The Use of Prisons in Australia: Analysis and Reform Directions

KEY FACTS

TO REDUCE UNNECESSARY INCARCERATION AND KEEP THE COMMUNITY SAFE, GOVERNMENTS SHOULD:

• Extend the use of alternative punishments like fines and home detention for nonviolent, low-risk offenders.
• Restore the requirement of mens rea ("guilty mind" — the mental element of a crime, either intention or recklessness) for regulatory criminal offences.  Where strict liability is imposed, alternative punishments to prison should be applied, provided the offender has not demonstrated a propensity for violence or anti-social behaviour.
• Not punish victimless crimes with incarceration.
• Allow offenders to make restitution to their victims and take this into account in sentencing.
• Broaden the applicability of fines and restitution by enabling alternative collection mechanisms, such as garnishing wages and reducing government benefits.
• Investigate amending bail and parole laws to impose certain but mild consequences on all breaches.
• Redirect resources saved from incarcerating fewer criminals to the police, strengthening their capacity to deter and detain criminals.

INCARCERATION IS GROWING

• In 1975, there were 8,900 people in Australian prisons.  In 2015, there were 36,000.
• The incarceration rate is now 196 per 100,000 of the adult population — the highest rate since just after federation.
• With the exception of the United States, Australia has a higher incarceration rate than other major common law countries (United Kingdom, Canada) and a much higher incarceration rate than continental European countries like France and Germany.
• The number of offenders and the offender rate has been fairly flat over recent years.
• However, the juvenile offender rate has decreased sharply and adult offender rate (which is what is relevant when analysing the incarceration rate) has increased sharply in line with the growth in the incarceration rate.
• About 46 percent of prisoners nationwide are incarcerated for nonviolent offences.
• A large part of the rise in the prison population is constituted by unsentenced prisoners.
• There is some evidence that parole has become more difficult to obtain and hold, which may affect the incarceration rate.

INCARCERATION HAS SUBSTANTIAL FISCAL AND SOCIAL COSTS

• Australian governments spend $3.8 billion on prisons each year.  Net criminal justice system spending is $15.2 billion per year, with 24 percent of this spent on prisons.
• The average annual cost per prisoner is $110,000.
• For nonviolent offenders, the annual cost of incarceration is up to $1.8 billion nationwide.
• Incarceration is associated with lower lifetime economic performance, with good reason to expect that this relationship is partly causative.
• Incarceration has an overall negative impact on the children of prisoners, leading to worse lifetime outcomes for those children and additional costs to society.

---

FOREWORD

My new criminal justice project aims to develop and promote ideas for the reform of Australia's criminal justice system, including policing, the courts, and the criminal law.  This report is the first in a series of research reports that will outline a reform agenda based on sound philosophy, the latest research, and the experiences of comparable jurisdictions.

The criminal justice project will extend my sceptical scrutiny of government overreach and waste to an area from which it has been noticeably absent.  Criminal justice is one of the most important functions of government but this does not mean that it should be exempt from public oversight.  On the contrary, because criminal justice is so fundamental to public safety and order, it is vital that the criminal justice system be held to account — that individual rights are respected and spending is subjected to cost-benefit analysis.  I believe in limited, rules-based government, without exception.

A corollary of the defence of individual freedom is a belief in personal responsibility.  The criminal justice system protects the community from those individuals who do not respect the rights of others.  It is essential to civilised, ordered liberty that lawbreakers are forced to take responsibility for their actions.  Criminal justice reform can never forget that crime is a choice and that the interests of victims must prevail.

Put another way, criminals are not victims;  criminals are criminals and they should be punished for their crimes.  The real victims are the members of the public who have been robbed, assaulted, deprived of liberty or belongings or who live in fear for themselves and their families.

The question of criminal justice reform, then, is not about how best to explain crime or about how to engineer a crime-free society.  It is about how best to defend the safety of the community and the rights of the individual.

This report applies this perspective to Australia's increasing use of incarceration, which is at an historical high point.  A high level of incarceration is not a problem per se — if there are more criminals who deserve prison then a high incarceration rate is a positive.  But it is a problem if there are people who are being incarcerated unnecessarily, where the interests of justice could be better served by the imposition of an alternative punishment.

It is often said that "the punishment must fit the crime".  The unique function of prison is the isolation of dangerous criminals from the public.  But in Australia there are a great number of nonviolent, low-risk offenders in prison, people for whom prison may not be the punishment that fits best.  And because prisons are not effective in rehabilitating criminals, incarceration may not always be the punishment that does the most to protect the community from future offending.

As this report details, incarceration is an expensive punishment, and applying it injudiciously carries high costs for the taxpayer.  There are also opportunity costs:  money spent on prison is money not spent on other parts of the criminal justice system that might achieve better results for community safety.  For example, on average criminals are more likely to respond to immediate incentives than are others in the population.  This means deterrence is better achieved through increasing the chances of being caught rather than increasing potential prison sentences.  And this in turn implies that money saved by reducing incarceration could be profitably invested in policing.

In the United States, solidly conservative states like Texas and Georgia have achieved great results by being "smart on crime":  looking for efficiencies in the criminal justice system and focusing on reducing crime and reoffending.  In Australia, I propose that this means being guided by community safety and proportionality in sentencing — putting violent offenders behind bars but looking for reasonable alternatives for the nonviolent.

This report makes a strong case for the need for rethinking incarceration as part of criminal justice reform in Australia.  It provides a thorough overview of incarceration in Australia and lays out a powerful understanding of the role that individual choice and personal responsibility should play in the much-needed national conversation about criminal justice reform.  I commend it to you, and trust that you will find it illuminating.

---

INTRODUCTION

One of the most important roles for government is the protection of people's lives, liberty and possessions through a well-functioning criminal justice system.  Keeping people safe from violence gives them confidence to live, work and raise a family.

And a key part of protecting the community is incarceration.  Dangerous and antisocial criminals simply must be kept in isolation so that they cannot continue to harm others.  This is the unique and defining function of prisons.

This is not to say, however, that public safety can only be secured through incarceration, or that it is better secured as incarceration rises.  In some cases, where the offender is nonviolent and of little risk to the community, an alternative punishment may better serve the interests of justice.

Approximately 46 percent of the prison population are incarcerated for nonviolent offences.  This may have been manageable in 1975 when there were only 8,900 people in jail.  But now that number is over 36,000 — an increase of more than 300 percent.  Over this same period the total population grew by just 70 percent, resulting in the incarceration rate increasing to 196 per 100,000 adult population.  This is higher than most other common law countries and the democracies of continental Europe (though much lower than the exceptional case of the United States).

For many of these nonviolent offenders, home detention, fines, restitution orders, and other such punishments might be preferable, either because they reduce the risk of recidivism or escalation of criminal behaviour or because they better realise the interests of victims.  In these circumstances, changing the punishment mix can improve community safety.

Alternatives to prison also have the advantage of being less burdensome for the taxpayer.

The costs of criminal justice in Australia are rising sharply.  In 2014-15 alone governments spent over $15 billion on criminal justice.  The growth in prison numbers has seen an attendant explosion in prison costs.  Australia spends nearly $4 billion each year on the construction and operation of prisons.  This equates to $300 per prisoner per day, or $110,000 per year.  This adds up to approximately $1.8 billion annually to incarcerate nonviolent offenders.  It is vital that criminal justice spending is subject to the same scrutiny as all other major government programs.  This means investigating and implementing more cost-effective approaches to criminal justice — and this implies a reconsideration of the role of prisons.

Unnecessary incarceration can also have downstream effects that lessen public safety and increase waste.  Prisons have a poor record for rehabilitating criminals.  Nationwide, 59 percent of prisoners have been previously incarcerated.  Incarceration is associated with unemployment and worse lifetime economic outcomes.  Imprisoning nonviolent, low-risk offenders can inadvertently turn them into hardened criminals who may never return to productive society.  Criminal acts need to be punished.  But where appropriate we should look to alternatives to prison that might better incentivise criminals to choose the right path in the future.

This paper presents the case for reform to Australia's incarceration policies by describing the operation of criminal justice in Australia;  investigating who is in the system;  examining why those people are in the system in growing numbers;  and suggesting directions toward an improved system.

A central finding of the paper is that prisons are being used for purposes broader than what is necessary.  The objectives of criminal justice include punishment, deterrence, public safety through incapacitation and victim restitution.  Only one of these objectives — public safety through incapacitation — can uniquely be achieved by prison.  The other objectives of criminal justice can be met through alternative measures.

To this end the paper argues that prisons should be returned to their core and unique purpose of incapacitating violent offenders who pose a threat to public safety and antisocial recidivists who have failed to respond to alternative punishments.

In Chapter 1, I outline the costs of Australia's criminal justice system.  This section includes an overview of the system and specific data relating to prisons.

In Chapter 2, I describe the increase in the incarceration rate across Australia nationally and in each state, the most serious offences committed by prisoners, and the proportion of the population imprisoned for nonviolent offences.  I also describe the demographics of the prison population based on known correlates of criminality.

Chapter 3 is an examination of the possible explanations for the rising incarceration rate.  This chapter begins with an outline of how criminals think.  This model of criminal behaviour provides context for the analysis that follows.  The chapter makes several findings in relation to the underlying crime rate, the number of people on bail and remand, changes in sentencing practice, the granting and revoking of parole, and the rate at which convicted criminals reoffend.

Based on these findings, Chapter 4 outlines future reform directions based on the moral need to have punishments fit the crimes to which they are applied.  Taking the principle of proportionality in sentencing seriously has a number of implications, including that we should focus prison resources on individuals who have demonstrated violent behaviour, and that deterrence is only a subsidiary concern in sentencing and can be better achieved by investing in the police.

---

1. THE COSTS OF CRIMINAL JUSTICE

Criminal justice constitutes a large and growing part of Australian government budgets.  The costs of criminal justice comprise government expenditures on police, courts and prisons, and indirect costs resulting from criminal punishment.  This chapter details these costs and illustrates why it is appropriate to place them under greater scrutiny.

FINANCIAL COSTS

Cost of crime to victims and the community

The most obvious financial cost of crime is the damage suffered by victims of crime.  In a 2011 analysis, the Australian Institute of Criminology (AIC) estimated the cost of crime to the economy, based on a calculation of actual losses, intangible losses, loss of output caused by the crimes, and related costs like medical expenses. ⁽¹⁾

This figure was more than $23 billion.  This figure rose over the decade 2001-2011 by more than 21 percent.

Table 1 The direct costs of crime ⁽²⁾

Note that this figure does not include the costs of administering the criminal justice system.  These costs are outlined in the following section.

Note also that these figures are for those crimes for which data was available.  Later sections of this report will use the Australian Bureau of Statistics crime categories, which do not map exactly on to these categories.

Finally, while it is important to attempt to put a dollar figure on the cost of crime to victims and their families, it is very difficult to quantify the harm they suffer.  Certainly this should not be read as an attempt to reduce human suffering to financial terms.  Nonetheless, for criminal justice reform, the savings that can be made by reducing crime are relevant to considerations of alternative policy approaches.

The direct cost of crime was approximately 48 percent of the total cost of crime.  Along with the criminal justice system, other costs include private spending to crime prevention measures, such as security systems.  It also includes assistance provided to victims of crime.

Table 2 Other costs of crime ⁽³⁾

All up, the AIC estimated that crime cost the Australian economy $47 billion, a 50 percent increase over the preceding decade.

Overall cost of the criminal justice system

In 2014-15 the state and commonwealth governments spent $15.2b on the criminal justice system, comprising police, civil and criminal courts, and corrective services. ⁽⁴⁾  The vast majority of spending is on police and corrective services, accounting for 67 percent and 24 percent of all criminal justice spending respectively. ⁽⁵⁾  By comparison, spending on the Australian Federal Police was about $1.1 billion in 2013-14 and spending on the High Court was about $18m in 2014. ⁽⁶⁾

Real expenditure has grown moderately in recent years at about 2.3 percent per year on average since 2010-11. ⁽⁷⁾  The fastest-growing expenditure items have been corrective services, which includes prisons, up by 4.3 percent on average per year, and police services, up by 1.9 percent per year. ⁽⁸⁾

Figure 1 Criminal justice expenditure 2014 - 2015

Source:  Productivity Commission Annual Report on Government Services 2016

Table 3 Real expenditure on the criminal justice system ⁽⁹⁾

The cost of prisons

Australian governments spent about $3.8 billion on prisons in 2014-15. ⁽¹⁰⁾  This equates to about $300 to incarcerate one prisoner for one day on average or about $110,000 per year. ⁽¹¹⁾  There are substantial cost differences across the jurisdictions — the Australian Capital Territory (ACT) has the highest operating costs at $421 per prisoner per day ($150,000 per year), over 75 percent higher than the lowest cost state, New South Wales (NSW) which houses its prisoners for $237 per day ($85,000 per year). ⁽¹²⁾

Nationwide the cost of incarceration per prisoner has grown by close to 20 percent from 2010-11 to 2014-15 in real terms, an average annual rate of about 4.5 percent. ⁽¹³⁾  The decrease in costs in NSW of 8 percent have been more than offset by increases in the other jurisdictions. ⁽¹⁴⁾

Figure 2 The cost of prisons

Source:  Productivity Commission Annual Report on Government Services 2016

Table 4 Prison operating costs (2014-15)

The costs of prisons can be broken down into capital (fixed) and operating (ongoing) costs.  On average across Australia 76 percent of total costs are ongoing. ⁽¹⁵⁾  This includes expenditure on items such as staffing and maintenance.  The marginal costs differ greatly across the country, ranging from $363 per prisoner per day in Tasmania to $180 in NSW. ⁽¹⁶⁾

Figure 3 Proportion of prison cost by type per prisoner

Source:  Productivity Commission Annual Report on Government Services 2016

It is worth noting that one extra prisoner will not increase costs to government by $110,000.  Adding more prisoners will, to a point, reduce the average costs of incarceration because the fixed costs (such as a prison) would be averaged across a larger number of prisoners.  It is the marginal costs, on average, that will increase:  $224 prisoner per day, or about $81,000 per year. ⁽¹⁷⁾

Nonetheless, as prison numbers increase more prisons need to be built or existing prisons expanded.  Utilisation is defined by the Productivity Commission as the annual daily average prisoner population as a percentage of the number of single occupancy cells and designated beds in shared occupancy cells provided for in the design capacity of the prisons. ⁽¹⁸⁾  Currently prison utilisation rates are estimated to be quite high, from about 75 percent in Tasmania to 120 percent in the ACT. ⁽¹⁹⁾  Being over capacity by the Productivity Commission's definition does not necessarily imply prisons are not reasonable, efficient, cost effective and humane.  Still, it raises the issue of further construction of prisons or expansion of existing prisons to house more inmates.  If the incarceration trend continues this seems like a likely outcome.

The creation of new prisons and expansion of existing prisons can be expensive.  The Melbourne Remand Centre and Marngoneet prison at Barwon involved constructing 900 beds including 600 maximum security at a net present cost of $275 million.  And the project to add 350 extra beds to the medium-security Hopkins Correctional Centre in Ararat is estimated to have cost $394 million in today's dollars (about $1.1 billion over 25 years).  Similarly, the Ravenhall prison located in Melbourne's outer-west is expected to cost $2.5 billion over 25 years to build and run, and will hold 1,300 prisoners.  This cost includes about $670 million in capital costs and $1.6 billion in total operating costs.  In New South Wales a 500-bed South Coast Correctional Centre which was opened in 2010 came at a cost of $155 million, and the state government committed in its 2016-17 budget to spend $3.8 billion over four years on 7000 new prison places (which will likely increase the state's annual per prisoner cost).

Key facts 1 Australian governments spend $3.8 billion on prisons each year. The average annual cost per prisoner is $110,000. This makes up 24 percent of total criminal justice expenditure, which is $15.2 billion per year.

Figure 4 Prison utilisation rates

Source:  Productivity Commission Annual Report on

INDIRECT COSTS

Aside from the large financial costs of incarceration, there are also a range of human costs.  These costs include the economic and social consequences of incarceration, including forgone employment, loss of skills and attendant loss of economic output, and impacts on families of the incarcerated.

Worse economic outcomes for prisoners

A large body of research suggests that people who have been incarcerated perform poorly in the labour market after release, experience high unemployment rates and, consequently, lower expected lifetime earnings and stunted social mobility.

A 2014 study in the United States by the National Research Council provides an extensive overview of the effects incarceration has on employment prospects of former prisoners.  They find that many studies of ex-prisoner populations estimate roughly half remain jobless up to a year after their release. ⁽²⁰⁾  For example, a 2010 study of US prisoners undertaken by Pew Charitable Trusts estimates that incarceration reduced hourly wages of men by approximately 11 percent, annual employment by 9 weeks and annual earnings by 40 percent.  And by age 48, the typical former inmate will have earned $179,000 less than if he had never been incarcerated. ⁽²¹⁾

The study further finds that of the former inmates who were in the lowest fifth of the male earnings distribution in 1986, two-thirds remained on the bottom rung in 2006, twice the number of those who were not incarcerated.  And only 2 percent of previously incarcerated men who started in the bottom fifth of the earnings distribution made it to the top fifth 20 years later, compared to 15 percent of men who started at the bottom but were never incarcerated. ⁽²²⁾

However, it is not entirely clear that these poor labour outcomes are caused by incarceration.  Those who are incarcerated have characteristics that are associated with an elevated likelihood of both poor employment outcomes and incarceration.  This includes, for example, low levels of schooling, higher incidence of drug and excessive alcohol use, mental illness, high discount rates and patchy work histories prior to incarceration.  It is possible that these people would have had just as poor employment outcomes independent of the time in jail.  Unemployment and low wages among the formerly incarcerated may therefore result not from incarceration but from preexisting low employability and productivity.  In this sense, time in prison would send a negative signal of productivity and workplace suitability just as education is often a signal of the opposite.

However, there a number of reasons why time in jail could be detrimental to job prospects upon release.  This includes the potential stigma of a criminal record, loss of work-relevant skills, and the development of behaviours that may be useful while in prison but not well adapted to the workplace.  Regardless of the causes, difficulty finding employment is highly problematic because it increases the pressure to gain income from illegal means which of course increases the chances of reoffending and returning to prison, which, in turn, further exacerbates the problem.

Observation 1 Evidence shows that incarceration is associated with lower lifetime economic performance, with good reason to expect that this relationship is partly causative.

Effects on the children of prisoners

A range of studies have found that children with incarcerated parents are at a heightened risk of developing behavioural problems, poor education, unemployment and imprisonment.  There are a range of possible reasons for this.  For example, insofar as children take on traits from their parents, or the environment in which they are raised, one could expect that they would possess some of the traits that led their parents to criminal behaviour.  If this were the case it would imply that criminal (and other antisocial) behaviour wouldn't be "caused" by having parents incarcerated, but independently though traits inherited or received.

Alternatively, there are some who argue that the actual or perceived stigma of parental incarceration may be a source of child problems.  This would occur where, for example, the stigma of incarceration that is attached to the person who is incarcerated is passed to the child of the incarcerated.  Some contend that this could result in social and economic exclusion, in the same way as that can be observed amongst the incarcerated.  More directly, though, the loss of a parent through incarceration would involve loss of income and associated education opportunities which would have substantial consequences for their children.

The incarceration of a parent could be beneficial for the spouse and children where the incarcerated parent is abusing their family and/or abusing drugs and alcohol.  In this case incarceration could lead to a more stable and less hostile environment and even potentially reinforce law-abiding behaviour with the children by demonstrating a link between incarceration and criminal behaviour.

The preponderance of evidence, however, suggests incarceration of the parent has a substantially negative effect on their children.  For example, one study found that 23 percent of children with a father who has served time in a jail have been expelled or suspended from school, compared with just 4 percent of children whose fathers have not been incarcerated. ⁽²³⁾  These results could be partly causative.  As Rucker C. Johnson notes:  "these results suggest that parental incarceration exposure leads children to develop greater behavioral problem trajectories … Imprisoning parents may cause greater deviant behavior and crime in the next generation, and thereby contribute to the intergenerational transmission of criminal involvement." ⁽²⁴⁾

In a similar study also from the United States, Dr Holly Foster and Professor John Hagan found that independent of crime and incarceration "selection" factors, such as low self-control, paternal imprisonment decreases the educational attainment of children in emerging adulthood. ⁽²⁵⁾  These results suggest that parental incarceration itself is a causal factor for the incarceration of their children.

Further, according to a 2014 meta-analysis, The Growth of Incarceration in the United States, many studies document negative outcomes for children through open-ended interviews with fathers and their families.  According to the book, mothers and fathers:

• believe their children perform more poorly or have more difficulties in school following their father's incarceration;
• report negative behavioral changes in their children, including becoming more private or withdrawn, not listening to adults, becoming irritable, or showing signs of behavioral regression;
• report changes in children's emotional or mental health, with children experiencing such feelings as shame or embarrassment about their father's incarceration, emotional strain, including a belief that the father did not want to live at home, a loss of trust in the father, grief or depression, and guilt. ⁽²⁶⁾

Overall, the costs of incarceration go well beyond the direct financial costs to taxpayers.  Potential forgone economic opportunities, earnings and attendant loss of economic output along with flow-on effects of incarceration on to prisoners' families can be substantial.

Observation 2 Incarceration has an overall negative impact on the children of prisoners, leading to worse lifetime outcomes for those children and additional costs to society.

---

2. THE INCARCERATION RATE AND PRISONER DEMOGRAPHICS

One of the main drivers of increased criminal justice costs across Australia is a rise in the number of people incarcerated.  In this chapter, I outline the increase in incarceration and examine the characteristics of the prison population.  The following chapter will then consider possible reasons for the increase.

THE GROWING PRISON POPULATION

Over the past several decades the number of people incarcerated in Australia has increased rapidly.  In 1975 there were 8,900 people in jail nation-wide. ⁽²⁷⁾  By 2015 that number had risen to over 36,000 — an increase of more than 300 percent. ⁽²⁸⁾  Over this same period the total population grew by just 70 percent, ⁽²⁹⁾ resulting in an attendant rise in the incarceration rate to 196 per 100,000 adult population, or 151 per 100,000 of the total population.  This is 145 percent higher than in 1975 and the highest rate since just after federation. ⁽³⁰⁾

Figure 5 Incarceration per 100,000 total population

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

This puts Australia around the middle of the pack compared to other Western democratic countries — above the UK and Canada, for example, but below the US and New Zealand.  Of course, a lower incarceration rate is not good in and of itself.  One would need to analyse the costs of incarceration along with comparative crime rates, which are a function of the incarceration rate, to form a view on this.

Table 5 Incarceration rate per 100,000 population ⁽³¹⁾

INCARCERATION BY STATE

There is a substantial difference in the incarceration rate across the states.  The Northern Territory has by far the highest rate at 885, while Tasmania has the lowest rate at 130. ⁽³²⁾  The main demographic reason for this is the relatively large proportion of Indigenous Australians in the NT, who are substantially overrepresented in prison (see below).  Western Australia's incarceration rate is elevated compared to the national average, while the Australian Capital Territory and Victoria join Tasmania with rates below the national average.  Queensland, New South Wales and South Australia each have rates which are close to the national average. ⁽³³⁾

There is also a substantial difference in changes to the incarceration rate across the states over the past decade.  For instance, SA and the NT saw the largest increases, of 65 percent and 56 percent, respectively.  While Tasmania's rate dropped by close to 15 percent.  Nationally, the incarceration rate increased by close to 20 percent. ⁽³⁴⁾

Figure 6 Incarceration rates by state per 100,000 adults

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

INCARCERATION BY TYPE OF CRIME

People are incarcerated for a wide rage of crimes, from homicide and assault through to drug use and traffic offences.  In 2015 the majority of people were in prison for relatively serious offences, such as acts intended to cause injury, sexual assault and unlawful entry with intent. ⁽³⁵⁾  A sizeable minority — about 13 percent — were incarcerated in relation to illicit drug offences, 70 percent of whom were charged with possessing and/or using drugs, as opposed to dealing, trafficking, manufacturing or importing or exporting drugs. ⁽³⁶⁾

One should be aware of the limitations in interpreting prisoner data by offence type.  This information is provided by most serious offence type.  This means the actual number of incarcerated who have been convicted of a given crime is not publicly reported.  For example, many offenders who have been incarcerated for assault might also be guilty of another offence, such as breaking and entering or drug use.  But the only statistic recorded is their most serious offence.  Thus this data provides an incomplete picture of the crimes committed by those in prison.

There has been a substantial change in the composition of the prison population by most serious offence since 2005.  For example, the proportion of prisoners whose most serious offence was in relation to Dangerous or Negligent Acts Endangering a Person increased by 140 percent, while Traffic and Regulatory Offences decreased by 40 percent. ⁽³⁷⁾

Figure 7 Prisoners by crime type

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

Figure 8 Drug offence by type

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

VIOLENT, NONVIOLENT AND PROPERTY OFFENCES

In 2014-15, 53 percent of sentenced offenders were in jail for crimes against another person, including homicide, acts intended to cause injury and sexual assault.  This has increased from about 50 percent in 2005. ⁽³⁸⁾

In categorising the prison population by crime type I follow definitions provided by the ABS. ⁽³⁹⁾  Violent offences committed against a person are defined as acts that:

• relate to culpable (i.e. intentional, negligent or reckless) acts that result in harm (i.e. physical injury/violation, or non-physical harm).  These acts are not necessarily completed;  they include attempts and conspiracies.
• must affect a specific person as opposed to the general public.  That is, the victim(s) can only ever be a distinct person or persons.  Thus, the acts cannot be committed against organisations, the state or the community.

Figure 9 Change in proportion of prisoners by most serious offence 2005-14

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

Violent offences are categorised as homicide and related offences;  acts intended to cause injury;  sexual assault and related offences;  dangerous or negligent acts endangering persons;  abduction, harassment and other offences against the person;  and robbery, extortion and related offences

Nonviolent offences are generally classified as offences against organisations, government (local, state or federal) and the community in general, rather than against particular individual persons, such as illicit drug offences;  prohibited and regulated weapons and explosives offences;  public order offences;  traffic and vehicle regulatory offences;  offences against government procedures, government security and government operations;  and miscellaneous offences.

Property offences are also included for our purposes in the nonviolent category as they do not generally involve an offence against another person.  They include unlawful entry with intent/burglary, break and enter;  theft and related offences;  fraud, deception and related offences;  property damage and environmental pollution.

There are two important caveats to this observation.

One is that these categories are not always cleanly separated into violent and nonviolent offenses.  For example public order offences include both riot and affray together with nonviolent offences such as censorship crimes;  and robbery, extortion and related offences which includes armed robbery (a violent offence) together with demands made via a letter (nonviolent).  So there may be some people counted in the nonviolent category whose crimes involved some violence.

The other is that a nonviolent offence, like a weapon, drug, or traffic offence may still pose a danger to other people.  While these offences do not involve harm, or the threat of harm, to another individual they may put others at risk, whether intentionally or unintentionally.  Unlike with a violent offender, however, the risk to others entailed by these offences is not necessarily inherent in the offender, who is therefore less likely to need to be isolated from the community.

Based on these definitions, in 2015 about 54 percent of those in prison had committed a violent crime against a person;  18 percent had committed a property crime;  and 28 percent a nonviolent crime. ⁽⁴⁰⁾

Table 6 Prison population, 2015, by most serious offence ⁽⁴¹⁾

CALCULATING THE COST OF INCARCERATING NONVIOLENT OFFENDERS

We can combine data on the incarceration of nonviolent offenders outlined above with our earlier data about the cost of incarceration to produce a rough figure for the amount spent annually nationwide and in each state on locking up nonviolent offenders.

Table 7 Annual cost of incarceration of nonviolent offenders ⁽⁴²⁾

By this measure, in 2015, Australian government spent more than $1.8 billion on the incarceration of offenders whose worst offence was a nonviolent offence.

This figure does come with some significant caveats, however.

Figure 10 Sentenced prisoners by type of offence

Source:  my calculations using Australian Bureau of Statistics Prisoners in Australia 2015

As previously stated, the number of nonviolent offenders is an estimate based on the most serious offence for which a prisoner is currently incarcerated.  Some of these people may have committed violent offences in the past or concurrent with a more-heavily punished nonviolent offence.  And given that 59 percent of prisoners have been incarcerated before, many of these nonviolent offenders will be recidivists, for whom alternatives to prison may be inappropriate and/or ineffective.  (The incarceration of recidivists is discussed in greater detail in Chapter 4.)

A more accurate calculation would require a case-by-case analysis.  Should public policy be based on the distinction between violent and nonviolent offenders, it will be necessary for the courts to closely enquire into the nature and circumstances of the offending before them.

This figure includes the capital costs of building and maintaining prisons.  These costs would not necessarily go down were incarceration reduced, for the reasons discussed above.  Moreover, it has been the experience of jurisdictions that have reduced their incarceration of nonviolent offenders that some savings are offset by the need to employ more parole officers, community corrections officers, and re-entry specialists, and that prison staff numbers do not go down in strict correspondence to the reduction of the prison population because violent prisoners require the most supervision.

Nonetheless, this figure indicates that there is a prima facie case for believing that there are significant savings to be made within the criminal justice system by emphasising the distinction between violent and nonviolent offenders.  A number of the implications of this distinction are detailed in Chapter 4 of this report.

Key facts 2 In 1975, there were 8,900 people in Australian prisons.  In 2015, there were 36,000. The incarceration rate is now 196 per 100,000 of the adult population — the highest rate since just after federation. With the exception of the United States, Australia has a higher incarceration rate than other major common law countries (United Kingdom, Canada) and a much higher incarceration rate than continental European countries like France and Germany. The highest incarceration rates are in the Northern Territory and Western Australia. About 46 percent of prisoners nationwide are incarcerated for nonviolent offences. Australian governments spend up to $1.8 billion annually on incarcerating nonviolent offenders.

PRISONER CHARACTERISTICS

In the main crime tends to be committed by those of low socioeconomic status, including those with low education attainment, patchy work histories, lack of stable housing, broken families and substance abuse.  This section examines the statistical correlation between these social phenomena and criminality.  While none of these correlations should be read as causing crime, much less excusing it, they are important to understand in developing appropriate policy responses to the growing incarceration rate.

Educational attainment

Lower education attainment is correlated with higher crime and incarceration rates.  For example, according to a 2015 report by the Australian Institute of Health and Welfare, only 16 percent of those entering prison had year 12 as their highest level of education attainment;  while 72 percent had completed year 10 or lower.  And over half had no formal education other than schooling, while one-third had completed a trade certificate.  Further, few prison entrants were educated at tertiary level, with 4 percent completing a diploma, 2 percent a Bachelor's degree and 1 percent a postgraduate qualification. ⁽⁴³⁾

Similarly, a 2008 report by the Griffith Institute of Social Research found about 20 percent of prisoners had year 12 as their highest level of education attainment, while 30 percent had only completed year 10. ⁽⁴⁴⁾  A 2015 academic piece published in the journal Health and Justice found that 31 percent had attained their high-school certificate, certificate or a degree while 34 percent had not completed year 10. ⁽⁴⁵⁾  And the NSW Inmate Health Survey found 52 percent of men and 45 percent of women in NSW prisons did not finish year 10. ⁽⁴⁶⁾

Evidence abroad is consistent with research in Australia.  For example, a 2012 survey by the U.K Ministry of Justice reported 59 percent of prisoners stated they had regularly played truant from school, 63 percent had been suspended or temporarily excluded, and 42 percent stated that they had been permanently excluded or expelled.  Prisoners with these issues were also more likely to be reconvicted on release than those without. ⁽⁴⁷⁾

Similarly, a 2003 study of American crime and incarceration rates found that schooling significantly reduces the probability of incarceration.  And this result is driven by a reduction in criminal behaviour rather than the probability of arrest or incarceration condition on crime.  Further, the authors find differences in educational attainment between black and white men explain 23 percent of the black-white gap in male incarceration rates. ⁽⁴⁸⁾

Observation 3 Lower educational attainment is strongly correlated with increased incidence of incarceration, with as many as 80 percent of Australian prisoners having left formal schooling before completing Year 12.

Employment

Prison entrants are likely to have patchier work histories than the general population and are less likely to be working immediately prior to entering prison.

According to a 2015 study by the Australian Institute of Health and Welfare, almost one-half of prison entrants were unemployed immediately prior to prison, and a further 14 percent reported being unable to work due to disability, age or health conditions.  Slightly more than one-half of those who were unemployed were looking for work. ⁽⁴⁹⁾

A similar study conducted by Corrections Victoria estimated that about two-thirds of repeat offenders were unemployed at the time they re-offend (Victorian Department of Justice 2000-2001 cited in Graffam et al. 2004). ⁽⁵⁰⁾  And a 2015 study by the Victorian Ombudsman estimated that 63 percent of male prisoners and 45 percent of female prisoners were unemployed at the time they entered prison. ⁽⁵¹⁾

Furthermore, according to the NSW Inmate Health Survey, 50 percent of men and 67 percent of women were unemployed in the six months before their incarceration.  The report also found this unemployment tends to be quite entrenched, with about 30 percent of men and 44 percent of women being unemployed for five years or longer prior to incarceration. ⁽⁵²⁾

One needs be careful when interpreting these statistics as it seems likely the non-employment rate for criminals is a bit overstated.  This is because some of the surveys interview people's situations "immediately prior" to prison.  By this time many have been caught, convicted and sentenced and, as a consequence, have lost their job.  Additionally, crime is disproportionately committed by younger people who are less likely to be employed than their older counterparts in the general population.  For example, the employment rate for 18-24 year olds in July 2015 for the general population was 58.6 percent and 44.1 percent for 15-19 year olds, compared with the national average of 61 percent. ⁽⁵³⁾

Still, the preponderance of evidence suggests that the incarcerated have much worse employment outcomes and prospects than the general population.

Observation 4 Evidence suggest that perhaps more than half of prisoners were unemployed at the time of their crime, with many having been unemployed long-term.

Housing

Many studies have found that a large minority of former prisoners are without stable housing when released.  For example, the Australian Institute of Health and Welfare found that approximately a quarter of those entering prison were homeless or in short-term or emergency accommodation. ⁽⁵⁴⁾  And the Melbourne Institute estimated about one in five lack stable housing. ⁽⁵⁵⁾

There is also a well documented relationship between unstable housing and recidivism.  A 2006 study published in the Australian and New Zealand Journal of Criminology found of the ex-prisoners who did not move or moved housing just once about 22 percent had been reincarcerated within nine months of release, whereas of those who moved twice or more 59 percent were back in prison. ⁽⁵⁶⁾  Many of those involved in the trial indicated they moved because they had to, not because they chose to (although they may have had to move because of choices they made).

Some international evidence provides similar results.  For example, a 2012 report by the UK Minister of Justice found 15 percent of prisoners in their sample reported being homeless before custody compared with 3.5 percent general population who reported having ever been homeless.  In addition, close to 80 percent of prisoners who reported being homeless before custody were reconvicted in the first year after release, compared with 47 percent of those who did not report being homeless before custody. ⁽⁵⁷⁾

Family

A large and growing body of research suggests that people who commit crime and are incarcerated are more likely to come from dysfunctional families and/or single parent households.

For example, a 1987 report from the US on youth in custody found that about 70 percent did not grow up with both natural parents. ⁽⁵⁸⁾  And a 1994 study of juveniles in Wisconsin found only 13 percent grew up with their married parents. ⁽⁵⁹⁾

Similarly, 24 percent of prisoners stated they had been in care at some point during their childhood, according to the UK's Ministry of Justice report. ⁽⁶⁰⁾  Those who had been in care were younger when they were first arrested, and were more likely to be reconvicted in the year after release from custody than those who had never been in care.

Observation 5 Homelessness is correlated with an increased rate of incarceration.  Released prisoners without stable housing are also more likely to reoffend and return to prison.

According to the book, the Index of Leading Cultural Indicators, in the US children from single-parent families account for 63 percent of all youth suicides, 70 percent of all teenage pregnancies, 71 percent of all adolescent chemical/substance abuse, 80 percent of all prison inmates, and 90 percent of all homeless and runaway children. ⁽⁶¹⁾

The Heritage Foundation found that a 10 percent increase in the proportion of children living in single-parent homes on average leads to a 17 percent increase in juvenile crime. ⁽⁶²⁾

As Kay Kymowitz wrote in The Atlantic:

The bottom line is that there is a large body of literature showing that children of single mothers are more likely to commit crimes than children who grow up with their married parents.  This is true not just in the United States, but wherever the issue has been researched. ⁽⁶³⁾

Observation 6 Evidence shows that criminality is correlated with growing up in a single-parent family or in out-of-home care.

Drug use

There is also evidence that prisoners are more likely to engage in illicit drug use than the general population.  According to the Australian Institute of Health and Welfare, for example, prison entrants were typically 2-3 times as likely as the general community to report recent use.  In particular, cannabis use was reported by more than one-half (53 percent) of 18 to 24-year-old entrants, compared with just under one-quarter (23 percent) of their general community counterparts. ⁽⁶⁴⁾

Similarly, the drug most likely to be used by prison entrants, methamphetamines, was also the drug type with the largest difference in use compared to the general community, being reported at least 10 times as often by prison entrants as by the general community.  Among 18 to 44-year-olds, more than 50 percent of prison entrants reported using methamphetamines in the previous 12 months, compared with 5 percent or less in the general community. ⁽⁶⁵⁾

Further, another study by Corrections Services NSW found illegal drug use six months prior to imprisonment was reported by close to 75 percent of inmates.  The use of "heavy-end" drugs such as heroin, amphetamines or cocaine in the six months prior to imprisonment was reported by one in two inmates.  Thirty-five percent reported they had injected drugs. ⁽⁶⁶⁾

Observation 7 Criminals use illicit drugs at a higher rate than the rest of the community.

Age and gender

Imprisonment is overwhelmingly a young male phenomenon.  About 92 percent of those in prison in 2015 were male and over 50 percent were under 35 years old.  The ratio of male/female imprisonment has been roughly constant over the past decade, although the representation of women has increased slightly over recent years. ⁽⁶⁷⁾

Figure 11 Imprisonment by gender

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

Figure 12 Imprisonment by age 2015

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

Observation 8 Prisoners are overwhelming male.  Criminality is most prevalent among young people, between the ages of 20 and 39.

Indigeneity

Indigenous Australians are far more likely to be imprisoned than the non-Indigenous population.  For example, Indigenous Australians represent about 2.5 percent of the general population, but about 27 percent of the prison population.  And the incarceration rate of Indigenous Australians stands at 2,253 compared with non-Indigenous of 146. ⁽⁶⁸⁾  This equates to an incarceration rate ratio of 16.

Some of this difference is because the Indigenous population has, on average, a younger age profile than the non-Indigenous population (given younger people have higher incarceration rates than older).  When we take into account this difference, by calculating the incarceration if the Indigenous and non-Indigenous population had the same age profile, the incarceration rate for the Indigenous drops to 1731, compared with 146 for non-Indigenous. ⁽⁶⁹⁾  This represents a ratio of 12, compared with 16 for the crude imprisonment rate.

The growth of Indigenous Australians in prison has also far outstripped the growth of the non-Indigenous.  There was an increase of 75 percent of Indigenous Australians in prisons over the last decade compared to 33 percent increase for the non-Indigenous. ⁽⁷⁰⁾

There is a substantial difference between Indigenous Australians and the non-Indigenous in terms of the most serious offence for which they are incarcerated.  The Indigenous are far more likely have their most serious offence as acts intended to cause injury, unlawful entry, offence against justice procedures and robbery.  Non-Indigenous people are far more likely to be incarcerated for illicit drug offences — 17 percent compared to 3 percent.  Higher rates of non-Indigenous incarceration can also be seen for sexual assault and homicide. ⁽⁷¹⁾

Figure 13 Imprisonment by Indigenous status

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

The high number of Indigenous Australians in prison also may help to explain why, as we saw in Key Facts 2, above, that the Northern Territory has the highest incarceration rate.  Indigenous Australians constitute about 30 percent of the population of the Northern Territory.

Some suggest high Indigenous Australian incarceration rates are the result of institutional bias or racism.  However, Indigenous Australians generally serve shorter sentence lengths than non-Indigenous.  For example, the mean and median sentence lengths for Indigenous Australian prisoners is 2.6 and 1.2 years, respectively, compared with a mean 4.4 years and median of 2.1 for non-Indigenous.  There are also substantial differences in some categories of crime.  For example, in the case of homicide the median sentence length for the Indigenous is 9.9 years compared with 14.6 years for non-Indigenous. ⁽⁷²⁾  If there were institutional bias one would expect this to be reversed.

Further, the evidence suggests that unemployment and low education attainment are the two greatest "risk" factors for Indigenous incarceration, as opposed to factors relating to race or socio-economic status.  For example, unemployed Indigenous Australians are 20 times more likely to be imprisoned than employed Indigenous Australians.  Indeed, the incarceration rate for Indigenous people who were employed at the time of their offending was 332 per 100,000, compared with 6,495 for unemployed Indigenous people. ⁽⁷³⁾  Similarly, the imprisonment rate was 164 for Indigenous people who had completed school compared with 2,217 for those who had not. ⁽⁷⁴⁾

Figure 14 Percentage of crime by Indigenous status 2015

Source:  Australian Bureau of Statistics Prisoners in Australia 2015

According to the ABS, about 58 percent of Indigenous Australians who are imprisoned returned to prison within ten years of release, compared to 35 percent for the non-Indigenous. ⁽⁷⁵⁾  Similarly, according to the Centre for Independent Studies, Indigenous offenders are more likely to exhibit factors that lead to custodial rather than non-custodial sentences, such as a lengthy criminal record, being convicted for a serious violent offence or multiple concurrent offences. ⁽⁷⁶⁾

Further, there are also likely to be a range of cultural factors that influence the rate of incarceration of Indigenous Australians.  For example, in his book on Indigenous incarceration Dr Don Weatherburn notes that "arrest, prosecution and imprisonment may have become a rite of passage for young Aboriginal offenders rather than a source of shame or embarrassment". ⁽⁷⁷⁾

Weatherburn goes on to note that "for older offenders the attraction of free accommodation and food, good health care, relative safety from violence and regular social contact with relatives and friends sometimes far outweigh the negative aspects of incarceration". ⁽⁷⁸⁾  In other words, the relative conditions outside of jail are so bad that some would rather a life in jail.  This could be a reflection of the poor quality of conditions outside of jail, or the quality of conditions inside jail.

The Centre for Independent Studies quotes a police officer's observation about Indigenous crime and incarceration:

Crime occurs more in low socio-economic areas.  These are places where parents do not know and usually don't care where their children are;  where a lot of people do not work and do not want to work;  and where there is little respect for the police or for other people.  The crime is committed by both Indigenous and non-Indigenous people but it is unfortunately a fact that many Aboriginal people live in low socio-economic areas. ⁽⁷⁹⁾

Another factor contributing to the high rate is more people are identifying as being Indigenous. ⁽⁸⁰⁾  If the number of people who identify as Indigenous increases, then the observed number of Indigenous Australians being incarcerated will increase, all else being equal.

Observation 9 Indigenous Australians commit crime and are incarcerated at a far higher rates than the non-Indigenous.  This may partly be explained by the strong correlation between indigeneity and the other factors correlated with criminality, but may also depend in part on cultural developments within some Indigenous Australian communities.

---

3. EXPLAINING THE RISE IN THE INCARCERATION RATE

In this chapter, I will discuss the possible explanations for why the people described in Chapter 2 are in Australian prisons.  This includes a description of why criminals choose to commit crimes, an analysis of how public policy responds to their choices, and the effects of that response on the incarceration rate.

INDIVIDUAL CHOICE AND CRIME

All work on criminal justice is based on assumptions of how criminals behave.  There are two broad ways to view the issues:  socio-cultural causation and personal choice.

Criminology and explanations from the social sciences see crime as a function of the social environment and cultural conditioning.  In this sense crime is not viewed as a conscious choice but the inevitable result of social and cultural circumstances, such as poverty, race or drug addiction.

Researchers who adopt this approach often talk of certain people as being "predisposed" to crime and analyse the correlates or personal characteristics which determine the "risk level" or probability of committing crime.  The implication is that society is primarily responsible for crime because it creates and sustains the conditions in which crime takes place.  On this view, policing and punishment cannot deter crime;  crime is reduced most effectively by social policy aimed at eliminating the factors which are held to cause crime.

An alternative perspective brought from economics sees criminals not as a special category of people for whom particular and unique theories of behaviour are needed, but as regular people who base their decisions to commit crime on analysis of the costs, benefits and risks of those crimes.  This approach suggests there is no fundamental difference between someone who chooses to steal for a living rather than work and someone who works instead of steals.  Both are making a decision that they see as being in their own interests.  As Gary Becker notes, "Some persons become criminals not because their basic motivation differs from that of other persons, but because their benefits and costs differ". ⁽⁸¹⁾  This approach ultimately assumes that people have free will and exercise that will when committing crime.

The benefit of the economic approach is that it accommodates the standard criminology and social sciences view:  that one's social environment, upbringing and culture affect individuals' attitudes toward crime, through influencing their preferences to crime relative to other options.  As an example, someone living in a neighborhood with a high prevalence of crime may be more likely to internalise crime as something normal and legitimate (because, in this situation it effectively is) and therefore develop different preferences with respect to crime than someone living in a low crime neighbourhood.  At the same time there also many who are brought up in these environments and do not commit crime.  The choice to commit crime — as with all other choices — is ultimately made by the individual.

Similarly, crime could also be normalised where someone's family member has a history of crime and incarceration.  This is borne out by the documented effect that parental incarceration has on children's life outcomes.  Similarly, where time in jail is seen as a rite of passage for acceptance into a group or gang, crime might seem more preferable.

In the economics view, the key to understanding why crime is committed is to analyse the costs and benefits of crime from the perpetrators' perspective.  There are two main costs of crime:  direct costs which involve punishment including incarceration and fines, and opportunity costs which comprise that which perpetrators of crime give up in order to commit crime, such as forgone earnings from engaging in legal work, now and in the future.  There are a range of offsetting benefits perpetrators receive from crime, from financial gain through to the satisfaction of revenge and gratification from assault or homicide.

Importantly, the economics framework can also help us make sense of the widely documented economic and social correlates of crime, detailed in Chapter 2.  For example, there is a large and growing body of research which suggests that criminals, in the main, have lower education attainment than the general population.  This means they are likely to be less skilled, less productive and have lower expected lifetime earnings from legitimate work than the general population.  Additionally, incarceration also means lost wages resulting from time out of the labour market as well as flow-on consequences for finding future work, resulting from one having a criminal record and loss of work-relevant skills whilst in jail.

These factors mean that criminals, and in particular those who have been incarcerated, have lower expected wages than their counterparts in the general population, which in turn means they have a lower opportunity cost of committing crime.  It therefore makes sense that people who commit crime are more likely to have lower education attainment, as relatively poorer workplace prospects reduce the cost of crime.

Similarly, a central element of the economic approach to understanding how people behave is analysing how they allocate their time and money in the present relative to the future (intertemporal substitution).  Generally speaking, people who are more patient and are willing to defer gratification discount the future less than those who are impatient and seek immediate gratification.  Crime itself is typically an example of this behaviour:  the payoffs are, in the main, immediate and certain, while the costs in terms of detection, arrest and conviction are distant and uncertain.  As noted in Chapter 2 of this report, criminals and the incarcerated are more likely to engage in a range of other behaviours that have upfront and certain benefits with distant and uncertain costs, such as heavily consuming alcohol, drugs and tobacco.  The inverse can also be observed:  criminals are more likely to avoid behaviour where the costs are immediate and certain but the benefits are uncertain and distant, such as education.

This has obvious implications for how to reduce crime — increase both the direct and opportunity costs of committing crime.  Increasing direct costs can be achieved through increasing the probability of detection (more police on the streets) and/or increasing the severity of punishment (such as sentence length).  Increasing the opportunity costs involves increasing the return to non-criminal activities.  This would occur, for example, through higher expected wages (either higher absolute wages or more aggregate opportunities at a lower wage) which is closely related to more economic opportunity and better education outcomes.

As James Q Wilson noted:

Deterrence and job-creation are not different anti-crime strategies;  they are two sides of the same strategy.  The former increases the costs of crime;  the latter enhances the benefits of alternatives to criminal behavior. ⁽⁸²⁾

With this model in mind, I will now consider the possible reasons for the increase in the incarceration rate.

THE CRIME RATE

The most obvious possible explanation for the increasing incarceration rate is that more people are choosing to commit crimes.

It is difficult to be certain about whether this is true or not.  However, a fair reading of the data suggests that the crime rate has been increasing over recent years.

It is important to remember that data is limited, and sometimes patchy and inconsistent across jurisdictions.  And even in a world of perfect data the actual crime rate would never be known because many crimes are unreported. ⁽⁸³⁾  The ABS has summarised some of the data issues:

When examining our statistics it must be remembered that not every crime is reported to the police, not every crime that is reported is recorded, not every crime that is recorded is investigated, not every crime that is investigated is cleared ("solved"), not every crime that is investigated yields a suspect, not every suspect is apprehended, not every apprehended person is charged, not every charged person is brought before the courts, not every person brought before the courts is convicted, and not every convicted person is imprisoned.

So then, taking a snapshot at any time, or taking a series of data over time on any activity (e.g. court appearances, prisoner numbers, etc.) does not tell us about crime in Australia, but rather about that particular activity or segment. ⁽⁸⁴⁾

No matter which way the data is analysed, an objective understanding of the crime rate is difficult to obtain.  For example, the public data available includes:

• The number of offenders;  but this does not provide information on the number of crimes committed per offender.
• The number of proceedings made by police per offender;  but this doesn't include the number of alleged crimes per proceeding.
• The number of victims;  but this doesn't include victimless crimes or the number of crimes committed against a given victim.

Additionally, the majority of data is not available over long time periods, so it can be difficult to draw strong conclusions about longer run trajectory of the crime rate.

With these limitations in mind it appears that:

• the overall offender rate has been relatively flat
• the adult offender rate has increased
• the juvenile offender rate has decreased
• the number of proceedings per offender has increased, and
• the victimisation rate has decreased.

The combination of a rise in the adult offender rate and the number of proceedings made by police per offender suggests the adult crime rate has increased, which explains some of the rise to the incarceration rate.  The fact that the victimisation rate has decreased suggests the rise in the adult crime rate is mostly from victimless crimes (or there has been a reduction in the number of crimes per offender and per proceeding.)

NUMBER OF OFFENDERS

In 2014-15 there were an estimated 411,000 offenders nation-wide.  About 80 percent, or 326,000, were adults (18+) and about 20 percent, or 85,000, were juveniles. ⁽⁸⁵⁾  On an average annual basis the number of offenders has grown roughly in line with overall population growth at about 1.7 percent. ⁽⁸⁶⁾  This leaves the aggregate offender rate relatively unchanged since 2008-9.

Breaking down by youth and adult shows that the steadiness in the overall offender rate is the result of the adult rate increasing and the youth rate decreasing.  The youth offender rate has decreased from 4898 per 100,000 youths in 2008-09 to 3739 in 2014-15 — a 30 percent decrease.  While the adult offender rate has increased by 15 percent from 1429 per 100,000 of the adult population to 1639 over the same time period. ⁽⁸⁷⁾

Table 8 Offender rates ⁽⁸⁸⁾

The adult offender rate has increased closely with the increase in the incarceration rate, which intuitively makes sense given we would expect a positive relationship between the number of offenders and the number of people in jail.

Of course, while the offender rate gives some indication of the crime rate — an increase in offenders would suggest more crime and vice versa — it doesn't give us the whole picture.  This is because a given offender can commit more than one crime, so knowing the number of crimes committed per offender is an important piece of information.

Unfortunately, the data on the number of crimes per offender is not publically available.  But the number of proceedings made by police per offender is available, which can be used as a rough proxy for the crime rate.  The table below gives the average number of proceedings made by police per offender by state (data for WA isn't available).

Figure 15 Adult offender rate versus incarceration rate

Source:  my Calculations using Australian Bureau of Statistics Recorded Crime — Offenders

Table 9 Mean number of proceedings per offender ⁽⁸⁹⁾

The average number of proceedings per offender has increased since 2008-09 in NSW, Victoria, Queensland, and the NT, stayed flat in South Australia and decreased slightly in Tasmania and the ACT.

Using a weighted average I estimate the national average police proceedings per offender to have increased by 14 percent from 1.4 in 2008-09 to 1.6 in 2014-15. ⁽⁹⁰⁾

The graph below plots the estimated crime rate, which is the number of offenders multiplied by the average number of proceedings per offender.  Again we see a sharp increase in the crime rate and a close association with the incarceration rate, consistent with intuition.

Figure 16 Crime rate and incarceration rate

Source:  my Calculations using Australian Bureau of Statistics Recorded Crime — Offenders

A limitation to using the number of proceedings as a proxy for crimes committed per offender is more than one offence can be committed per proceeding.  So theoretically it is possible that the number of offences per proceeding has increased, decreased or stayed the same.

The vast majority of offenders are only proceeded once by police, ranging from 85 percent in the ACT to 66 percent in the NT. ⁽⁹¹⁾

Table 10 Number of proceedings per offender (percentage of total) ⁽⁹²⁾

Key facts 3 The number of offenders and the offender rate has been fairly flat over recent years. The adult offender rate (which is what is relevant when analysing the incarceration rate) has increased sharply. The growth in the adult offender rate is in line with the growth to the incarceration rate. The juvenile offender rate has decreased sharply. The average number of proceedings made per offender has increased slightly from 1.4 in 2008-09 to 1.6 in 2014-15.

VICTIMS

The number of victims has decreased substantially since 2000, down from 1.3 million to 750,000 in 2014. ⁽⁹³⁾  Crimes counted are:  homicide, sexual assault, kidnapping, robbery, blackmail/extortion, unlawful entry with intent, motor vehicle theft and other theft.  Unfortunately, ABS data for assault is only for certain states, which does impose a limitation on my interpretation of the overall number of victims given assaults are not a trivial occurrence.

Similarly, the victimisation rate — the number of reported victims as a percentage of the population — has decreased for a number of crimes from 2008/9 to 2014/15, for example:

• physical assault has decreased from 3.1 to 2.1;
• face-to-face physical assault has decreased from 3.9 to 2.9;
• robbery is down from 0.6 to 0.3;
• sexual assault is relatively flat at 0.3;
• break-ins are down from 3.3 to 2.7;  and
• malicious property damage is down from 11.1 to 5.7. ⁽⁹⁴⁾

Figure 17 Number of reported victims, selected offences

Source:  Australian Bureau of Statistics Crime Victimisation, Australia 2014-15

Changes in recording and reporting make it difficult to compare data over time and across jurisdictions, as can changes in the prevalence of reporting amongst victims.

That adult offender rates and number of proceedings per offender have increased while the victimisation rate has decreased could indicate that the rise in offending is largely constituted by victimless crimes.  It could also indicate there are more crimes being committed with multiple offenders per victim.

Figure 18 Victimisation rate as percentage of Australian population

Source:  Australian Bureau of Statistics Crime Victimisation, Australia 2014-15

Since 2008-09, the number of nonviolent offenders has increased by close to 20 percent, violent offenders increased by 5 percent and the number of people who have committed offences against property has decreased by 1 percent.  Illicit drug offenders have increased the most, by about 40 percent since 2008-09. ⁽⁹⁵⁾  Obviously, it is not clear if this is because more people are committing illicit drug offences, or more are being caught.

Figure 19 Offenders by type of offence

Source:  Australian Bureau of Statistics, Recorded Crime — Offenders 2014-15

Figure 20 Percentage change in incidence of crime type 2008-9 to 2014-15

Source:  my Calculations using Australian Bureau of Statistics Recorded Crime — Offenders

Finding 1 Based on available data regarding offending and the number of victims, the growth in the incarceration rate is partially attributable to a rise in the adult offending rate.

Reoffending rates

While a rising crime rate may be contributing to the incarceration rate increase, there is one aspect of offending that is not playing a role.  Approximately 59 percent of the prison population has been incarcerated more than once. ⁽⁹⁶⁾  However, this has changed little over the last 10 years.

State and territory figures on this measure have also been generally stable over this period.  Higher rates are generally found in jurisdictions with smaller populations, with the exception of South Australia, which has the overall lowest rate.

Figure 21 Percentage of prisoners who have been imprisoned before

Source:  Australian Bureau of Statistics, Prisoners in Australia 2006-2015 inclusive ⁽⁹⁷⁾

The stability of this measure means that the overall growth in the incarceration rate is driven equally by the imprisonment of reoffenders and new offenders.

In turn, this suggests that the poor performance of prisons in preventing re-offending is one of the drivers of the growing incarceration rate in the sense that were they performing better the overall growth would be lower (all else being equal).  But they are not performing any worse in this respect than they have done in the past;  the population of those who have been imprisoned has increased and therefore so has the number of people jailed more than once.

Figure 22 Percentage of prisoners who have been imprisoned previously by state

Source:  Productivity Commission Report on Government Services 2016

And for most of those who are jailed multiple times, their return comes quite swiftly.  Nationally 44.3 percent of prisoners released in 2014-15 returned to prison within two years. ⁽⁹⁸⁾  This figure has increased by more than 12 percent over the last five years.

This suggests that the performance of prisons in prevent reoffending is getting worse.  Reversing this trend would make a substantial contribution to limiting the growth of the incarceration rate.

Figure 23 Nationwide percentage of released prisoners who return to prison within two years of release

Source:  Productivity Commission Report on Government Services 2016

Finding 2 The proportion of prisoners with prior convictions has been stable over the past decade.  It therefore has not contributed to the rise in the incarceration rate.  However, there is reason to believe that prisons are getting worse at preventing re-offending.

PUBLIC POLICY:  COURTS AND THE ADMINISTRATION OF CRIMINAL JUSTICE

Another possible contributor to the increase in the incarceration rate are changes in the administration of justice:  remand, sentencing, parole.  In this section we find that in addition to being partially explicable in terms of the behaviour of individuals, the increase in the incarceration rate has also been caused in part by public policy.

Bail and remand

A large proportion of the increase to the total incarceration rate was due to increases in unsentenced offenders (those on remand). ⁽⁹⁹⁾  For example, since 2005 the number of unsentenced prisoners has increased by 93 percent, compared with 29 percent for sentenced prisoners. ⁽¹⁰⁰⁾  This has resulted in the proportion of total prisoners who are unsentenced to increase from about 20 percent to 28 percent.

People on remand have increased as a proportion of Australian prisoners, from 22 percent in 2006 to 27 percent in 2015. ⁽¹⁰¹⁾  Unlike the increase in the sentenced prison population this is not likely attributable to people being remanded longer.  As the figure below shows, the average time spent on remand has not changed over the last 10 years across all courts.

Figure 24 Sentenced and unsentenced

Source:  Australian Bureau of Statistics, Prisoners in Australia 2015

People spend an average of 4.9 months on remand.  While there has been an increase in time spent on remand in higher courts, from an average of 9.1 months to an average of 10.4 months, most proceedings take place in lower courts and so overall people are not spending more time on remand.

This can be seen more clearly by breaking down the unsentenced prison population by the time they are spending on remand.

Figure 25 Mean time spent on remand by court

Source:  ABS Prisoners in Australia 2006 and 2015

This suggests that inefficient processing by the judicial system is not a contributor to the increase in the prison population.  In turn, this implies that the number of people on remand has increased because bail is either not being granted or is being revoked more frequently.

New South Wales is the state with the most remandees and has one of the highest proportions of unsentenced prisoners.  Recent analysis by the NSW Bureau of Crime Statistics and Research (BOCSAR) shows that in that state the court and police bail refusal rates have increased in the last three years, since changes were made to the NSW Bail Act. ⁽¹⁰²⁾

Figure 26 Proportion of unsentenced prisoners by time spent on remand

Source:  ABS Prisoners in Australia 2006 and 2015

BOCSAR further attributes the growth in the remand population to an increase in bail breaches leading to bail refusal and an increase in the number of people against whom criminal proceedings were commenced.

A rising refusal rate along with more breaches necessarily means that more people are being remanded.  If other states are experiencing a similar phenomenon, this would partially explain the increase in the nationwide remand population.  There is some evidence to suggest that they are.

Figure 27 NSW rate of bail refusal at first court appearance

Source:  NSW Bureau of Crime Statistics and Research

The Senate Inquiry into Justice Reinvestment heard that more stringent bail conditions make it difficult for many offenders to successfully request bail and meet court requirements.  Following legislative changes in March 2011, the Northern Territory saw an increase of 67 percent in the number of bail breaches recorded.  There has also been a significant rise in the remand population in Western Australia. ⁽¹⁰³⁾

Figure 28 NSW number of established breaches of bail

Source:  NSW Bureau of Crime Statistics and Research

Finding 3 A large part of the rise in the prison population is constituted by unsentenced prisoners.  The growth in unsentenced prisoners is partly attributable to more stringent conditions on the granting and keeping of bail.

Guilty pleas

It is sometimes argued that the rise in the incarceration rate is partly attributable to more people taking plea bargains.  It is further implied that they are submitting to these deals under duress resulting from delays in the administration of justice and prosecutors charging defendants with more serious crimes even where they are not readily supported by the evidence.

In Australia, the evidence for this thesis is weak.  As noted in the section on remand, the average time spent on remand has not increased in lower courts, where the great majority of criminal cases are decided, while there was a small increase in the nation's higher courts.  The increase, however, has not led to a large increase in the number of guilty pleas in those courts.

This rise in the number of guilty pleas is offset by the fact that the total number of higher court adjudications in Australia has fallen in the past five years by 5.4 percent.  This has led to a 4.3 percent reduction in the number of people being found guilty in higher courts.  An estimate based on incomplete data indicates that this translates to a rough increase of less than 1 percent in the number of guilty pleas in higher courts across the country. ⁽¹⁰⁴⁾

There is little evidence that courts are over-worked.  The number of adjudications has stayed stable or decreased in almost every jurisdiction over the past five years, with the exception of Victoria.

Figure 29 Higher courts:  guilty pleas as a proportion of total adjudications 2010-11 vs 2014-15

Source:  ABS Criminal Courts Australia 2010-11 and 2014-15

In Australia in recent years, defendants are not spending more time on remand and are not pleading guilty at higher rates.  The nation's courts are processing roughly the same number of defendants each year.  Therefore, if defendants are being pressured to plead guilty, this problem has not worsened in this period and therefore is not a contributor to the rising levels of incarceration in the states and territories.

Finding 4 There is no evidence to suggest that the increase in the prison population has been driven by more defendants pleading guilty.

Sentencing

All else being equal, longer sentences necessarily contribute to increased incarceration rates.  In the United States, it has been estimated that longer sentences have contributed up to half of the increase in the prison population (at state level). ⁽¹⁰⁵⁾  The lengthening of prison terms in that country has been driven by the use of mandatory minimum sentences, "three strikes" and similar laws aimed at re-offenders, and truth-in-sentencing laws making suspended sentences and parole rarer.

However, these laws are not used frequently in Australia.  Australian courts and parole boards have retained more discretion to ensure proportionality in sentencing.  Some states have implemented versions of them:  for example, and as discussed later in this chapter, Victoria has abolished suspended sentences, while WA and the Northern Territory have used mandatory minimums.  But these are exceptions.  In Australia then, only part of the increase in the prison population is likely attributable to convicted criminals being given longer sentences.  And this contribution is reduced further by the large number of prisoners who do not serve their full terms.

Figure 30 Total number of adjudications by jurisdiction (all courts) 2010-11 v 2014-15

Source:  ABS Criminal Courts Australia 2010-11 and 2014-15

Average sentences have gotten longer

Over the last 10 years, the average sentence for Australian prisoners has increased.

As Figure 30 shows, however, the median sentence has increased only slightly.

Together, these statistics suggest that offences which already attracted custodial sentences now attract heavier penalties.  The rates of increase in the number of sentences greater than the median and in those less than the median have been similar.  But those sentences in the former group have grown substantially, while those in the latter group are now closer to the median, boosting the overall average.

This can be seen by breaking down the prison population by sentence length.

Figure 31 Median and mean sentence lengths 2006 v 2015

Source:  ABS Prisoners in Australia 2006 and 2015

There has been a substantial increase in the number of sentences greater than 12 months, with the largest increase in the five to ten year bracket.  Although there has been a reduction in the proportion of sentences of less than 12 months, given that the overall number of sentences handed down has increased, it seems likely that this is because a substantial number of people who would have previously received short sentences are now receiving sentences closer or greater than the mean.  Every bracket beyond the 12-month mark has increased, suggesting an across-the-board toughening of sentences.

And in fact sentence lengths have increased on average for most of the crime categories for which there are statistics.

Figure 32 Proportion of prisoners serving different sentence lengths

Source:  ABS Prisoners in Australia 2006 and 2015

But the amount of time served has not increased

Although sentence lengths have increased, offenders are not generally spending more time in prison than 10 years ago.  ABS statistics reveal that prisoners' expected time to serve (ETTS) has been stable over the past 10 years, with the only substantial increase being for homicide.

This graph implies that there are more offenders being released after serving less than the median time but that those who are serving more than the median are serving slightly longer sentences on average than they did before.

This pattern, however, is quite inconsistent across the various crime types.

Figure 33 Median and mean sentence lengths by crime type 2006 v 2015

Source:  ABS Prisoners in Australia 2006 and 2015

Figure 34 Median and mean expected time to serve 2006 vs 2015

Source:  ABS Prisoners in Australia 2006 and 2015

Figure 35 Median and mean expected time to serve by crime type 2006 v 2015

Source:  ABS Prisoners in Australia 2006 and 2015

Figure 36 Median sentence length and median expected time to serve 2006 v 2015

Source:  ABS Prisoners in Australia 2006 and 2015

Prisoners are being incarcerated longer for homicide but not necessarily for other violent offences, with time served getting shorter for abduction and robbery.  There are similar small variations among nonviolent offences.

Combining the data, it becomes apparent that even though sentence lengths have gotten longer, this hasn't necessarily translated into prisoners staying in prison longer.

Indeed, for drug and weapon offences the expected time to serve has decreased even as sentences have gotten harsher.  For serious crimes like homicide and sexual assault, prisoners are staying in prisons longer although the length of their expected stints in prison have not grown as quickly as their sentences.

However, for no crime is the median expected time to serve in 2015 equal to or greater than the median sentence length for that crime in 2006.  This may not be explained by parole being given to more prisoners — as the section on parole statistics below will show, there is good reason to believe that parole is being refused more frequently.

Nor is it explained by legal and policy differences at the jurisdictional level.

Figure 37 Percentage changes in number of prisoners in each ETTS bracket 2006 vs 2015 (mainland states)

Source:  ABS Prisoners in Australia 2006 and 2015

Across the mainland states, ⁽¹⁰⁶⁾ the number of prisoners increased uniformly in the ETTS brackets less than the nationwide median value.  There was also a general increase in the number in the longest ETTS brackets, especially in the 20 years and more bracket.

This has slightly changed the composition of the prison population, with the proportion of prisoners expected to serve longer stints in prison increasing most significantly (although this is offset in part by the reduction in the proportion of prisoners serving life sentences).

There are more prisoners in the longest ETTS brackets and there are also more in the shortest brackets.  The bulk of the growth in the sentenced prisoner population are expected to serve long or short sentences — there has been less growth in the number of prisoners expected to serve between five and 15 years (despite the growth in the number of sentences of between five and ten years).  This explains why the mean ETTS has not increased.

Figure 38 Change in percentage of prisoners in each ETTS bracket 2006 vs 2015

Source:  ABS Prisoners in Australia 2006 and 2015

It further shows that any attempt to reduce the overall incarceration rate by reducing time spent in prison would confront the political and moral reality that offenders serving increased sentences are in prison for good reason and there will be no public appetite for shortening their terms of imprisonment.

Conversely, reducing the prison population by allowing more people currently expected to serve short sentences to serve those sentences outside the prison system would increase the mean ETTS:  there would be fewer people in prison but those who are in prison would stay there longer, on average, all else being equal.  This would be more politically and morally practical as it would concentrate prison system resources on the worst offenders.

Finding 5 While sentence lengths have increased, this has not translated to prisoners spending more time in prison.  Expected time to serve has increased for the worst offenders.  Reform efforts could focus on using alternatives punishments in place of short prison sentences.

Prison population changes

The composition of the prison population in terms of crimes committed has changed over the past 10 years.  There are proportionally more people in prison for acts intended to cause injury and for drug offences than there were 10 years ago.

Figure 39 Prison population (including remandees) by crime type 2006 v 2015

Source:  ABS Prisoners in Australia 2006 and 2015

Prison population changes do not track increases in sentence lengths

Reading the charts in the two previous sections together, we can see a relationship between sentence lengths and the incidence of particular crimes emerge.

Stiffer penalties have coincided with reductions in the proportional incidence of certain types of crime.  Significant examples include: ⁽¹⁰⁷⁾

• The average sentence for homicide has increased 27.1 percent, from 14.5 years to 19.9 years.  Over the same period, the percentage of prisoners incarcerated for homicide fell 18.7 percent from 10.2 percent to 8.3 percent.
• The average sentence for sexual assault has increased 16.8 percent, from 7.6 years to 9.1 years.  Over the same period, the percentage of prisoners incarcerated for sexual assault fell 9.3 percent, from 12.4 percent to 11.3 percent.

Similarly, weaker penalties on average have coincided with an increase in the proportional incidence of other crimes:

• The average sentence for acts intended to cause injury has decreased slightly (approximately 0.8 percent), while the percentage of prisoners incarcerated for this offence has risen 27.6 percent from 22.9 percent to 29.2 percent.
• The average sentence length for abduction, harassment, and other offences against the person has decreased by 14.9 percent, while the percentage of prisoners incarcerated for this offence has increased, from a low base, by 85 percent from 0.7 percent to 1.3 percent.
• The average sentence length for property damage and environmental pollution has decreased by 18.9 percent, from 2.6 to 2.2 years, while the percentage of prisoners incarcerated for this offence has increased by 20.6 percent from 1.1 percent to 1.3 percent.

Figure 40 Proportion of population (including remandees) by crime type 2006 v 2015

Source:  ABS Prisoners in Australia 2006 and 2015

While these data may indicate that there is a correlation between sentence lengths and the incidence of a crime, some offences do not fit this pattern:

• Robbery and extortion, theft, and fraud and deception have all seen reductions in both average penalties applied and the proportion of prisoners incarcerated for those crimes.  Note however that prisoners incarcerated for these crimes (and indeed, for almost every category of crime) have increased in absolute terms, just at a lower rate than other crimes.
• Conversely, prisoners incarcerated for weapons, negligence and drug offences have all proportionally increased even as penalties for those offences have also increased.

Mandatory sentences

By definition, mandatory sentencing laws mean a prison sentence is imposed on some offenders regardless of whether the judge would have supported such a sentence.  However, in Australia, mandatory minimum sentences only exist in Western Australia and the Northern Territory.

Therefore the overall contribution of these laws to median sentence lengths and thereby to the current incarceration rate is quite small.

Should these laws become more common across the country, they will likely contribute to a rise in the incarceration rate in the future.  This has been the experience in the United States, where these laws are widely used.

Figure 41 Community-based corrections orders v full-time custody in Victoria 2014-15

Source:  ABS Corrective Services December Quarter 2015

Suspended sentences

On 1 June 2014, Victoria abolished suspended sentences.  Offenders for whom immediate incarceration was not deemed essential to justice were from that date to be placed on Community Corrections Orders (CCOs), which had been introduced in 2012.

Over 2014-15, the number of Victorians serving community-based corrections orders (as classified by the ABS, this category includes all such orders and not just CCOs) ⁽¹⁰⁸⁾ has increased substantially.  The prison population has also grown over that time.

It seems likely, however, that in Victoria a number of people who might have otherwise been given suspended sentences are now receiving CCOs.  The differences between suspended sentences and CCOs is that the latter may include a number of impositions on the liberty of offenders, like curfews and restrictions on movement.  Because of the level of monitoring required, CCOs are more expensive than suspended sentences.

A recent NSW study found that the reoffending rate for offenders serving suspended sentences was similar to that for offenders who were incarcerated.  Given that suspended sentences have lower social and financial costs, abolishing them is possibly counter-productive economically. ⁽¹⁰⁹⁾

However, even though CCOs are slightly more expensive than suspended sentences, they are also potentially more coercive too.  This means that they may be a more appropriate punishments than the minimally-coercive suspended sentence.  This in turn might make it possible to use these orders as an alternative to prison for some offenders for whom a suspended sentence is inappropriate.  While the effects of the shift to CCOs in Victoria will need to be monitored, this measure may prove to be an effective alternative to prison for suitable offenders.

PAROLE

Evidence suggests that parole is becoming more difficult to obtain.

All else being equal, as the number of people serving custodial sentences increases, the number of people released into the community on parole should rise.  As the graph above shows, however, this is not the case in Australia.

Figure 42 Parolees v sentenced prisoners 2006-2015

Source:  ABS Corrective Services Reports

There could be a number of reasons for the divergence of these two numbers.  It could be explained by changes in the composition of the prison population — the proportion of prisoners whom it is safe to release on parole may have gotten smaller, for example.  It could also be explained by regulatory changes making parole requirements more difficult to satisfy, such that some people who in previous years might have been granted parole are now refused it.  Given that the median expected time to serve has been steady, parole may be being refused more often to offenders serving long sentences.

Parole conditions vary from jurisdiction to jurisdiction, with different legislative requirements affecting the circumstances of automatic release and the factors considered in parole hearings.  There are no national statistics for how frequently parole is refused and not every jurisdiction reports the data required to determine the parole refusal rate.

From the data that is available, however, there is evidence from some jurisdictions that parole boards have become less generous in their consideration of applications.

In Victoria, for example, the parole refusal rate in 2014-15 was 38.5 percent, up from 13.6 percent in 2005-06.  Over this period, the number of paroles granted went down and the number of refusals went up. ⁽¹¹⁰⁾  In Western Australia, the trend was even stronger.  In 2014-15, 64.5 percent, up from 28.9 percent in 2005-6.  The ACT's refusal rate increased from 22.7 percent in 2005-6 to 38.1 percent in 2014-15. ⁽¹¹¹⁾

Of jurisdictions with data available, the only one to buck the trend was NSW, where the refusal rate in 2014 was 27.9 percent compared to the 2005 figure of 36.8 percent. ⁽¹¹²⁾  However, compared to the average rate for 2006-2010 of 32.9 percent, the reduction is not as pronounced.

There are no national statistics for the revocation of parole.  However, some states do track these figures, and in those states the number of revocations has risen.

Emerging evidence from the United State suggests that applying minor punishments, such as short stints in jail, for minor and technical parole breaches is a more effective way to manage parolees. ⁽¹¹³⁾

Figure 43 Revocations of parole

Sources:  NSW Parole Authority Annual Reports;  Adult Parole Board of Victoria Annual Reports;  Queensland Department of Corrective Services Annual Report 2005-6;  Queensland Parole Boards Annual Report 2014-15;  Western Australia Parole Board Annual Report 2006;  Prisoners Review Board (WA) Annual Report 2014-15

Finding 6 While complete data is not available, there is some evidence that parole has become more difficult to obtain and hold, which may affect the incarceration rate.

CONCLUSIONS

The rise in the incarceration rate is a product both of more individuals choosing to commit crimes and public policy choices that favour imprisonment.

In this chapter, I argued that crime is a choice that individuals make when they deem it to be in their best interests.  Their understanding of their interests is coloured by a range of factors, including some of the characteristics described in Chapter 2.  We saw that there is good reason to believe that there are more people now choosing to commit crime and this is contributing to the rising prison population.

While more crime may be being committed, more people in prison is not a necessary consequence.  There are also policy decisions that are contributing to the increase.  The evidence suggests that the bail and parole have become more difficult to get and this has led to an increase in the prison population.  Courts have also been imposing longer sentences, which could be a contributing factor as well, although it should be noted that prisoners are serving on average the same amount of time as in the past.

---

4. REFORM DIRECTIONS

The prison population and its attendant costs are growing.  Up to 46 percent of people being incarcerated committed nonviolent offences.  And governments are making public policy choices that promote the use of incarceration as a punishment.  It is therefore fair to question whether all the resources being devoted to criminal justice are being used efficiently and beneficially.

It is my contention that any consistent approach to reducing incarceration begins with addressing the large number of nonviolent offenders in Australian prisons because:

• Punishments must be proportional to the offence;  and
• The primary purpose of prison is the protection of the community.

The guiding principle of punishment is proportionality:  the punishment must fit the crime.  The growing prison population implies that the number of people or offences for which prison is appropriate has grown.  But there is no reason to expect that this is the case.  Instead, the growing prison population reflects a widespread failure in the criminal justice system to respect the principle of proportionality.

Reducing the prison population is a matter of rationalising which criminals should and should not be subject to incarceration.  I argue that the general rule upon which to conduct this rationalisation is the distinction between violent and nonviolent crime.  Prison is proportional for the violent and antisocial because they are the only ones who need to be isolated from the community, which is a unique function of prison.

Nonviolent offenders are less likely to pose a physical risk to the community and therefore less likely to need to be isolated in prison.  Furthermore, they can still be punished in a way that fits their crimes, even without prison.

This chapter describes how the principle of proportionality informs criminal punishment and makes some observations about the implications of this principle.  I argue that proportionality implies that as a general rule nonviolent offenders should be given punishments other than prison, provided there is reason to believe they will respond to those punishments.  I go on to discuss alternatives to prison, the proper limits of the criminal law, and strict liability issues.  I also argue that it follows from proportionality that policing should be the focus of justice reinvestment efforts.

THE PRINCIPLE OF PROPORTIONALITY

Proportionality in sentencing is the principle that the punishment must fit the crime.  Typically, this involves matching the severity of the punishment to the harm caused by the defendant, taking into account any mitigatory factors like remorse and prior good character.  By linking punishment to the harm caused by the offender, this principle places a limit on the extent to which society can use criminal justice to pursue its interests, including its safety.

This principle has been criticised for being uncertain.  Any judgement of proportionality is necessarily only approximate.  Criminal justice does not operate as a market and as such there is no system akin to pricing whereby we can measure the preferences of criminals and victims.

In place of this information we have only the priorities of the government, which are usually politically determined and therefore operate as distortions of the criminal justice system. ⁽¹¹⁴⁾  For example, politicians need achievements to campaign on when running for re-election and therefore favour the creation of new laws and new penalties to the rationalisation of laws and penalties.

However, while this political reality is undeniable, it is also true that the principle of proportionality has intuitive force:  we do not impose the death penalty for jaywalking, no matter how we might desire to deter jaywalkers.  Unless it can be seriously argued that the strongest punishments can be justifiably applied to the most trivial offences, the argument around proportionality is one of degree, not principle.

Taking this principle seriously has a number of implications for how nonviolent criminals should be punished.  The rest of this chapter makes some observations about policy changes that follow from this principle.

THE PURPOSE OF PRISON IS TO PROTECT THE COMMUNITY

Violent offenders should receive jail sentences

Prison serves a number of purposes.  It satisfies the community's need for retribution and it deters people from committing crime.  For those who are given custodial sentences, prison can also be a part of their rehabilitation and reintegration into society.  But prison's main function is to isolate individuals whose liberty threatens the safety and tranquillity of the community.

There are a variety of philosophical arguments for why this should be so, ⁽¹¹⁵⁾ but it is sufficient for our purposes to note that community safety is the only aim of criminal justice that cannot be achieved by other reasonable means.  Rehabilitation can be facilitated through community-based programs and deterrence can be achieved through inflicting other types of hardship on offenders.  Society might content itself with other forms of punishment.  But while partial measures like home detention offer some security, the only way for the community to be safe from a criminal — capital punishment aside — is a prison.

For this reason, prison is a proportionate punishment for those criminals who pose a risk to the community because they have demonstrated a propensity for violence or a pattern of anti-social behaviour.  Physical harm demands a harsher punishment for two reasons:

• Physical harm causes suffering throughout the healing process, which may never end.  This is a cost to the victim that cannot be repaired.
• Physical harm imposes additional costs on the public that other harms do not, such as fear.

Nonviolent offenders should have the chance to avoid jail

The inverse of the above is that, if prison is for community protection then, as a general rule, wherever public safety can be reasonably assured, nonviolent offenders should, in the first instance, be punished by means other than prison.

There are a range of punishments that might be imposed on nonviolent offenders in lieu of prison, ranging from home detention to professional disqualification to fines.  These punishments can be levied to a degree sufficient to deter reoffending just as effectively as a prison sentence.

As noted above, almost half of Australian prisoners were charged with or convicted of nonviolent offences.  Many of these individuals do not pose a physical threat to the community and as such they may not need to be isolated through imprisonment.  Recognising that they may not need to be in prison would be the boldest reform we could make.

The expansion of home detention laws in South Australia are a good example of what can be done in this area.  In May 2016, the South Australian parliament passed the Statutes Amendment (Home Detention) Bill 2015, enabling judges to permit low-risk, nonviolent offenders to serve their sentences in home detention and some suitable prisoners to be released into home detention earlier in their sentences.

Reform direction 1 Extend the use of alternative punishments like fines and home detention for nonviolent, low-risk offenders.

Justifying additional punishment for recidivism

It has been argued that stronger punishments for recidivists are a form of double punishment.  Mirko Bagaric has written that:

Each time a person is sentenced for an offence, they should be principally sentenced for that offence only, not for what they have done in the past — they have already been punished for this.  To do otherwise involves either punishing a person for their supposed "character" (which is far too nebulous a commodity to justify inflicting pain on a person through the process of criminal sanctions) or double punishment. ⁽¹¹⁶⁾

Applying this principle in conjunction with the argumentation above would mean that repeated nonviolent offending would not justify incarceration.  This is certainly a counterintuitive result;  we would expect that punishments would get harsher as a pattern of offending emerges.

The question of whether prior offending should be an aggravating factor in sentencing has been debated for more than a century.  It is beyond the scope of this paper to attempt to resolve this question.  We do not need to know whether longer prison sentences for recidivists than for first-time offenders are justified;  we need only to note why in the case of recidivists prison might be preferred to other punishments which we might apply to first-time offenders.

As we have seen, Australian prisons are not very effective in preventing reoffending by released inmates.  Australian performance in this respect mimics the experience of other jurisdictions. ⁽¹¹⁷⁾  This suggests that prisons are not effective at rehabilitating criminals.  The high proportion of Australian prisoners who have been imprisoned before also suggests that prison does not work especially well as a specific deterrent.  If it is possible that the public's need for retribution can be satisfied through alternative punishments, this leaves only community safety as a possible justification for imprisoning recidivists.

A pattern of antisocial behaviour can be taken as an indication that an offender is not responding to punishment.  We can say that the recidivist has shown through his repeated offending that the only way the community will be rid of his criminality is to put him in prison.  Prisons may not prevent future offending by released inmates but prisoners cannot further harm the community for as long as they are imprisoned.  This is sufficient reason to prefer prison to alternative punishments in the case of recidivists.

Community safety comprises both physical safety and freedom from criminality.  Where the latter interest is infringed upon by an offender who has shown that he won't respond to other punishments, as in the case of recidivists, then prison is a proportionate response.

The use of strict liability should be limited

Traditionally, the criminal law has required that a guilty act (actus reus) be accompanied by a guilty mind (mens rea).  There are however a great many regulatory offences that have eliminated the mental element, such that an offender can be held accountable for an act even if he did not intend to break the law.  To choose just one example, in Australia you can be sentenced to up to seven years in prison for damaging a protected wetland, whether or not you knew the wetland was protected.

Strict liability regulatory offences can impose prison sentences on individuals who have demonstrated no intent to cause harm.  It is difficult to see incarceration for these offenders as necessary for the protection of the community and its norms.

In the United States, legal reformers have begun to push for the restoration of mens rea in the criminal law.  Reformers have proposed a law that would make all federal laws in that country that are silent on intent to be held to imply a requirement of a knowing state of mind, judged objectively with a "reasonable person" test. ⁽¹¹⁸⁾

This effort has met with some resistance.  Its opponents argue that it would make it much more difficult to prosecute those whose conduct adversely affects the environment, financial systems, public health.  However, with the expanding number of criminal offences in the US (the exact amount is unknown but may exceed 5000 at the federal level alone) ⁽¹¹⁹⁾ it is impossible to know the exact range of conduct that is now considered criminal, and therefore difficult to say that everyone in breach of these laws should be punished, much less punished with incarceration.  Nonetheless, opponents of mens rea reform have threatened to walk away from bipartisan criminal justice reform over the issue, seeing it as unjustly favoring corporate interests. ⁽¹²⁰⁾

The exact contribution that strict liability makes to over-criminalisation in Australia, and thus to the growing culture of incarceration, requires further study.  However, it is clear that in order to address the rising incarceration rate, we must be more consistent in our application of the principles of punishment, paying closer attention to the real purpose of prison and the circumstances in which it is unnecessary.  This implies that mens rea reform should be considered as part of broader criminal justice reform efforts.

Reform direction 2 Restore the requirement of mens rea for regulatory criminal offences.  Where strict liability is imposed, alternative punishments to prison should be applied, provided the offender has not demonstrated a propensity for violence or anti-social behaviour.

FURTHER IMPLICATIONS OF PROPORTIONALITY FOR NONVIOLENT OFFENDERS

Deterrence is only a subsidiary consideration in punishment

Proportionality implies that deterrence plays only a subsidiary role in determining the appropriate level of punishment for a criminal.  If punishment is guided by the harm done by the offender, then future acts, either by the offender himself or by other individuals, do not seem to be relevant.

This result can be avoided if we construe the harm done by the offender to include not just the damage done to the victim but to the societal norm which the law expresses.  We can then see how deterrence might justify a stiffer punishment.  But given that a single crime is unlikely to do much damage to a societal norm, the amount that might be added to a sentence in the name of deterrence is small.

In practice, this implies that victimless crimes such as drug use should only carry slight penalties, and certainly not prison sentences.

Reform direction 3 Do not punish victimless crimes with incarceration.

White-collar crime should not be treated exceptionally

The above arguments about how proportionality is to be assessed and about the limits of deterrence are often disregarded in the case of so-called white-collar crime.

The term "white-collar crime" was invented by Edwin Sutherland for expressly political purposes. ⁽¹²¹⁾  He believed that the upper classes of society committed just as much crime as the lower classes but were punished and stigmatised much less.  White-collar crime is therefore founded on the idea that special attention needs to be given to a particular group, whose behaviour needs to be constrained by exceptional legal means.

Today, white-collar crime is usually taken to refer to corporate crimes, typically where there is a breach of trust.  It includes such offences as:

• Insider trading
• Mishandling of trust accounts
• Regulatory infractions
• Director liability

These crimes tend to attract very heavy penalties. ⁽¹²²⁾  Because white-collar offenders are typically nonviolent and generally don't require rehabilitation, their incarceration is normally justified by the public's desire for retribution and the need to deter similar acts.  Given that retribution can be gotten with different punishments, the weakness of general deterrence as a justification for incarceration, and the lack of risk posed to the community by these offenders, there is good reason to consider incarceration for white-collar crime to be disproportionate and unnecessary.

The strength of this proposition is obscured by the powerful stigma that activists have successfully attached to this class of offenders.  In practice, this stigma is the main reason for the heavy custodial penalties that these crimes attract.  For example, white collar crime is also one of the reasons that mens rea reform is so controversial.  Opponents of that reform are particularly concerned that it would hinder the prosecution of corporate crime. ⁽¹²³⁾  That is, they explicitly argue that the state needs to be able to prosecute these crimes more easily than others.  This is indicative of the general inconsistency with which white-collar crime is considered.

The criminal justice system should have incentives for restitution

Of the alternatives to prison, restitution is the only punishment that explicitly puts the needs of the victim first;  before inquiring into any other interests of justice, where possible we ought to incentivise making the victim whole.

This argument is strengthened by the observation that incarceration actually requires the victims of crime to pay to punish the victim as their taxes are spent on incarceration.  And given the host of issues prisoners face upon release, such as poor health and employment outcomes, there is likely to be a greater impost on the victim through publicly funded health care and unemployment payments.

Some people argue that fines and restitution amount to letting rich people buy their way out of prison.  But fines can help the state recover the costs of justice, and restitution is of benefit to the victim.  The concern that people with the means to pay these penalties will avoid jail is offset by these benefits.  It has been argued that the possibility of avoiding prison operates as an incentive to offenders to make restitution where otherwise they would not. ⁽¹²⁴⁾  Moreover, the state could use alternative collection methods, such as garnishing wages and reducing benefits.  Doing so would allow courts to impose fines and restitutions orders on a wider range of offenders.

Undoing the harm caused by the offending is a good first step towards proportionality.  Additional punishment for the inconvenience to the victim, the cost to the state, and to cause pain to the offender will likely still be necessary, but this does not necessarily imply a prison sentence. ⁽¹²⁵⁾

Reform direction 4 Allow offenders to make restitution to their victims and take this into account in sentencing.  Broaden the applicability of fines and restitution by enabling alternative collection mechanisms, such as garnishing wages and reducing government benefits.

Strengthening the police could be effective justice reinvestment

Gary Becker initiated a long line of research which applied rational choice theory to why crime occurs, and how to achieve cost-effective reductions to crime.  The basic framework is that crime is a function of costs and benefits.  Criminal justice policy can raise the costs of crime in two ways:  raising the probability of being arrested, and increasing the penalty of crime upon arrest (i.e., the length of a jail sentence).

Under certain assumptions, an implication of this framework is that an almost infinite combination of probabilities of arrest and severity of punishments can be made to achieve a given crime rate.  Becker suggested that resources could be saved by spending less on detection and arrest, compensated for by longer prison sentences to leave the overall expected cost — and therefore the supply of crime — unchanged.

This framework continues to form the foundation of analysis of criminal justice.  But a key underlying assumption of Becker's original paper can be challenged:  that the average criminal has similar preference for risk and future discounting as the average person in society more generally.  In fact, there is some evidence to suggest that the average prisoner discounts the future heavier than does the average person (see chapter 3 for a fuller discussion of this).

This implies that increasing the expected cost of crime through the probability of arrest, rather than increasing sentence lengths, could more cost-effective than vice-versa. ⁽¹²⁶⁾  And Alex Tabarrok has argued that it follows from this observation that punishment for crime should be "quick, clear, and consistent". ⁽¹²⁷⁾

This argument has one clear practical consequence.  It is common to talk of justice reinvestment: ⁽¹²⁸⁾ the idea that expenditures on prisons can be more profitably deployed to address the root causes of crime.  What the evidence suggests is that a particularly effective reinvestment would be in policing.

Note however that this is not an argument for spending more money policing matters that do not rightly fall within the purview of the criminal law.  That is, this observation is limited by the earlier observation that the current bounds of the criminal law are far too expansive.  Moreover, it can be expected that tightening those bounds would make resources dedicated to policing more effective.

Reform direction 5 Redirect resources saved from incarcerating fewer criminals — particularly low risk, nonviolent criminals — to the police, strengthening their capacity to deter and detain criminals.

A related point is that policing of people on bail or parole should also focus on detecting more breaches and imposing weaker punishments more frequently.  The consequences of breaching parole, in particular, are often severe, with the parolee returned to prison to serve out the remainder of his sentence.  It has been argued that this severity leads to parole officers and judges trying to avoid punishing breaches until a pattern of breaches emerges.  This is perceived by parolees as arbitrary:  some breaches go unpunished while others result in prison.  The alternative is to punish all breaches with milder punishments, such as detention in holding jails or fines. ⁽¹²⁹⁾

As we have seen, a large contributor to the increase in Australia's prison population is constituted by people on remand, and parole has become more difficult to obtain.  Changing the conditions of bail and parole may make them more accessible to more offenders and more acceptable to the courts.

Reform direction 6 Investigate amending bail and parole laws to impose certain but mild consequences on all breaches.

---

5. CONCLUSION

Australia puts a lot of people in prison.  The incarceration rate is higher now that at any point since just after federation.  In the last 40 years the number of people in Australian prisons has increased by more than 300 percent.  We have proportionally more people in prison than almost all comparable countries, and the number is increasing every year.

A rising incarceration rate is not a problem in and of itself.  There are a number of criminals who need to be separated from the community.  Incarceration only becomes a problem if it is overused — that is, if the public is paying to lock up individuals who could safely and more productively be punished in different ways.  In Australia, along with the rising incarceration rate and prison population, we see large numbers of nonviolent offenders being imprisoned, high levels of recidivism, and a persistent level of crime.

The price of this approach is substantial.  Almost $4 billion is spent on prisons each year.  And then there are the downstream effects:  unproductive citizens, broken families, and repeat offending.  On all standards — public safety, economics, and morality — the rapid rise of incarceration in Australia demands investigation.

Unfortunately, one of the main contributors to this trend is simply that more individuals are choosing crime more frequently.  Research in recent decades has yielded real insight into how the criminal mind works, by focusing on criminality as an option that may be adopted by individuals who perceive, for whatever reason, that it is in their interests.  In short, crime is a choice.

The people in Australia's prisons tend to share a particular set of characteristics.  They have less formal education than their peers and they have been employed less in their lives.  Many have been among the country's long-term unemployed.  Along with this, prisoners are more likely to use illicit drugs.  They are also more likely to be Indigenous Australians.

These correlations can easily be mistaken for causes.  But while these factors may contribute to the reasoning behind criminal activity, they are not definitive.  Criminals, like everyone else, determine their preferences and act accordingly, which is why they can be held to account by the criminal justice system.  What this suggests is that to reduce crime, we need to make it less attractive.  In this respect, reducing crime is more than simply a matter for criminal justice reform, and likely involves reforms in workplace relations and the education system that promote individual agency and personal responsibility.

Nonetheless, there have been some criminal justice policy choices made by Australian governments that favour prison over alternative punishments and approaches.  Bail has become more difficult to get, and as a result Australia's remand population has ballooned.  Similarly, parole is being granted less frequently and judges are handing down longer sentences.  Australians and their governments are both, in their own ways, choosing prison at higher rates than in the past.

We can begin to address this change by thinking more seriously about what prisons are really for and in which circumstances it is necessary or desirable to put people in them.  Prisons are primarily for the protection of the community.  This implies that the prison population should be constituted of violent and antisocial offenders.  Yet 46 percent of prisoners are incarcerated for nonviolent crimes, at an approximate annual cost of $1.8 billion.  This in turn raises the question of proportionality:  whether prison is a punishment that fits all of the crimes to which it is applied.  Where the threat to the community is low, alternatives to prison should be preferred, both in order to respect the moral force of proportionality and to ensure that the limited resources of the criminal justice system are deployed effectively.

It follows too from these principles that the criminal law should not sprawl into domains traditionally governed by the civil law.  In cases where the remedies available to civil law courts would suffice, the criminal law is not needed.  Relatedly, since the criminal law's aim is to protect society from particular individuals, and since prison is the punishment uniquely suited to providing such protection, it makes sense that those individuals should have demonstrated some ill intent before the community isolates them.  Restoring the requirement of mens rea to many regulatory offences would signal the importance of intent to the criminal law, better capturing the moral argument for punishment, which rests on the free choice of the criminal.

And because crime is a choice, we know that deterrence plays a role in criminal justice.  It cannot, however, be a reason to forget about proportionality in sentencing.  Such a trade-off is not necessary anyway.  We know that criminals respond more to the chance of being caught than they do to the harshness of the punishment that awaits them if they are apprehended.  The best way to deter crime is to make it a less attractive choice by increasing the chance of being caught.  This means that the savings we make by being more judicious in our use of prisons can be usefully reinvested in the police force.

None of this is to say that nonviolent criminals should not be punished.  And it is certainly not to suggest leniency for the violent.  As this report very clearly shows, however, Australia is spending a fortune on imprisoning some people to little or no benefit.

This report is a suggested beginning for a reform process that is badly needed.  Society requires standards, and we should never shirk from enforcing our laws.  But for safer streets and more prosperous lives, we need to think carefully about how we impose the law on those who would choose to break it.

---

BIBLIOGRAPHY

Government reports

Adult Parole Board of Victoria (2015), Annual Reports 2005-6 and 2014-15

Australian Bureau of Statistics (2001), Year Book Australia 2001, Table C8.11

Australian Bureau of Statistics (2011), Australia and New Zealand Standard Offence Classification, Category 1234.0

Australian Bureau of Statistics (2010), Australian Social Trends Category 4102.0 March 2010.

Australian Bureau of Statistics (2015a), Australian Demographic Statistics December 2015 Category 3101.0

Australian Bureau of Statistics (2015b), Corrective Services, Category 4512.0 December Quarter 2015

Australian Bureau of Statistics (2015c), Criminal Courts Australia, Category 4513.0

Australian Bureau of Statistics (2015d), Prisoners in Australia 2015, Category 4517.0

Australian Bureau of Statistics (2015e), Recorded Crime — Offenders 2014-15, Category 4519.0

Australian Bureau of Statistics (2015f), Crime Victimisation, Australia 2014-15, Category 4530.0

Australian Bureau of Statistics (2015g), Labour Force Survey, Category 6202.0, July 2015

Australian Securities and Investments Commission (2014), Report 387: Penalties for corporate wrongdoing, March 2014

New South Wales Health Department (2009), New South Wales Inmate Health Survey 2009

Productivity Commission (2016), Annual Report on Government Services 2016, Vol. C

Senate Legal and Constitutional Affairs References Committee (2013), Value of a justice reinvestment approach to criminal justice in Australia, June 2013

Smith, P., Goggin, C., & Gendreau, P. (2002), The effects of prison sentences and intermediate sanctions on recidivism: General effects and individual differences, User Report 2002-01, Solicitor General Canada.

Smith, RG et al (2014), Counting the cost of crime in Australia: A 2011 estimate, Australian Institute of Criminology

United Kingdom Ministry of Justice (2012a), Accommodation, homelessness and reoffending of prisoners: Results from the Surveying Prisoner Crime Reduction (SPCR) Survey

United Kingdom Ministry of Justice (2012b), Prisoners' childhood and family backgrounds

United States Department of Justice (1988), Survey of Youth in Custody 1987, September 1988

Victorian Ombudsman (2015), Investigation into Rehabilitation and Reintegration of Prisoners in Victoria, September 2015

Western Australia Parole Board (2006) Annual Report 2006

Western Australia Prisoners Review Board (2015) Annual Report 2014-15.

Wisconsin Department of Health and Social Services (1994), Division of Youth Services, Family Status of Delinquents in Juvenile Correctional Facilities in Wisconsin, April 1994

Monographs and reports

American Bar Foundation, "The Social Costs of Incarceration", Researching Law Vol. 21 No. 3 Summer 2010

Australian Institute of Health and Welfare (2015), The Health of Australian Prisoners

Bennett, Dr. William J. (1994) Index of Leading Cultural Indicators, Broadway Publishers

Dotson, Greg and Cassady, Alison (2016), "Three ways congressional mens rea proposals could allow white collar criminals to escape prosecution", Center for American Progress, 11 March 2016

Fagan, Patrick (1995), The Real Root Causes of Violent Crime: The Breakdown of Marriage, Family and the Community, Heritage Foundation, March 1995

Foster, Helly and Hagan, John (2009), The Mass Incarceration of Parents in America: Issues of Race/Ethnicity, Collateral Damage to Children, and Prisoner Reentry (2009). The Annals of the American Academy of Political and Social Science May 2009 623:179-194

Griffith Institute for Social Research (2008), Literacy Unbarred: Investigating the Literacy and Numeracy Levels of Prisoners Entering Queensland Correctional Centres

Hudson, Sarah (2013), Panacea to Prison? Justice reinvestment in Indigenous Communities, Centre for Independent Studies

Johnson, Rucker C. (2009) Ever-Increasing Levels of Parental Incarceration and the Consequences for Children. In Do Prisons Make Us Safer? The Benefits and Costs of the Prison Boom, ed. Steven Raphael and Michael Stoll, 177–206. New York: Russell Sage Foundation

Kevin, Maria (2013), Drug Use in the Inmate Population — prevalence, nature and context, Corrections Services NSW, Research Publication No. 52, June 2013

National Research Council (2014), The Growth of Incarceration in the United States: Exploring Causes and Consequences National Academies Press 2014

The Pew Charity Trusts (2010), Collateral Costs: Incarceration's Effect on Economic Mobility, The Pew Charitable Trusts

Villasenor, John (2015) "Over-criminalization and mens rea reform: a primer", Brookings Institution, 22 December 2015

Walmsley, Roy (2016), World Prison Population List 11th Edition, Institute of Criminal Policy Research

Weatherburn, Dr Don (2014) Arresting Incarceration — Pathways out of Indigenous Imprisonment, Aboriginal Studies Press 2014

Weatherburn, Don and Fitzgerald, Jacqueline (2015), The impact of the NSW Bail Act (2013) on trends in bail and remand in New South Wales, NSW Bureau of Crime Statistics and Research, Issue Paper No. 106, August 2015

Articles

Bagaric, Mirko and Alexander, Theo (2014) "A rational approach to sentencing white-collar offenders in Australia", Adelaide Law Review, No. 34, 2014

Baker, Jr., John S. (2014), "The Sociological Origins of White-Collar Crime", Heritage Foundation Legal Memorandum No. 14, 4 October 2004

Baldry et al (2006), "Ex-Prisoners, Homelessness and the State in Australia", The Australian and New Zealand Journal of Criminology, 39(1), pp. 20-33

Becker, G (1968), "Crime and Punishment: An Economic Approach", Journal of Political Economy, vol. 76, 1968

Cutcher et al (2014), "Poor health and social outcomes for ex-prisoners with a history of mental disorder: a longitudinal study", Australian and New Zealand Journal of Public Health, Vol 38 pp. 424–429

D'Amico, Daniel J. (2015) "Knowledge problems and proportionality", Criminal Justice Ethics Vol. 34, Issue 2 2015

Doyle et al. (2015), 'Alcohol and other drug use among Aboriginal and Torres Strait Islander and non-Aboriginal and Torres Strait Islander men entering prison in New South Wales' Health and Justice, Vol 3 2015

Fagan, Patrick (1995), The Real Root Causes of Violent Crime: The Breakdown of Marriage, Family and the Community, The Heritage Foundation, March 1995

Giles, Margaret and Le, Ahn Te (2007) "Prisoners' Labour Market History and Aspirations: A Focus on Western Australia" Economic Record Vol 83, No 260, pp. 31-45, March 2007

Lochner, Lance and Enrico Moretti, (2004), "The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports.", American Economic Review, 94(1), pp. 155-189

Raphael, Steven (2014) "How do we reduce incarceration rates while maintaining public safety?", Criminology and Public Policy, Vol 13, Issue 4, 2014 pp.579-597

"Punishment", Stanford Encyclopedia of Philosophy http://plato.stanford.edu/entries/punishment/ (accessed 16 August 2016)

Journalism

Bagaric, Mirko (2016) "Indigenous Australians and African Americans deserve a sentencing discount", The Guardian, 5 April 2016.

DeBonis, Mike (2016) "The issue that could keep Congress from passing criminal justice reform", Washington Post, 20 January 2016.

Hymowitz, Kay (2012), "The Real, Complex Connection Between Single-Parent Families and Crime", The Atlantic, December 2012

"Longer jail sentences do deter crime, but only up to a point", The Economist, 29 March 2016

McArdle, Megan (2015), "What's fair? Swift, automatic consequences", Bloomberg View, 9 September 2015

Alex Tabarrok (2015), "What was Gary Becker's biggest mistake", Marginal Revolution, 16 September 2015

Wilson, James Q. (1983), "Thinking about crime" The Atlantic Monthly, September 1983, Volume 252, Number 3, pages 72-88

---

ENDNOTES

1. Smith, RG et al (2014), Counting the cost of crime in Australia: A 2011 estimate, Australian Institute of Criminology p.IX

2. Ibid p. XIII

3. Ibid

4. Productivity Commission (2016), Annual Report on Government Services 2016, Vol. C, p. C.8

5. Ibid

6. Ibid

7. Ibid

8. Ibid

9. Ibid

10. Ibid, Table 8A.6

11. Ibid, Table 8A.7.

12. Ibid

13. Ibid

14. Ibid

15. Ibid

16. Ibid

17. Ibid

18. Ibid

19. Ibid

20. National Research Council (2014), The Growth of Incarceration in the United States: Exploring Causes and Consequences.

21. The Pew Charitable Trusts (2010), Collateral Costs: Incarceration's Effect on Economic Mobility

22. Ibid

23. Rucker C. Johnson (2009) Ever-Increasing Levels of Parental Incarceration and the Consequences for Children. In "Do Prisons Make Us Safer? The Benefits and Costs of the Prison Boom", ed. Steven Raphael and Michael Stoll, 177–206. New York: Russell Sage Foundation

24. Ibid

25. Helly Foster and John Hagan, The Mass Incarceration of Parents in America: Issues of Race/Ethnicity, Collateral Damage to Children, and Prisoner Reentry (2009). The Annals of the American Academy of Political and Social Science May 2009 623:179-194

26. National Research Council (2014), op. cit.

27. Australian Bureau of Statistics (2015d), Prisoners in Australia 2015, Category 4517.0

28. Ibid

29. Australian Bureau of Statistics (2015a), Australian Demographic Statistics December 2015, Category 3101.0

30. Australian Bureau of Statistics (2001), Year Book Australia 2001, Table C8.11

31. Roy Walmsley (2016), World Prison Population List 11th Edition, Institute of Criminal Policy Research Note: Some countries report incarceration rates per 100,000 of the population, rather than 100,000 of the adult population. For comparison purposes per 100,000 of the population is used.

32. Australian Bureau of Statistics (2015d), op. cit.

33. Ibid

34. Ibid

35. Ibid

36. Ibid

37. Ibid

38. Ibid

39. Australian Bureau of Statistics (2011), Australia and New Zealand Standard Offence Classification, Category 1234.0

40. Australian Bureau of Statistics (2015d), op. cit.

41. Ibid

42. Ibid, and Productivity Commission (2016) op. cit. Tables 8A.6 and 8A.7

43. Australian Institute of Health and Welfare (2015), The Health of Australian Prisoners

44. Griffith Institute for Social Research (2008), Literacy Unbarred: Investigating the Literacy and Numeracy Levels of Prisoners Entering Queensland Correctional Centres

45. Doyle et al. (2015), 'Alcohol and other drug use among Aboriginal and Torres Strait Islander and non-Aboriginal and Torres Strait Islander men entering prison in New South Wales'

46. New South Wales Health Department (2009), New South Wales Inmate Health Survey

47. United Kingdom Ministry of Justice (2012b), Prisoners' childhood and family backgrounds

48. Lance Lochner and Enrico Moretti, (2004), "The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports."

49. Australian Institute of Health and Welfare (2015), op. cit.

50. Margaret Giles and Ahn Te Le (2007) "Prisoners' Labour Market History and Aspirations: A Focus on Western Australia" Economic Record Vol 83, No 260, pp. 31-45, March 2007

51. Victorian Ombudsman (2015), Investigation into Rehabilitation and Reintegration of Prisoners in Victoria, September 2015

52. New South Wales Health Department (2009), New South Wales Inmate Health Survey

53. Australian Bureau of Statistics (2015g), Labour Force Survey, Category 6202.0, July 2015

54. Australian Institute of Health and Welfare (2015), op. cit

55. Cutcher et al (2014), "Poor health and social outcomes for ex-prisoners with a history of mental disorder: a longitudinal study" Australian and New Zealand Journal of Public Health, October 2014; 38(5):424-9

56. Baldry et al (2006), "Ex-Prisoners, Homelessness and the State in Australia" The Australian and New Zealand Journal of Criminology, Vol. 39, No. 1

57. United Kingdom Ministry of Justice (2012a), Accommodation, homelessness and reoffending of prisoners: Results from the Surveying Prisoner Crime Reduction (SPCR) Survey

58. United States Department of Justice (1988), Survey of Youth in Custody 1987, September 1988

59. Wisconsin Department of Health and Social Services (1994), Division of Youth Services, Family Status of Delinquents in Juvenile Correctional Facilities in Wisconsin, April 1994.

60. United Kingdom Ministry of Justice (2012b), op. cit.

61. William J. Bennett (1994), Index of Leading Cultural Indicators Broadway Publishers

62. Patrick F. Fagan (1995), The Real Root Causes of Violent Crime: The Breakdown of Marriage, Family and the Community Heritage Foundation

63. Kay Hymowitz (2012), "The Real, Complex Connection Between Single-Parent Families and Crime", The Atlantic, December 2012

64. Australian Institute of Health and Welfare (2015), op. cit.

65. Ibid

66. Maria Kevin (2013), Drug Use in the Inmate Population — prevalence, nature and context Corrective Services NSW, Research Publication No. 52 June 2013

67. Australian Bureau of Statistics (2015d), op. cit.

68. Ibid

69. Ibid

70. Ibid

71. Ibid

72. Ibid

73. Sara Hudson (2013), Panacea to Prison? Justice Reinvestment in Indigenous Communities The Centre for Independent Studies, p. 8

74. Ibid p. 9.

75. Australian Bureau of Statistics (2010), Australian Social Trends Category 4102.0 March 2010.

76. Hudson (2013), op. cit. p. 7

77. Don Weatherburn (2014), Arresting Incarceration — Pathways out of Indigenous Imprisonment, Aboriginal Studies Press p. 7

78. Ibid

79. Hudson (2013), op. cit.

80. Weatherburn (2014), op. cit. p. 3

81. Becker, G (1968), "Crime and Punishment: An Economic Approach", Journal of Political Economy,vol. 76, 1968, p. 9

82. James Q. Wilson (1983), "Thinking about crime" The Atlantic Monthly, September 1983, Volume 252, Number 3, pages 72-88

83. Although this would only affect the absolute numbers of crime reported. Provided errors in reporting (such as underreporting or misreporting) were similar across time, analysis of growth rates or relative changes to crime rates would not be affected.

84. Australian Bureau of Statistics (2001), op. cit.

85. Australian Bureau of Statistics (2015e), Recorded Crime — Offenders, 2014-15, Category 4519.0

86. Australian Bureau of Statistics (2015a), Australian Demographic Statistics, December 2015, Category 3101.0.

87. Australian Bureau of Statistics (2015a, 2015e)

88. Ibid

89. Australian Bureau of Statistics (2015e), op. cit.

90. Ibid

91. Australian Bureau of Statistics (2015e) op. cit.

92. Ibid

93. Australian Bureau of Statistics (2015f), Crime Victimisation, Australia 2014-15, Category 4530.0.

94. Ibid

95. Ibid

96. Australian Bureau of Statistics (2015d), op. cit.

97. NB: This data has been compiled from the annual reports in category 4517.0 over the years 2006-2015.

98. Productivity Commission (2016), op. cit. p. C.18.

99. Remand means being held in custody while awaiting trial.

100. Australian Bureau of Statistics (2015d) op. cit. NB: data taken from the 2006 and 2015 reports.

101. Ibid

102. Weatherburn, Don and Fitzgerald, Jacqueline (2015), The impact of the NSW Bail Act (2013) on trends in bail and remand in New South Wales - Issue paper no. 106, NSW Bureau of Crime Statistics and Research, Issue Paper No. 106, August 2015 p. 1.

103. Senate Legal and Constitutional Affairs References Committee (2013), Value of a justice reinvestment approach to criminal justice in Australia, p. 8

104. In the latest ABS reporting, there are no statistics for how defendants pleaded in Western Australia. This calculation is based on filling this gap in the data by ascribing to Western Australia the average of the guilty plea rates from the other states and territories. Western Australian data are included in the calculation of adjudications and guilty verdicts. Australian Bureau of Statistics (2015c), Criminal Courts Australia, Category 4513.0

105. Raphael (2014), "How do we reduce incarceration rates while maintaining public safety?" Criminology and Public Policy, Vol 13, Issue 4, 2014 p. 584.

106. Tasmania, Northern Territory, and the ACT were excluded because of small sample sizes. Note also that while the graph's maximum value is 300 percent, there was one value greater than this: the increase in the number of South Australian prisoners expected to serve more than 20 years increased by 1785 percent from 7 to 132. This was driven by a reduction in the number of life sentences and an overall increase in the expected time to serve.

107. The following statistics are drawn from: Australian Bureau of Statistics (2015d), op. cit. Annual reports in this category from 2006 and 2015.

108. "Community-based corrections orders are non-custodial orders served under the authority of adult corrective services agencies and include restricted movement, reparations (fine options and community service), supervision orders (parole, bail, sentenced probation) and post-sentence supervision orders." Australian Bureau of Statistics (2015b), Corrective Services, December Quarter 2015

109. David Tait and Karen Gelb, "The effectiveness of suspended sentences in reducing reoffending", Presentation to NSW Bureau of Criminal Statistics and Research, 18-19 February 2015.

110. Adult Parole Board of Victoria (2015), Annual Reports 2005-6 and 2014-15

111. Western Australia Parole Board Annual Report 2006; Western Australia Prisoners Review Board (2015) Annual Report 2014-15

112. NSW Parole Authority (2014) Annual Reports 2005 and 2014

113. Megan McArdle (2015), "What's fair? Swift, automatic consequences", Bloomberg View, 9 September 2015

114. Daniel J. D'Amico (2015) "Knowledge problems and proportionality", Criminal Justice Ethics Vol. 34, Issue 2 2015

115. See eg. "Punishment", Stanford Encyclopedia of Philosophy http://plato.stanford.edu/entries/punishment/ (accessed 16 August 2016)

116. Mirko Bagaric (2016) "Indigenous Australians and African Americans deserve a sentencing discount", The Guardian, 5 April 2016

117. For example, a 2002 meta-study by the Canadian Solicitor-General found that imprisonment slightly increases the risk of reoffending. P. Smith, C. Goggin, & P. Gendreau (2002), The effects of prison sentences and intermediate sanctions on recidivism: General effects and individual differences, User Report 2002-01, Solicitor General Canada.

118. John Villasenor (2015) "Over-criminalization and mens rea reform: a primer", Brookings Institution, 22 December 2015

119. Ibid

120. Mike DeBonis (2016) "The issue that could keep Congress from passing criminal justice reform", Washington Post, 20 January 2016

121. John S. Baker, Jr. (2014), "The Sociological Origins of White-Collar Crime", Heritage Foundation Legal Memorandum No. 14, 4 October 2004

122. For example, insider trading is punishable by up to 10 years in prison. ASIC notes that penalties for corporate crimes in Australia are generally similar to those in comparable jurisdictions. Australian Securities and Investments Commission (2014), Report 387: Penalties for corporate wrongdoing, March 2014

123. Greg Dotson and Alison Cassady (2016),"Three ways congressional mens rea proposals could allow white collar criminals to escape prosecution", Center for American Progress, 11 March 2016

124. Mirko Bagaric and Theo Alexander (2014), "A rational approach to sentencing white-collar offenders in Australia", Adelaide Law Review, No. 34 2014

125. That said, where restitution is made impossible by deliberate actions taken by the defendant to thwart recovery of ill-gotten gains, this may well indicate in those cases that alternatives to prison would be futile, and fail to measure up to the harm caused.

126. See, for example, "Longer jail sentences do deter crime, but only up to a point", The Economist, 29 March 2016.

127. Alex Tabarrok (2015), "What was Gary Becker's biggest mistake", Marginal Revolution, 16 September 2015

128. Senate Legal and Constitutional Affairs References Committee (2013) op. cit.

129. McArdle (2015) op. cit.