Abstract
There are stark and enduring differences between the Australian States and Territories in their rates of imprisonment. The Australian Capital Territory, for example, has an imprisonment rate of 113/100,000 population. The Northern Territory imprisonment rate is more than 8 times the Australian Capital Territory rate, and 41% higher than the imprisonment rate in Louisiana; the US state with the highest imprisonment rate in that country. The disparity is not limited to the Australian Territories. The imprisonment rates of Western Australia, Queensland, South Australia, and New South Wales are between 1.5 and 2.4 times higher than the Victoria rate. The aim of the present article is to determine whether the lower rate of entry into Victorian prisons is due to less frequent use of imprisonment or a lower arrest/court appearance rate. We find that the principal reason is a lower court appearance rate, particularly in relation to traffic and vehicle regulatory offences, justice procedure and related offences, burglary/break and enter, and illicit drug offences.
Keywords
Introduction
After 20 years of falling crime rates (Weatherburn & Rahman, 2021), Australia now finds itself in the invidious position of having incarcerated a larger share of its adult population than at any point since the late 19th century (Leigh, 2020). Those unfamiliar with the evidence might think that the fall in crime is due to rising imprisonment rates, but that would be a mistake. Two rigorous studies of the effect of imprisonment on crime in Australia have been conducted over the last few years. The first, by Wan et al. (2012) concluded that a 10% increase in the likelihood of prison would reduce property crime by 1.2% and violent crime by 1.7%. The second, by Bun et al. (2020), concluded that a 10% increase in the risk of a prison sentence would reduce property crime by 0.02% and violent crime by 0.03%. Neither study found any significant effect of increasing sentence length. If the Wan et al. (2012) study is accepted, less than 6% of the decline in crime in Australia between 2001 and 2020 is due to the rise in imprisonment risk. If the Bun et al. (2020) study is accepted the contribution of prison is less than 1%.
The rise in Australian imprisonment rates over the last two decades has been the subject of considerable attention on the part of both scholars and Government (Bushnell, 2019; Productivity Commission, 2021; Tubex et al., 2015; Tubex, 2021; Weatherburn, 2018). But it is not the only feature of penal policy in Australia worthy of closer attention. As Tubex et al. (2015) point out, although State and Territory imprisonment rates have all risen since the 1980s, there are stark and enduring differences between them in their rates of imprisonment. The Australian Capital Territory (ACT) has an imprisonment rate of 113/100,000 population (Australian Bureau of Statistics, 2021a). The Northern Territory (NT) imprisonment rate (960.6/100,000) is more than 8 times the Australian Capital Territory (ACT) rate, and 41% higher than the imprisonment rate in Louisiana; the US state with the highest imprisonment rate in that country (Australian Bureau of Statistics, 2021a; World Population Review, 2021). The disparity is not limited to the Australian Territories, which have small populations. The imprisonment rates of WA, QLD, SA, and NSW are between 1.5 and 2.4 times higher than the VIC rate (Australian Bureau of Statistics, 2021a).
This article seeks to shed light on the question of why there are stark interjurisdictional differences in rates of Australian imprisonment. It is a question that has received far less scholarly attention than the question of why State and Territory imprisonment rates have risen. Indeed, the small number of studies that have been conducted have focused almost exclusively on the difference in imprisonment rates between NSW and VIC. Porritt et al. (1980) analyzed rates of prison reception and length of study in VIC and NSW and concluded that differences in the length of stay in prison explained the large differences in the rate of sentenced prisoners per 100,000 population in NSW and Victoria. Babb (1992), also focusing on the difference between NSW and VIC in imprisonment rates, found little difference in the proportion of convicted offenders imprisoned but a large difference in the rate of appearance in court in NSW. He concluded that the higher sentenced prisoner population rate in NSW reflected a higher court appearance rate rather than a greater proclivity on the part of NSW to imprison convicted offenders. Gallagher (1995) and Weatherburn et al. (2010) drew similar conclusions, although Weatherburn et al. (2010) noted that the percentage of convicted offenders imprisoned in NSW (7.5%) was slightly higher than VIC (5.4%), while the mean expected time to serve in custody in NSW (45.4 months) was slightly shorter than in VIC (45.4 months).
Only three national studies of interjurisdictional differences in imprisonment rates have been conducted. The first by Tubex et al. (2015) drew attention to differences between jurisdictions in sentencing and bail policies but did not actually demonstrate that these policies were responsible for the differences between jurisdictions in rates of imprisonment. The second, by Brignell and Donnelly (2015), compared sentencing practices across the States and Territories for five offences that were similarly defined. They found that NSW imprisoned the highest proportion of convicted offenders for each of the five offences, but they did not examine differences in rates of arrest/court appearance. The third, by the Productivity Commission (2021), addressed the question of why State and Territory imprisonment rates had grown rather than why they differ. Its analysis of the differential rates of growth could have shed some light on differences in State and Territory imprisonment rates but was hampered by the fact that there are no national data on the percentage of defendants refused bail, the average length of stay on remand, the proportion of those on remand who receive a prison sentence, or the average prison sentence length. The Commission had to “guestimate” these parameters and, judging from the results of the “performance test” the Commission conducted, the results of their analysis for some jurisdictions are open to question (Productivity Commission, 2021, Table B.1, p. 110).
The lack of comparable national data on bail and remand makes it difficult to give a complete explanation for interjurisdictional differences in rates of imprisonment. We can make some headway in understanding those differences, however, by focusing on the process responsible for generating differences in the sentence prisoner imprisonment rate. This is important because sentenced prisoners account for between 55% and 70% of all adult inmates in Australian prisons (Australian Bureau of Statistics, 2021a). In this article two key drivers of the sentenced prisoner imprisonment rate are examined: the rate at which people are arrested and brought before the criminal courts (hereafter referred to as the court appearance rate), and the fraction of those appearing in court who are given a full-time custodial penalty (hereafter referred to as the imprisonment fraction). To anticipate our conclusion, we find that the primary driver of interjurisdictional differences in the rate at which sentenced prisoners to flow into the prison system lies in differences in the court appearance rate rather than differences in the imprisonment fraction. These differences in court appearance rates are particularly noteworthy in the case of justice procedure offences, burglary/break and enter, and illicit drug offences. This suggests that differences between the States and Territories in crime and policing policy are major contributors to differences in their rates of imprisonment.
In the next section of this article, we explain our approach to judging the relative importance of the court appearance rate and the imprisonment fraction in shaping the rate at which sentenced prisoners to enter prison. To keep the analysis within reasonable bounds, however, we do not include the Australian Capital Territory or the Northern Territory. Their populations are too small to obtain reliable data on the percentage imprisoned broken down by offence type. Further, rather than compare each jurisdiction to every other jurisdiction (a herculean task), we compare every State to VIC: the jurisdiction with the lowest imprisonment rate.
Method
To determine the relative importance of the court appearance rate and the imprisonment fraction in shaping the rate at which sentenced prisoners to enter prison we exploit the fact that the prison sentence rate for any jurisdiction can be obtained by multiplying its court appearance rate by its imprisonment fraction. Since we can do this with any combination of court appearance rates and imprisonment fractions, we can determine what would happen to the prison sentence rate in each of the other Australian states if they had either the same court appearance rate as Victoria (VIC) or the same imprisonment fraction. The approach is the same as that adopted by Weatherburn (2018) in analyzing the relative importance of crime, policing, and penal policy in shaping the growth in Australian imprisonment rates between 2002 and 2016. In essence, we contrast the effect on the expected number of sentenced prisoners by: (a) Holding the court appearance rate constant and allowing the fraction imprisoned to vary: (b) Holding the fraction imprisoned constant and allowing the court appearance rate to vary.
Having determined which component of the prison sentence rate (viz. the court appearance rate or the imprisonment fraction) plays the dominant role we can then examine the offence profiles of those coming before the criminal courts in different states to determine which offences are the primary drivers of differences between VIC and other states in their prison sentence rates. Finally, because rates of appearance in some offence categories (e.g. offensive behavior, justice procedure offences) are more clearly policy driven than others (e.g. homicide, break and enter, robbery), we can gain some insight into the role that policy plays in accounting for the low Victorian imprisonment rate.
Before we begin, a few definitions are in order. In what follows we define the term “court appearance rate” as the number of finalized court appearances involving adults in a jurisdiction divided by the population aged over 17 in that jurisdiction and multiplied by 100,000. We define the term “imprisonment fraction” as the number of adults given a prison sentence divided by the number of finalized adult court appearances in that jurisdiction. 1 Finally, we define the term “prison sentence rate” as the number of adults given a prison sentence in a jurisdiction, divided by the population aged over 17 in that jurisdiction and multiplied by 100,000. All data in what follows are drawn from the Australian Bureau of Statistics Prisoners in Australia data files (Australian Bureau of Statistics, 2021b). 2 In most cases, we provide data at the second level in the Australian and New Zealand Standard Offence Classification. On occasion, however, we present data at the third level (Australian Bureau of Statistics, 2011). Population data on Indigenous status are drawn from the Australian Bureau of Statistics (2022b).
Results: Court appearance rate or fraction imprisoned?
Columns 2, 3, and 4 of Table 1 show, respectively, the court appearance rate, the prison sentence rate, and the imprisonment fraction. It is obvious that the court appearance rate, the prison sentence rate and the imprisonment fraction vary greatly across the Australian State. All, however, have a higher court appearance rate than VIC. The effect of this on the prison sentence rate is muted in some cases by a lower imprisonment fraction but exacerbated in others by a higher imprisonment fraction.
Court appearance, fraction imprisoned and prison sentence rates by State.
NSW: New South Wales, VIC: Victoria, QLD: Queensland, SA: South Australia, WA: Western Australia, TAS: Tasmania.
Table 2 shows what would happen to the prison sentence rate in each jurisdiction if its courts imprisoned the same fraction of those appearing in court as VIC but had its own court appearance rate (middle column) or if they had their own imprisonment fraction but the same court appearance rate as VIC (third column).
Counterfactual scenarios.
NSW: New South Wales, VIC: Victoria, QLD: Queensland, SA: South Australia, WA: Western Australia, TAS: Tasmania.
The effects are surprising. Only two jurisdictions would see a reduction in their prison sentence rates if their imprisonment risk was the same as that which prevails in VIC. In TAS, the prison sentence rate would decline by 9.4%. In QLD, the prison sentence rate would decline by 10.7%. In the remaining jurisdictions, the prison sentence rate would actually rise—in some cases (e.g. SA and WA) by a large margin. If WA imprisoned the same fraction of those appearing in court as VIC, for example, it would experience a 57% increase in its prison sentence rate. The net effect on the Australian prison sentence rate if all states shared the VIC imprisonment fraction would be close to zero (0.2%).
The third column in Table 2 shows what would happen if all jurisdictions shared the VIC court appearance rate. The results are very different. In this situation, the prison sentence rate of all jurisdictions would fall—in some cases by a substantial margin. If WA shared the court appearance rate of VIC, its prison sentence rate would fall by 56%. In QLD the prison sentence rate would fall by 46%. The smallest decline would occur in SA, but even its prison sentence rate would fall by 19%. VIC has a lower prison sentence rate than other states, not because it is less likely to imprison offenders, but because the rate of appearance in court in VIC is substantially lower (per head of population) than in other states. The question we now turn to is what accounts for the fact that VIC has a lower court appearance rate than any other Australian state.
Given the gross over-representation of Indigenous Australians in custody, an obvious point of departure is the relative rate at which Indigenous and non-Indigenous Australians appear in court. Only four jurisdictions publish data on the Indigenous status of adults appearing in court. They are NSW, QLD, SA, and the NT. The court appearance rates for Indigenous Australians in these jurisdictions exceed the court appearance rates for non-Indigenous Australians by a factor of 6.7 in QLD, 7.9 in NSW, 12.9 in the NT, and 15.4 in SA (Australian Bureau of Statistics, 2022c). Unfortunately, the Australian Bureau of Statistics does not provide information on the relative proportions of Indigenous and non-Indigenous adult Australians were given a prison sentence. 3 It is therefore impossible to determine what the rate of entry into prison in each jurisdiction would have been if the fraction of Indigenous defendants they imprison were the same as that in VIC.
The only information we have that might shed light on the reason for the low court appearance rate in VIC, relative to other States, is the offence profile of those appearing in court. In the next section, therefore we compare the offence profile of those appearing in court in VIC, with the offence profiles of those appearing in court in the other states.
Results: What offences drive the difference in court appearance rates
NSW versus VIC
Figure 1 shows the ratio of the NSW court appearance rate to the VIC court appearance rate for each of the major offence categories. The offences to the right of the ordinate (vertical axis) have higher rates in NSW than in VIC. The reverse is true for offences where the ratio is less than one (i.e. those to the left of the vertical line). These features have the same interpretation in Figures 2 to 6.
The NSW court appearance rate is higher than the VIC court appearance rate in all offence categories other than dangerous/negligent acts; abduction/harassment and related offences; miscellaneous offences; and justice procedure and related offences. The rates of a court appearance in NSW are more than twice those of VIC in the categories of acts intended to cause injury, public order offences and illicit drug offences.
New South Wales/Victoria court appearance rate ratio by offence type.
QLD versus VIC
Figure 2 shows the percentage difference between QLD and VIC in court appearance rates for each offence. As in NSW, court appearance rates in QLD are higher than in VIC in almost all categories of crime; the exceptions being abduction/harassment and related offences: acts intended to cause injury; and miscellaneous offences. The court appearance rates in QLD are more than twice as high as the VIC rates in the categories of theft and related offences; burglary/break and enter; justice procedure and related offences, illicit drug offences and public order offences. The most striking differences, however, are those associated with illicit drugs and public order. In these cases, the QLD/VIC court appearance ratios are 4.6 and 7.3, respectively.
Queensland/Victoria court appearance rate ratio by offence type.
SA versus VIC
Figure 3 shows the percentage difference between SA and VIC in court appearance rates by type of offence.
South Australia/Victoria court appearance rate ratio by offence type.
In contrast to the other States just reviewed, the court appearance rates in SA are mostly either lower or close to the VIC rate. The four exceptions are weapons and explosives offences; illicit drug offences, burglary/break and enter; and public order offences. The court appearance rates are more than twice as high in the case of burglary/break and enter; and almost 3.5 times as high in the case of public order offences.
Figure 4 shows the contrast between VIC and WA in court appearance rates, broken down by offence type.
Western Australia/Victoria court appearance rate ratio by offence type.
WA has higher court appearance rates in all categories of a crime other than homicide and related offences. The standout offences: (i.e. those where the WA court appearance rate is more than twice the VIC rate) are robbery, extortion, and related offences; property damage/environmental pollution; dangerous, negligent acts; theft and related offences; traffic and vehicle regulatory offences, burglary/ break and enter; justice procedure and related offences, illicit drug offences; and public order offences. The QLD court appearance rates in the last two offences are, respectively, 5.8 and 8.5 times higher than the corresponding rates in VIC.
Figure 5 shows the offence-specific contrasts between VIC and TAS in court appearance rates.
Tasmania/Victoria court appearance rate ratio by offence type.
As in NSW, QLD, and WA, TAS has higher court appearance rates for nearly all offences. This is particularly true in relation to acts intended to cause injury (the TAS court appearance rate is 2.19 times the VIC rate); burglary/break and enter (the TAS court appearance rate is 2.75 times the VIC rate); and public order offences (the TAS rate is 4.87 times the VIC rate). A check of the differences at the lowest level of ANZSOC (see Appendix Table A1) for acts intended to cause injury reveals that this is largely because the two jurisdictions differ greatly in their court appearance rates for assault (TAS: 394/100,000; VIC: 174/100,000 pop). In the case of public order offences, the primary driver is the difference in court appearance rates for offensive contact (TAS: 35/100,000; VIC: 9.3/100,000).
Finally, to give an overview of the differences in court appearance rates between VIC and the remaining states, we examine the ratio of the Australian court appearance rate to the VIC court appearance rate for each offence (see Figure 6).
Australia/Victoria court appearance rate ratio by offence type.
Inspection of Figure 6 shows that the most noteworthy differences in court appearance rates between VIC and Australia as a whole (i.e. those where the Australian rates are more than 1.5 times higher than the VIC rates) are traffic and vehicle regulatory offences (1.6 times higher), justice procedure and related offences (1.64 times higher), burglary/break and enter (1.75 times higher), illicit drug offences 2.77 times higher) and public order offences (3.79 times higher).
Results: Crime or policing policy?
Figures 1 to 6 raise the question of whether the manifest differences in court appearance rates are the product of policy or crime (or both) This is a difficult question to answer; partly because public information on the policies being pursued by police and other enforcement agencies is very scarce and partly because reliable measures of crime are simply not available for many offences (e.g., child sexual assault; drug trafficking; traffic and vehicle regulatory offences; dangerous/negligent acts). 4 What follows, is therefore avoidably somewhat speculative in nature.
We begin with the difference in relation to traffic and vehicle regulatory offences. The most common offences in this category (see Appendix Table A1) are driver’s license offences (most of which involve driving without a license or driving when the person's license has been disqualified or suspended) and regulatory driving offences (most of which involve driving under the influence of alcohol or drugs). As there are no independent data on the prevalence of driving while disqualified/suspended or driving under the influence of alcohol or drugs, it is impossible to tell whether the court appearance rate difference between VIC and other states arises because these offences are less prevalent in VIC or because enforcement of traffic and vehicle regulatory offences is pursued less vigorously in VIC than in other states. VIC has the second lowest rate of road fatalities in Australia (the ACT has the lowest) (Australian Government, 2022), but this could signal (a) more responsible driving (b) a lower rate of driving without a license or (c) a lower risk of detection. The first two of these suggest lower rates of offending. The third suggests a difference in policing.
The difference between VIC and Australia in rates of a court appearance for public order offences, though very large and almost certainly the result of interjurisdictional differences in policing policy, is not a major contributor to differences in prison sentence rates. This is because less than 1% of those convicted of a public order offence receive a prison sentence (see Appendix Table A2).
The lower rate of a court appearance in VIC for burglary/break and enter is probably attributable, at least in part, to a lower rate of break and enter in VIC. QLD, SA, WA, and TAS all have substantially higher recorded rates of unlawful entry with intent than VIC and all have higher rates of appearance in court for this offence. The combined figure for break-in and an attempted break-in in the national crime victim survey reflects the pattern for recorded crime. It is worth noting in this connection that VIC also has one of the lowest rates of a motor vehicle and “other theft.” Theft offences also involve a relatively high risk of imprisonment (see Appendix Table A2).
The low court appearance rate for illicit drug offences in VIC admits to several interpretations. Most of those entering prison for this category of the offence has been convicted of importing, trafficking, manufacturing, or cultivating prohibited drugs rather than using or possessing them (see Appendix Table A2). This means that jurisdictional differences in the prevalence of illicit drug use shed little light on the reason for the lower prison sentence rate for drug offences in VIC. Data on drug seizures and arrests for drug importation, cultivation, manufacture, and supply provide the only evidence we have on the prevalence of drug trafficking The Australian Crime Intelligence Commission publishes data annually on drug seizures and drug providers (viz., suppliers). According to their latest Illicit Drug Data Report (Australian Crime Intelligence Commission, 2022), the number of arrests for drug supply in NSW in 2019/20 was more than 2.5 times higher than in VIC. Drug seizures were also far higher in NSW than in VIC, both in terms of number and weight. The weight (in grams) of ATS seized by State Police in NSW in 2019/20, for example, was more than 40 times that in VIC. Similar weight differentials between NSW and VIC were recorded for cannabis (2.7 times higher in NSW), heroin (10.5 higher in NSW), and cocaine (19.9 higher in NSW). These differences might be thought to reflect greater investment in drug law enforcement by the NSW Police, but they are reflected in the seizures made by the Australian Federal Police, who arguably have no reason to focus their efforts anywhere but on the major sources of drug trafficking. It would seem likely, therefore, that at least part of the reason for the high rate of drug offences outside VIC is that much of the country's drug trafficking occurs outside VIC, most notably in NSW.
Acts intended to cause injury and sexual assault are two other offence categories that account for a high proportion of imprisonments. VIC does not publish data on recorded rates of assault; however, the latest crime victim survey indicates that the prevalence of assault in VIC is lower than in South Australia, Western Australia, and Tasmania (Australian Bureau of Statistics, 2022a). VIC also has much lower recorded rates of sexual assault than NSW, QLD, and WA. In NSW, for example, the recorded rate of sexual assault is 54.5% higher than that of VIC (135.4/100,000 pop. vs. 87.6/100,000 pop.). In WA, the recorded rate of sexual assault is 20% higher than in VIC (104.9/100,000 pop. vs. 87.6/100,000 pop (Australian Bureau of Statistics, 2021c). These higher rates could be due to a higher rate of reporting sexual assault, but it is not obvious why the citizens of VIC would be substantially less willing to report sexual assault to police than victims of this offence in other states. The national crime victim survey, unfortunately, does not publish a jurisdictional breakdown of the prevalence of sexual assault or the willingness to report the offence to the police. It is therefore unclear how much of the difference between VIC and other States in court appearance rates for assault and sexual assault is driven by policy differences, differences in willingness to report, differences in rates of discovery and differences in the actual incidence of these offences.
There are some offences where the role of policy is more transparent. The offence category Justice Procedure Offences provides the clearest example of this. The dominant subcategory here involves breaching a violence or non-violence restraining order (see Appendix Table A1). There is no reason to believe that Victorian residents are less likely to breach these orders than residents of other states, however, the court appearance rate for justice procedure offences in VIC, is 50% lower than in the other states combined. The three most likely explanations for this are (a) VIC issues fewer violence and non-violence restraining orders (per capita) than other States (perhaps because the prevalence of domestic violence is lower or victims are less likely to report it), (b) VIC police devote fewer resources to enforcing compliance with violence and non-violence restraining orders (or accord such enforcement a lower priority), or (c) the conditions placed on violence and non-violence restraining orders by VIC courts are easier to comply with than those placed on similar orders in other states. Further research is necessary to determine which of (a) to (c).
Conclusion
The primary aim of the article was to determine how much of the difference in the rate of flow into prison is due to higher arrest rates and how much is due to a greater proclivity to send convicted offenders to prison. A secondary aim was to comment on the relative contributions of crime and penal policy to the effects observed. The evidence presented in this article indicates that the primary reason for the higher prison sentence rate in VIC, compared with other States, is that the court appearance rate for offences likely to result in imprisonment is much lower in VIC than in other states. The lower court appearance rates in VIC are particularly noteworthy in the case of justice procedure offences, burglary/break and enter, and illicit drug offences. These offences are all associated with a high risk of imprisonment. They are not, however, the only types of crime in relation to which the VIC court appearance rate is lower. The VIC rate is also lower in the case of sexual assault and related offences; robbery, extortion, and related offences; acts intended to cause injury; and theft and related offences.
Some of these differences are likely attributable to lower crime rates in VIC but certainly not all. Victoria, for example, certainly appears to arrest fewer people (per capita) for breaching restraining orders and for public order offences than other States. Although few people are imprisoned for public order offences, a conviction for these offences lengthens a defendant's criminal record, thereby increasing the risk of imprisonment for other offences at a later point in time. One possible explanation for the lower court appearance rate in VIC is that the percentage of Victorian residents who identify as Indigenous (0.8%) is much lower in VIC than in all other States (Australian Bureau of Statistics, 2022b). As we saw earlier, Indigenous Australians are arrested at much higher rates than non-Indigenous Australians in NSW, QLD, SA, and NT (Australian Bureau of Statistics, 2021c). Unfortunately, VIC does not publish data on the Indigenous status of those arrested by police or those appearing in court. This makes it impossible to determine to what extent its low prison sentence rate is attributable to the small percentage of its population who are Indigenous. This problem could be addressed if the Australian Bureau of Statistics provided separate tabulations for Indigenous juveniles and Indigenous adults given full-time custodial sanctions.
Earlier we noted that factors in the rate of arrest/court appearance and the proportion of convicted offenders given a prison sentence are not the only determinants of a jurisdiction's imprisonment rate. The rate of entry on remand is another important factor. The VIC remand rate per head of population (62/100,000 pop.) is lower than that of all other States except Tasmania (Australian Bureau of Statistics, 2021a). If time already spent on remand is any guide, however, this is not because prisoners spend much less time on remand in VIC than in other states. Unsentenced prisoners in VIC spend on average 5.9 months on remand, compared with 6 months on average across Australia as a whole (Australian Bureau of Statistics, 2021a). The lower remand rate might be because VIC courts are less likely than courts in other States to refuse or revoke bail. However, as with the sentenced prisoner population, the reason for the lower VIC remand rate may lie in the fact that fewer people per head of the population are being arrested and prosecuted in Victorian criminal courts.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
