Payment by results pilot at HMP Doncaster
August 5th, 2015 by Dan Branigan.
Our colleagues at Criminal Justice Alliance have recently compiled a report, outlined below, which investigates the ‘payment by results’ model tested at HMP Doncaster in 2011.
Mini-briefing: Payment by results pilot at HMP Doncaster
- In 2011 a pilot began at HMP Doncaster, a privately-operated prison managed by Serco, to test the ability of a ‘payment by results’ model to reduce the reconviction rate of individuals leaving prison. The pilot, started in October 2011 and delivered with Catch22, was originally for four years but cut short to three by mutual agreement. All offenders leaving HMP Doncaster were eligible for the services, although the programme was not compulsory.
- The cohort 2 re-conviction rate for offenders, just published by the Ministry of Justice, was only 3.3 percentage points lower than the 2009 baseline year and therefore Serco must now reimburse 10 per cent of the core contract value for the year 2012-2013. During the same period there was a reduction of 0.4 per cent in the national reconviction rate. (The first cohort had successfully met the target by reducing the reconviction rate by 5.7 per cent.)
- The results will cause concern for organisations involved in the recent Transforming Rehabilitation changes, and now contracted under Community Rehabilitation Company arrangements. The terms of those contracts are not identical to the Doncaster pilot but it is worrying that the target outcomes were not achieved within a headline programme.
- There is one conspicuous success in the pilot. When only analysing offenders serving sentences of less than 12 months, the group that now for the first time have access to supervision under the changes within Transforming Rehabilitation, there was a 6.3 per cent reduction in their reconviction rate, compared to a 1 per cent fall nationally. (Using this figure alone Serco would have met their target.)
- Unfortunately, worse results were produced for those serving sentences of more than 12 months. Their reconviction rate increased by 3.7 per cent, while falling nationally by 0.4 per cent. It appears that resources may have shifted to the sub-12 month group to the detriment of other offenders.
- Finally, under the pilot the reoffending rate fell by 4 per cent, compared to 0.8 per cent nationally. At first this would appear to be positive. Unfortunately, despite this drop, the frequency of offending among offenders actually increased by 0.26 offences per individual. The ‘binary’ measure, either there was or wasn’t a reconviction, meant that frequency or severity of reoffending wasn’t taken into consideration and therefore support was withdrawn from individuals who had reoffended within a cohort year.
The pilot ‘payment by results’ contract required that if the re-conviction rate in each year was not at least five percentage points lower than the 2009 baseline of 58 per cent, the Ministry of Justice would reclaim ten per cent of the core contract value. For every additional percentage point about five per cent Serco would have been entitled to additional payments. The value of the contract, and thus of the penalty to be paid by Serco, has not been published.