How do we break down a $3.4b prisons bill? What can it tell us?

Longer-term objectives of prison, such as their cost as a deterrent or the cost of failures to rehabilitate, are much harder to put a price on. shutterstock

This article is part of the Beyond Prison series, which examines better ways to reduce re-offending, following the recent State of Imprisonment series.


Australia’s states are struggling to bear the costs of their growing prison population.

  • It costs at least A$80,000 to house each prisoner for a year.

  • Recurrent spending for each Victorian on prisons was $83.95 in 2011-12.

  • It costs around $180 million a year to operate South Australia’s prisons.

  • Restrictive funding of alternatives like community corrections in Queensland has implications for the success and utilisation of these programs.

  • In Western Australia, the average cost of keeping someone in prison is $342 a day, compared to $43 for community service supervision.

  • In the ACT, at $394 per day, prisoner costs are thought to be the highest in the country.

  • The Northern Territory spent $500 million on the construction of a prison with 1000 additional beds to house a prison population that is worryingly over-represented by Indigenous Australians.

  • In NSW, prisons are bursting at the seams, operating 9.4% above capacity.

These snapshots offer a “user friendly” picture of prison costs. But the recasting of total costs in these various ways is not without problems.

What do we know about the cost of corrections?

Prisons serve multiple purposes. Some of these purposes are relatively short-term and concrete in their objectives, such as the act of imprisonment itself. These are not without problems, but are relatively easy to cost.

Other objectives are longer-term and harder to define, such as the cost of prisons as a deterrent, or the costs of vocational programs over the longer term, or the cost of failures to rehabilitate and high rates of recidivism. These are much harder to cost. Despite their importance, they have tended to be ignored in the cost data that is used within the realm of fiscally constrained policymaking.

In Australia, at least theoretically, the modern prison is cast as a correctional institution. As such, the importance and cost of its rehabilitative role is particularly relevant.

“Correcting” offenders and minimising rates of re-offending will always be a challenge. But rehabilitation involves equipping inmates with a variety of skills so they can successfully re-enter the community.

Inmates with family support and educational opportunities, particularly those focused on vocational training, are more successful at securing a crime-free future. Both require significant resources.

Unfortunately, tracking the costs of many of these programs is easier than monetising their benefits.

Comparing costs

If we start with the total cost of prisons in Australia, which was $3.4 billion in 2013-14, we can then cut this figure in a number of ways to draw inferences about relative efficiencies, performance and value for money.

From this, we can track whether the total cost has gone up or down over time. We can also calculate figures such as the cost per prisoner per day and the costs allocated to each prison per year based on its budget allocation.

Much of this data appears compelling and accurate – the figures are based on accounting numbers. But many don’t realise that the cost calculations produced by accountants are quite crude approximations, which are always the result of judgement, assumptions and estimations. These can also be highly politicised.

For example, the calculation of “total costs” and how these are allocated to different prisons and projects are based on a series of decisions. These decisions can distort the relative costs of one prison or program over another. The total cost of prisons in Australia is a reflection of the amount spent on prisons and prison-related services (such as health, vocational programs, transportation).

When we reduce this to a figure per prison, as an outsider, it is hard to determine which costs are allocated to which prison and for what reason.

For instance, how are the respective departmental overheads allocated to each prison? This problem is only amplified with the inclusion of private prison providers.

The costs data should facilitate comparisons of relative performance, value for money and efficiency. But limitations on the quality of the data mean that, more often than not, they don’t.

If we are to come up with ways to facilitate equitable, efficient and effective prisons, then we need to know a lot more about how the costs of each prison are determined. This includes the effect of location, the impact of the age and design of prisons, and the experience profile of staff.

How well the data reflect the true cost of each prison or the sector should be a matter of significant concern to fiscally constrained governments, as they experiment with ways to meet the needs of a growing prison population, and to the public – who are largely convinced that “value for money” is key to public policy.

The problem with costs

Costs have clear policy relevance and should be part of public debate. But we also need to talk more openly about the cost information’s limitations in the public domain and the effect this has on the quality of debate.

While cost allocations will be determined by established practices and professional judgement, these change over time and are unavoidably political.

With this in mind, it would be unrealistic to suggest that we “perfect” the costing process. This would simply replace one set of assumptions with another.

We need to stop assuming that costs provide a neutral starting point to determine optimal policy. Instead, in complex policy settings like prisons, the best we can hope for is greater transparency to enable richer public discussion.


You can read other articles in the Beyond Prison series here.