islander adults held in incarceration by at least 15% by 2031. This is a pathetically unambitious target, as (on my calculation) it merely aims to reduce the incarceration rate from 2549 per 100k down to 2167 per 100k. This extreme lack of ambition raises serious questions regarding whether the Commonwealth (and its state and territory government partners) were ever serious about addressing the underlying disadvantages arising from this target that the Closing the Gap framework claims to be focussed on.

 

Even so, according to the Productivity Commission (PC) dashboard (link here), incarceration rates nationally have worsened since 2019, albeit there has been improvement in some jurisdictions. Of greater concern, incarceration rates nationally are projected by the PC to continue to rise through to 2031. Given the lack of progress, it is unsurprising to observe that governments have singularly failed to outline a comprehensive methodological model (or hypothesis) underpinning their actions directed at decreasing incarceration rates for Indigenous people.

 

These concerning trends, and the concomitant social and economic pain that they impose on First Nation communities, provide an unassailable rationale for further policy reform and action by governments at all levels. The key question then is what should those policy changes focus on?

 

The AIC Research Paper is framed quite narrowly as a statistical exercise aimed at testing a particular hypothesis. The Abstract to the research paper describes it in the following terms:

The Australian Indigenous imprisonment rate is currently 16.7 times the non-Indigenous imprisonment rate. The leading proximate cause of this over-representation is a high rate of Indigenous arrest. In this report we develop and test a model of Indigenous arrest in which the primary drivers of risk are substance use, stress and trauma, adverse social environment, exposure to arrest, human/economic/social capital, and state/territory of residence. We test the model using data from the 2014–15 National Aboriginal and Torres Strait Islander Social Survey (emphasis added).

 

The following paragraphs are extracts from the Research Paper designed to elucidate its major findings. I strongly recommend that interested readers look at the Research Paper as there is much nuance and detail (including a detailed account of the statistical analyses undertaken) that I have passed over.

 

From the Executive Summary:

Though the point is often overlooked, most Indigenous Australians are never arrested or imprisoned. It is impossible to understand Indigenous over-representation in prison without coming to grips with the factors that differentiate those who are arrested and, in many cases, imprisoned, and the majority who are not. In this report we outline and evaluate a preliminary theory of Indigenous arrest. The explanation we give treats Indigenous arrest as the interplay of two sets of factors, one of which increases the risk of arrest and the other of which reduces that risk. The first set includes factors such as age, gender, psychological distress, membership of the stolen generation, illicit drug use, alcohol use and state and territory laws and policies. The second (protective) set includes social embeddedness, income, school completion, marital status and living conditions that reduce contact with police. We evaluate the theory using the National Aboriginal and Torres Strait Islander Survey (NATSIS), a large nationally representative survey of Indigenous Australians. The results are largely consistent with the explanation we put forward. The risk of arrest is higher for males than females, rises to a peak around 21–30 years, and declines rapidly thereafter. Illicit drug and alcohol use increases the risk of arrest, as does a high level of psychological distress, being a member of the stolen generation, living in a problem-affected neighbourhood and having a higher level of exposure to police. The risk of arrest is lower among Indigenous Australians who are married, who have never been homeless, who have people they feel they can confide in, who have completed school, and who have an income in the top four deciles.

The strongest risk factor is having used illicit drugs and alcohol over the preceding 12 months, which increases the marginal risk of arrest by 14 percentage points…The strongest protective factor is school completion, which reduces the risk of arrest by 7.9 percentage points….Measures to reduce illicit drug and alcohol use, improve school retention and improve economic outcomes for Indigenous Australians are essential if Australia is to achieve any longterm reduction in the scale of Indigenous over-representation in prison(emphasis added).

 

I omit any summary of the model description and the detailed statistical results, though I recommend readers examine Figures 1 through to 4. I have taken the following text from the concluding ‘Discussion’ section of the paper.

Discussion. We set out in this report to test a theory that linked the risk of Indigenous arrest to age and gender, illicit drug and alcohol use, stress and trauma, environmental factors associated with a person’s neighbourhood, state laws and regulations, exposure to police, and human, economic and social capital. Broadly speaking, our findings are consistent with that hypothesis…

As in past studies, illicit drug and alcohol use emerged as having the strongest relationship with the risk of arrest…More than one in 10 of those who do not use illicit drugs or alcohol had been arrested at least once in the past five years, compared with almost a third of those who use illicit drugs and alcohol. The fact that this difference in risk persists over such a long period (35 years), even after controlling for a wide range of other factors, underscores just how large a contribution illicit drug and alcohol use makes to the volume of Indigenous arrests and, therewith, to Indigenous imprisonment (emphasis added) ….

The two strongest protective factors are completing school, which was associated with a 7.9 percentage point (or 35%) reduction in the risk of arrest, and having an income in the top four deciles, which was associated with a 7.4 percentage point (or 33%) reduction in risk…Reducing the risk of arrest by improving Indigenous school completion and income clearly requires a concerted effort to address the sources of Indigenous disadvantage in the home, school and labour market (emphasis added)….

The benefits associated with living in a remote area and having a permanent home are interesting, given that these variables have received little research attention in the literature on Indigenous arrest. Both are strong protective factors. If, as we assume, they measure reduced exposure to police, they suggest that simply being visible to police increases the risk of Indigenous arrest, even if no serious offence is committed. This conclusion is consistent with evidence that police make less use of diversionary alternatives when dealing with minor offences (eg offensive language, offensive behaviour) committed by Indigenous Australians than when dealing with similar offences by non-Indigenous Australians (emphasis added).

 

The authors of the AIC Research report conclude by arguing for better data to enable more rigorous testing of new or alternative hypotheses; for policy to be based more closely upon theoretical analyses of the correlates of Indigenous arrest rates; and finally for the immediate development of policies where their research finding are consistent with rigorous studies of factors that increase the risk of involvement in crime. In relation to these factors, they state:

 Research on illicit drug and alcohol use, school completion, employment and income are just four examples [references removed]. There is no need to wait for further research before developing policies to improve outcomes on these dimensions. Progress on them is essential if we are to achieve any long-term reduction in the scale of Indigenous over-representation in prison (emphasis added).

 

The AIC research paper is valuable precisely because it provides an evidence-based policy roadmap which if implemented would begin to address rising Indigenous incarceration rates. Of course, the policy reforms required will themselves take time to be designed and implemented, tasks which are themselves complex and not without risk of failure. Even when implemented, it will take time for them to gain traction and have an impact. The consequence is that even with the right policy response not only is it highly unlikely that the 2031 target will be achieved, but it is also likely that it will take at least another decade to turn around the current trends.

In these circumstances, what is required is for governments at all levels, and most importantly, for the Commonwealth to first step up and devise and implement the required reforms, and second, to stay the course. This would require the Commonwealth to come clean with the Australian community regarding the extended time frames required and the financial costs and implicit risks in the strategies being pursued.

 

The reason I consider this to be one of the most important policy relevant research papers in recent times for the Indigenous policy domain is that it substantially strengthens the policy case not just for incarceration reform, but for much more ambitious policy action on a range of other fronts, including housing, education, income support and employment policies. In other words, the AIC Research Paper is so much more than a criminological study on Indigenous  incarceration rates. With good policy design, and adequate investment of financial and human resources, each of these policy issues have the potential in their own right to substantially improve the quality of life for hundreds of thousands of the most disadvantaged Indigenous citizens. In turn, the longer term pay-off for the broader society in improved social cohesion and reduced health costs arising from widespread existence of social determinants driving sub-optimal health outcomes will be significant.

 

Yet notwithstanding the clear benefits of addressing these policy issues, our political elites and thus our political system has demonstrated a longstanding and deep-seated aversion to acting in the public interest on these issues. The likelihood that necessary action will now be taken to address incarceration is thus remote. In a very real sense, the nation’s incarcerated, its unemployed, its unhoused and its under-educated citizens continue to inhabit a realm of sustained exclusion. There is no shortage of simplistic and instant solutions: many of us have a predisposition to blame the victim, and notwithstanding the existence of systemic constraints, there is no exemption for anyone from taking responsibility for their own life. Nevertheless, the extraordinary imbalance between government rhetoric and action, and between the preparedness of democratically elected governments to play politics rather than act in the public interest, suggests that there is something very awry in the economic and social governance of the Australian nation.

 

I previously published posts dealing with incarceration issues in 2019 (link here) and 2020 (link here and link here). On re-reading them, what strikes me most is how over the past five years and notwithstanding a change of government in Canberra, nothing has changed. The underlying dynamic of apparent policy intractability and avoidance of responsibility by governments remains ubiquitous and deeply embedded.

 

While the AIC Research Paper is of much wider policy relevance than its title suggests, it does not offer a pre-fabricated policy solution to the policy issues it identifies as crucial. It merely provides a roadmap. Each of the policy areas identified requires innovative and determined policy work at both political and bureaucratic levels to transform the roadmap into real world policy. Yet I fear that without sustained advocacy from Indigenous interests, there will be no appetite within policy circles to begin the journey so persuasively mapped out by the AIC Report authors. Even with Indigenous advocacy, there is no guarantee that policymakers will listen.

 

In other words, for all our self-confidence regarding the seemingly unquestionable merit of our existing systems of democratic governance, it remains the case there exists a substantial ‘underclass’ of excluded citizens, many of whom are Indigenous. This should prompt serious reflection by thinking Australians. Might it not be the case that the deeper drivers of continuing stratospheric rates of Indigenous incarceration can be traced to the self-imposed captivity of mainstream Australians by beliefs and mindsets that refuse to recognise the continuing existence of systemic exclusion embedded in our political systems.

 

25 May 2024