Our $3b-a-year system is flying blind in supporting ex-prisoners

Prisoners are released every day, but we don’t know how many. The lack of basic data is an obstacle to effective services that would minimise their risk of re-offending. AAP/Dean Lewins/Image digitally altered

This article is one of several following up The Conversation’s series, State of Imprisonment, which provides snapshots of imprisonment trends in each state and territory.


Nobody knows how many people get out of prison in Australia each year. This fact is so striking that it bears repetition. Despite recurring investment of more than $3 billion a year in our correctional systems, we simply cannot determine how many people move through those systems each year.

It’s not that this information is difficult to find, or that it’s not publicly available. We simply don’t know.

Understanding throughput is important

For other large, state-based and publicly funded systems, such as hospitals and schools, information on throughput is readily available, and rightly so. These systems are funded by the taxpayer, for the taxpayer, and routine public reporting is critical to ensuring transparency and accountability.

Information on throughput is also critical to service planning: if we don’t know how many people use the service, or what these people look like, how can we possibly ensure that the service is appropriate in scale and character?

Yet we can only estimate how many people move through our prisons each year. We don’t have even a basic demographic description of these people. This information is important because almost everyone who goes to prison comes back out again, and effective support during the transition from prison to community is critical to preventing re-offending.

Effective transitional support can also reduce the risk of other poor outcomes that disproportionately affect ex-prisoners, such as preventable death, the spread of infectious disease and expensive, avoidable hospitalisation.

Effective support for people coming out of prison is therefore critical to public safety, public health and the public purse. But we don’t know how many people in Australia need this sort of transitional support.

What do we know?

It’s not that we don’t know anything about the people we incarcerate: for well over a decade the Australian Bureau of Statistics (ABS) has produced a quarterly publication that reports how many people were in prison on an average day, broken down by basic demographic characteristics. For example, we know that Indigenous Australians are over-represented in our prisons by an age-adjusted factor of 13. The ABS also produces an annual publication reporting on the number and characteristics of people in prison on June 30 of each year.

Notwithstanding the redundancy in these two publications, so far so good. The problem is, people being released from prison look different to those in prison, and there are a lot more of them. How can this be?

For the purposes of illustration, let’s consider our hospitals. According to the Australian Institute of Health and Welfare (AIHW), in 2012-13 there were about 86,300 hospital beds in Australia and for the 42% of patients who stayed in hospital overnight, the average length of stay was just 5.6 days. Because of this staggering throughput, there were almost 9.4 million hospital separations – when someone is discharged from hospital – in the 2012-13 financial year. This is more than 100 times the number of hospital beds, which is a pretty good proxy for the number of people in hospital on the average day.

Detailed patient discharge records enable efficient management of hospitals, so why is this not done for prisons? AAP

Now let’s consider our prisons. Across the country we have almost 34,000 people in prison on the average day and we’re spending billions building more prison beds, at a rate well in excess of population growth. Prisons are, by definition, a growth industry.

The average expected length of stay for sentenced prisoners is 1.8 years, but for those on remand – around one in four prisoners – the average length of stay is just three months. Almost two in five of those released from prison return within two years, most of them within the first year. Those with unresolved substance use and mental health problems are more likely to return to custody.

So how many “prison separations” do we have each year in Australia? How many people does this represent? We don’t know.

In 2012, the AIHW asked the states and territories to provide this information, for inclusion in a report on the health of Australia’s prisoners. On June 30 of that year there were 29,236 prisoners in Australia. Based on the information it received, the AIHW estimated that 33,751 individuals – 15% greater than the average daily number in prison – were released during the 2011-12 financial year.

Unfortunately, this estimate was hobbled by the fact that one jurisdiction was unable to provide a count of either receptions or releases, while another provided a count of separations rather than individuals.

What did our research find?

More recently, we attempted to estimate this figure by extrapolating from detailed throughput data provided by the NSW Bureau of Crime Statistics and Research. After accounting for demographic differences between the jurisdictions, we estimated that the number of people released from prison each year in Australia is around 25% greater than the daily number in prison. Applied to the most recent ABS statistics, this equates to an estimated 42,239 persons released from prison in 2013-14 – 8,448 more than the “number of prisoners” reported by the ABS.

Perhaps more importantly, we found that the characteristics of people being released from prison differ meaningfully from those in prison. Those being released were disproportionately young, Indigenous and female. For every young Indigenous woman in prison on the average day, we estimated that about 3.7 young Indigenous women were being released from prison each year.

Is this dramatic over-representation of particularly vulnerable people taken into account in the planning and funding of transitional programs for prisoners in Australia? It seems unlikely.

Fix the data to fix the system

So what needs to be done? First, an appropriate national body, probably the ABS, needs to commit to annual reporting of prison throughput in Australia. This should include at least the basic demographic characteristics of those released. It’s not rocket science, and would for the first time provide a platform for considering whether transitional programs for prisoners are appropriate in scale and character.

Then comes the hard bit: bringing evidence-based transitional programs to scale and ensuring that they are appropriate to the target population. With a rapidly increasing incarceration rate, enormous capital and recurring expenditure on the prison system, and predictably poor health, economic and offending outcomes for those released from prison, it’s about time we stopped flying blind in service planning for ex-prisoners.


Alex Avery was a co-author of the research paper published in the Australian and New Zealand Journal of Public Health, on which this article is based.


You can read the other articles in the State of Imprisonment series here.

Help combat alt-facts and fake news and donate to independent journalism. Tax deductible.