15 February 2011

Comparative Incarceration Rates

Recent posts in this blog (eg here and here) have noted statistics on incarceration rates. A new Bulletin [PDF] from the NSW (BOCSAR) unpacks some statistics with questions about differences in imprisonment rates in NSW and Victoria.

BOCSAR's 'Why does NSW have a higher imprisonment rate than Victoria?' uses a descriptive analysis of national crime, court and prison data in commenting that -
The NSW imprisonment rate is about twice that of Victoria (204 per 100,000 population vs. 104 per 100,000 population). This fact is widely believed to indicate that sentencing policy and practice in NSW is much harsher than in Victoria. A higher imprisonment rate, however, does not necessarily indicate tougher sentencing. The difference between the two states in their imprisonment rates may, for example, arise from differences in the rate of arrest for serious crime, differences in their bail laws or differences in their parole policies.

The NSW court appearance rate is 26% higher than that in Victoria. The overall conviction rate in NSW is 85.7%, compared with 79.0% in Victoria. The overall percentage imprisoned is significantly higher in NSW (7.5%) than in Victoria (5.4%). The mean expected time to serve among prisoners dealt with by Victorian courts is slightly longer than the mean expected time to serve among prisoners dealt with by NSW courts. The NSW remand rate is approximately 2.5 times the Victorian remand rate.
BOCSAR attributes the higher NSW imprisonment rate to a higher rate of court appearance, a slightly higher conviction rate, a higher likelihood of imprisonment and a higher likelihood of remand in custody.

The bulletin's authors ask what explanation can be given for the higher court appearance rate in NSW, the higher proportion of defendants convicted, the higher proportion of convicted offenders imprisoned and the higher remand rate.

They suggest that -
The higher court appearance rate in NSW is likely to be due, at least in part, to higher rates of crime. In 2004, for example, (the last year in which comparable figures were published) the NSW recorded armed robbery rate was nearly 1.9 times that of Victoria. The higher rate of court appearance for drug offences is probably also at least partly a reflection of crime. NSW has a much higher rate of court appearance for importing illicit drugs. It also happens to be the port where the largest quantities of illicit drugs are seized. The much higher NSW court appearance rate for acts intended to cause injury is harder to explain. It would not be surprising if NSW, given its significantly larger Aboriginal population, had higher rates of assault. 21% of the NSW prison population is Indigenous, compared with 6% of the Victorian prison population. Assault is the most common offence for which Indigenous offenders in NSW are imprisoned. In 2004, (the last year in which comparable figures were published), NSW did have much higher recorded rates of assault. The Victorian police figures on assault, however, have been called into question by the Victorian Ombudsman. Survey figures, moreover, show no difference between NSW and Victoria in the prevalence of assault. The surveys conducted by the Australian Bureau of Statistics measure the prevalence rather than the incidence of assault. It is possible that NSW has a higher incidence of assault than Victoria. At this stage, however, it is impossible to say to what extent the higher NSW court appearance rate for acts intended to cause injury is attributable to higher assault rates, as opposed to differences in the way NSW and Victorian police and prosecutors respond to incidents of assault.

In other cases, the higher NSW court appearance rate is more likely to reflect differences between NSW and Victoria in policing or penal policy. NSW, for example, has double the number of people appearing in court for breaching apprehended violence orders (2,976 for NSW vs. 1,057 for Victoria). This difference is much too large to be plausibly attributed to a greater proclivity on the part of domestic violence offenders living in NSW to breach domestic violence orders. The more likely explanation is that the number of domestic violence orders issued in NSW is much higher than in Victoria and/or that police in NSW are more likely to take action in response to an alleged breach of an apprehended violence order. The same applies to the large difference between the two states in the rate of appearance in court for traffic/motor vehicle regulatory offences. Much of this difference stems from the fact that NSW has nearly four times as many people appearing in court for drink-driving offences. It is possible that NSW residents are nearly four times more likely to drink and drive than their Victorian counterparts but a more likely explanation for the high rate of drink-driving appearances is that levels of enforcement for drink-driving are higher in NSW than in Victoria.

There are several possible explanations for the higher proportion of defendants convicted in NSW. Juries in NSW may be more likely to convict defendants who plead not guilty than their Victorian counterparts. Likewise, NSW magistrates may be more likely to convict defendants who plead not guilty than Victorian magistrates. The proportion of defendants pleading guilty may be higher in NSW than in Victoria. Since the guilty plea rate varies from offence to offence, the difference in the percentage convicted may arise from differences in the offence profile of cases coming before the criminal courts. The higher percentage of convicted offenders given a prison sentence in NSW may also be due to several factors. It could, of course, reflect a greater proclivity on the part of NSW courts (regardless of offence and offender characteristics) to impose a custodial sanction. It is also possible, however, that NSW courts deal with a more serious population of offenders3 or that prosecutors in NSW are more likely to lay multiple charges.

Although this analysis has answered some questions, it raises many others. The NSW court appearance rate for acts intended to cause injury is more than double that of Victoria yet national survey data show little if any difference between NSW and Victoria in the prevalence of assaults. Is the difference in court appearance rates for this offence due to differences between NSW and Victoria in the incidence of assault or is it due to differences between the two states in the way they respond to assault? The percentage of convicted offenders sent to prison in NSW is 39%higher than in Victoria. Is the higher imprisonment rate attributable to differences between the two states in the profile of offenders coming before the court system or do NSW courts imprison offenders who, had they appeared in a Victorian court, would be given some kind of non-custodial sanction? NSW has a far higher remand rate than Victoria. Is this because NSW courts are less likely to grant bail at first instance, because police in NSW are more likely to take action in response to alleged breaches of bail or because NSW courts are more likely to revoke bail following evidence of a breach?

... The length of time spent in custody is only partly a function of the sentence imposed by the courts. For many offenders it is also shaped by the willingness of parole authorities to grant or revoke parole. Do NSW and Victoria differ in the willingness of parole authorities to grant or revoke parole? There is clearly a great deal more work to do before the differences between NSW and Victoria in their imprisonment rates are fully understood.