It's only when crime doesn't pay that criminals will be deterred

SINCE 1999 Australia's prison population has increased by 50 per cent. And with high and likely rising recidivism rates, it seems set to continue growing.

As well as the social costs prison imposes, each day of imprisonment costs taxpayers close to $300.

With total spending on corrective services exceeding $3.5 billion annually, state and territory budgets are struggling to cope. Little wonder the cost-effectiveness of our criminal justice system has become increasingly controversial.

Now a landmark study by the NSW Bureau of Crime Statistics and Research on the effects of arrest and imprisonment on crime provides some important answers. Its results show increased law enforcement could cost-effectively reduce crime rates.

But it cannot be the complete solution, especially to the deeply troubling problem of indigenous offending.

The study finds a 10 per cent increase in the risk of arrest produces a 3 per cent reduction in violent crime, while a 10 per cent increase in the risk of imprisonment reduces violent crime by 2 per cent. The impacts are less pronounced for property crimes, with a 10 per cent increase in the risk of arrest causing a 1.5 per cent reduction in crimes, while a 10 per cent increase in imprisonment cuts crimes by about 1 per cent.

That difference between violent and property crime is not unexpected. Law enforcement affects the crime rate through two channels. First, the risk of arrest and jail deters potential offenders; second, by removing offenders from the community, imprisonment incapacitates them from committing further crimes.

But while jailing one violent criminal does not simply create a job opening for another, putting a burglar away leaves the heist he would have committed for a new recruit. That replenishment of the stock of offenders means incapacitation has less effect on property crimes than on crimes of violence.

At the same time, penalties are significantly lower for property crime, so any deterrent effect is also weaker. Indeed, data presented in the study implies that if you are planning to commit a burglary in NSW, the jail time you could expect to serve is less than a week.

This is because you stand an 11 per cent chance of being arrested and charged; if arrested and charged, you stand a 7.5 per cent chance of being imprisoned; and if imprisoned, the average sentence for property crimes is 8.7 months. As a result, the expected jail term (which is the product of those probabilities) barely exceeds 2 days. No surprise the study finds a 10 per cent increase in the duration of imprisonment, which would lengthen expected jail time by all of five hours, would not significantly reduce crime.

Do the study's results imply we should devote greater resources to law enforcement? No, say the NSW Greens. They point to the finding that rising incomes reduce crime more than arrests and imprisonment do, with a 10 per cent increase in household income reducing property crimes by 18.9 per cent and violent crime by 14.6 per cent in the long term. That, they say, means increased welfare and education spending would reduce crime more cost-effectively than ramped-up law enforcement.

In fact, the study's estimates imply the opposite. Increasing welfare and public education spending by 10 per cent would cost some $18bn. But going on the Australian Institute of Criminology's assessment of the social costs of crime, cutting all crime by 15 per cent would yield a social benefit of barely $5bn, leaving a $13bn net loss. In contrast, increasing imprisonments by 10 per cent would cost Australian taxpayers $350 million annually. But if it allowed a 2 per cent reduction in offending, especially in violent crime, it could lower the cost of crime by around $600m, and so be well worthwhile.

That greater law enforcement could reduce crime more efficiently than increased social outlays should be obvious. Criminal justice is much more tightly focused on crime than any other social service. Every child needs an education. Most low-income households receive social transfers. But few people become offenders, and even fewer repeatedly engage in violent crime. So the criminal justice system benefits from high target efficiency: its impact goes largely where the problem is, rather than splattering over where it is not.

But that does not mean greater deterrence is the entire answer. For the fact of the matter is that as the prisoner population has grown, it has become ever more skewed to indigenous Australians. In the 1950s, Aborigines accounted for some 10 per cent of prisoners in Western Australia and 6 per cent in Queensland; they now account for 40 per cent of WA's prisoner population and nearly one-third of Queensland's.

The proximate causes of those high incarceration rates are not hard to find: they lie in high rates of violent offending, persistent reoffending and frequent breaches of parole. One-quarter of indigenous offenders tried in NSW in 2005 had eight or more prior convictions; only 7 per cent of non-indigenous offenders did.

Those facts, however, are merely symptoms of the collapse of Aboriginal society since the 60s. And with Aboriginal incarceration rates now more than 10 times those for the population generally, imprisonment can feed on itself.

Arrest and incarceration, when widespread, lose their stigma. Instead, crime and punishment become rites of passage, as they have for many young Aborigines, closing off other life chances. And as communities suffer the strains of mass imprisonment, fragile social structures become even more dysfunctional.

But throwing in the towel on law enforcement would make no sense. Rather, incarceration rates tend to be greatest precisely where the police presence is weakest and deterrence least vigorously applied. And those are the communities that also bear the greatest costs not only from violence itself, but from the anguish it causes.

More intense local policing therefore needs to be part of the response. And until we find a way of addressing the underlying problems, mass imprisonment will inevitably have to absorb some of their consequences, protecting society from violent offenders.

Yet those are hardly solutions Australians can be proud of. Yes, the bureau's results confirm deterrence and incapacitation do work. Coercion is therefore every bit as necessary as improved social conditions. But it remains a stubborn measure of our failure, not of our success.

Sources and supporting material at http://blogs.theaustralian.news.com.au/henryergas/