A multimillion-dollar system designed to stop prisoners reoffending after their release appears to have failed.
The "integrated offender management" (IOM) system was described by the Corrections Department in 2000 as "the biggest single initiative the department has undertaken to reduce reoffending".
But it has failed to move the two-year reconviction rate of released prisoners from 55 per cent - the rate when the system started.
The number of released prisoners being re-imprisoned within two years has increased, from 34 per cent when the system started in 2000-01 to 39 per cent in the year to June, although this may partly reflect tougher sentencing laws.
Canterbury University criminologist Dr Greg Newbold told an Australasian sociology conference in Auckland yesterday that the system had become "a large and expensive failure".
"It's another wreck on the scrapheap of abandoned fads of criminal rehabilitation," he said.
The system is a computerised way of managing all offenders.
"The idea is that all offenders are classified on reception into security and treatment categories, a sentence plan is developed and people suitable for treatment are directed into programmes which are consistent with their criminogenic needs," Dr Newbold said.
"It's like a hospital - you go to hospital, you get diagnosed, you get treated.
"The expectations were extremely high. The significant cost of establishing and operating IOM, about $40 million a year, was expected to be offset by the increasing effectiveness of the interventions.
"They predicted up to a 25 per cent reduction in annual inmate intakes."
But Dr Newbold, who served seven and a half years in Paremoremo Prison in the 1970s, said he predicted in 2002 that the system would not work because most prisoners, unlike patients, did not want to be "cured".
"They like the way they are. They want to give up going to jail. They don't want to give up offending."
He said the system was also beyond the capability of prison staff.
"Just as I predicted in 2002, assessment information about offenders was mostly missing because the screws couldn't be bothered putting it in, or they made mistakes in a very complex computerised system. Prison officers are not trained computer operators," he said.
"Staff performed less than half the required assessment tasks necessary, and approximately a third of the computerised judgments were overridden by staff or judges because they didn't agree with them."
Instead of reducing prisoner numbers, the system was overwhelmed by a big jump in numbers caused by tighter sentencing and parole laws in 2002.
As a result, the Ombudsmen reported in 2005 that only 26 per cent of prisoners had any proper employment, and high-security inmates spent most of the day locked in their cells.
Dr Newbold said he still believed prisons should provide rehabilitative services, such as education and drug and alcohol counselling, to make prisoners "better parents and neighbours" after release.
But the programmes should be only for those who freely chose to take them, and early parole should not be used as an inducement.
"Identify the small number of inmates who want to do something, give them programmes, but give them no early release," he said.
"Do their lag. Their reward will come when they get out of jail and they don't go back."
But the Corrections Department's director of psychological services, David Riley, said international evidence showed that programmes could reduce reoffending if properly implemented.
Even in New Zealand, he said, programmes for sex offenders had reduced sexual reoffending, psychological treatment reduced reoffending by 13 per cent, a programme for disqualified drivers had cut re-offending "significantly" and from two intensive anti-violence programmes had shown "encouraging results".
"It's a new ball game now. We are recruiting people who have attributes such as a social science background or experience in occupational therapy, psychiatric nursing and education.
"They don't have normal prison officer duties. Their role is not custodial."
Released prisoners reoffending within 2 years
Reconvicted 55.1% Reimprisoned 34.0%
Reconvicted 55.1% Reimprisoned 34.6%
Reconvicted 55.2% Reimprisoned 37.2%
Reconvicted 55.4% Reimprisoned 38.8%
Sources: Corrections Department