The most striking fact to emerge is the total number of people in jails around the nation reached a monthly average of nearly 31,000 in the June quarter, an increase of 4.1 per cent in 12 months. This is particularly disappointing as just three or four years ago the rate of increase was slowing down markedly; now it is moving in the opposite direction.

If one looks beyond the crude numbers it becomes clear that the increases have occurred in the most vulnerable groups of prisoners: Aboriginal and Torres Strait Islanders, remandees and females. Over this period indigenous prisoners increased a further 7.4 per cent, so that they now constitute almost exactly 30 per cent of the national total. This should cause us all to hang our heads in shame and ask ourselves how we allowed this disgraceful situation to develop.

Also over this period female prisoners increased 9.5 per cent, more than twice the increase of male prisoners. The actual numbers of female prisoners may still seem to be relatively small but their rate of increase per year is higher than any of the other vulnerable groups, except remandees, and therefore should be a matter of particular concern.

The number of unsentenced prisoners, or remandees, increased by 11.3 per cent, making them well over 25 per cent of all Australia's prisoners, the highest proportion ever recorded. About 40 years ago the proportion of prisoners who were remandees was 10 per cent.

The final major finding in this report is the fact that over the past 12 months the number of offenders serving non-custodial correctional orders actually decreased while the number behind bars went up. For many years there was a reasonably constant ratio of almost exactly two offenders serving sentences in the community for every one in custody. This has changed to a ratio of only 1.75 offenders on community-based orders for every one in custody. This change is completely in the wrong direction. The overuse of imprisonment, especially the imprisonment of suspected offenders or remandees, is to be found in virtually every state and territory. This not just a failure to provide the basics of justice to offenders and their victims, it is also an enormous waste of money.

The Productivity Commission has estimated it costs about $100,000 to keep a person in custody for one year. This suggests that nationally we spend well over $300 million a year on our prison systems. Nearly all of this money comes from the states and territories, not the Commonwealth, but it is still taxpayers' money. All of these changes to our jail systems have occurred during a time when the rates of reported crime are falling markedly around the country, and therefore the question must be asked: what is the justification for the increasing use of imprisonment when the need for prisons seems to reducing? That is a nice little conundrum that could be appropriately addressed by a graduate class of criminology students, even though I am a little doubtful about whether they would produce useful explanations.

The evidence cited to establish whether crime is increasing, decreasing or staying at about the same level is usually quite complex and difficult to master, but, fortuitously, two reports on this subject were released by the NSW Bureau of Research and Statistics in August and both are readily accessible and eminently readable.

The first, titled The great property crime drop: a regional analysis, examined the incidence of reported offences in NSW and other jurisdictions and found dramatic falls in nearly all crime categories between 2000 and 2012, especially in urban areas. All jurisdictions had falls in burglary, motor vehicle theft and other theft in this period.

The second report, The decline in robbery and theft: interstate comparisons, found that the national robbery rate had fallen 49.1 per cent between 2000 and 2009, but the NT and the ACT bucked this trend and experienced small increases over this time. This report also found that the falls in crime had been much more substantial in urban areas than in remote or regional areas.

In response to this array of facts and figures some readers may be tempted to suggest that the explanation of the conundrum of simultaneous decreasing crime and increasing use of imprisonment is obvious: fewer people are now committing offences because more potential offenders are incarcerated. As obvious as that conclusion might seem, it does not tally with many other facts.

We are all aware of the huge differences in imprisonment rates between jurisdictions but no one has yet been able to show that higher imprisonment rates are associated with lower crime rates. For many decades NSW has had an imprisonment rate close to twice as high as the rate in Victoria and yet crime is not lower in NSW. The NT, which has by far the highest imprisonment rate in Australia, has the highest crime rate, not the lowest as one might expect. Simple explanations may seem obvious but they are not always helpful.

David Biles is a consultant criminologist in Canberra.

biles@netspeed.com.au