April 26th saw the MoJ publish the reoffending rates relating to CRCs and 11 out of 21 of them hit PbR targets. It looks like the MoJ are going to have to find another way of cooking the books in order to bail them out:-
This publication provides the final results for the third offender cohort, April to June 2016. Final results for October to December 2015 and January to March 2016 were published in October 2017 and January 2018 respectively. The results are based on a one-year proven reoffending measure for adult offenders being managed in the community in England and Wales by Community Rehabilitation Companies (CRCs) under Payment by Results (PbR) arrangements, and by the National Probation Service (NPS). This is the measure against which CRCs will be assessed for the PbR element of the Transforming Rehabilitation reforms.
The following two reoffending measures will be used to assess CRC and NPS performance:
- the binary rate (proportion of offenders who reoffend)
- the frequency rate (the average number of reoffences per reoffender)
The performance of each CRC in reducing reoffending, on both the binary and frequency measures, will be assessed against the baseline year 20112 . Furthermore, the binary rate for each CRC is subject to adjustment for changes in the case mix of offenders being supervised, using the Offender Group Reconviction Scale (OGRS), version 4/G3, to allow performance to be assessed against the baseline.
In addition to the final results, the publication also includes interim proven reoffending statistics for the July to September 2016, October to December 2016, January to March 2017 and April to June 2017 offender cohorts. The Ministry of Justice (MoJ) included proposals in its July 2015 consultation, to provide early insights into CRC and NPS performance in reducing reoffending. These have been produced since October 2016 and are based on a reoffending-to-date measure.
This bulletin was developed in response to the consultation and will provide final results and interim proven reoffending statistics for the following offender cohorts:
- PbR eligible offenders managed by CRCs
- Offenders managed by the NPS who meet the same eligibility criteria as those in the CRC PbR cohorts
It is important to note that, while interim results provide useful and timely information, they will only give a broad indication of progress and, therefore, care should be taken when interpreting them. The measure against which CRCs will be assessed for PbR will be based on the final results, compared against a 2011 baseline.
Final results for the July to September 2016, October to December 2016, January to March 2017 and April to June 2017 CRC offender cohorts will be published in July 2018, October 2018, January 2019 and April 2019 respectively.
2. Final reoffending rates for CRC and NPS – April to June 2016 quarterly cohort
Final results for the October to December 2015 offender cohort and the January to March 2016 offender cohort were published in October 2017 and January 2018 respectively. This publication contains the third set of final results, for the April to June 2016 quarterly offender cohort. They are based on a cohort of offenders being managed in the community under Payment by Results arrangements by CRCs following probation reforms. All offenders have been subject to the full one-year follow-up period and the additional six-month waiting period as detailed in the guide to proven reoffending statistics. Comparisons of performance between different CRCs and previous cohorts can now be made by comparing the adjusted binary rates.
1. There have been statistically significant reductions in the adjusted binary reoffending rate for 11 of the 21 CRCs in the April to June 2016 cohort when compared to the 2011 baseline reoffending rates.
2. For one CRC, Warwickshire & West Mercia, there has been a statistically significant increase in the binary reoffending rate.
1. We cannot say which CRCs are meeting their frequency targets from a single quarterly cohort, as frequency targets are based on annual cohorts. The next annual cohort will be the 2016/17 annual cohort, due to be published in January 2019. It will include eligible offenders that began their probation between April 2016 and March 2017.
2. This is the second publication to feature interim results for all the quarterly cohorts that will make up the 2016/17 annual cohort but it is too early to assess the progress of CRCs against their frequency targets.
3. It remains the case that comparisons cannot be made between CRCs and NPS due to the difference in the nature of offenders being managed.
Comparison to previous final results:
The final results for April to June 2016 are the third set of final results to be published. Figure 1 compares this most recent set of final results to the previous sets of final results.
1. Five CRCs have been in the non-payment region in all three final cohorts; seven have been in the payment region in all three final cohorts.
2. Warwickshire & West Mercia CRC are in the deduction region for a second time after the October to December 2015 cohort.
3. South Yorkshire CRC were in the deduction region in October to December 2015 but moved into the non-payment region for the January to March 2016 and April to June 2016 cohorts.
3. Interim results
3. Interim results
The interim results provide a broad indication of progress. They should be more reliable for earlier cohorts, where the follow-up and waiting periods are closer to having fully elapsed. The figures presented in the tables should be interpreted with caution for two main reasons:
1. They are interim estimates which are based on provisional data and a reoffending-to-date measure, rather than a measure with defined follow-up and waiting periods.
2. The binary results have not been adjusted for the mix of offenders in the cohort. Before performance is assessed against the 2011 baseline, the final set of binary results for each cohort will be adjusted for changes in the case mix of offenders being supervised using the OGRS4/G.
Furthermore, the number of offenders identified in the measurable cohort may still change and, hence, change the characteristics of the cohort. This could impact both the binary rate and the frequency rate. It, therefore, remains the case that no conclusions can be drawn until final results are published.
1. With the exception of the Bedfordshire, Northamptonshire, Cambridgeshire & Hertfordshire October to December 2016 quarterly offender cohort, the average OGRS4/G scores for all four offender cohorts show all CRCs are managing offenders that are less likely to reoffend compared to the baseline year of 2011. It should, therefore, be expected that the OGRS adjusted binary rates will be higher than the unadjusted binary rates for most CRCs.
2. No CRC in any cohort is currently in the deduction region based on their current interim figures, although it is possible this will change when the binary reoffending rates are OGRS adjusted.
3. South Yorkshire; Staffordshire & West Midlands and Warwickshire & West Mercia CRCs are currently in the non-payment region based on their respective unadjusted binary rates for the July to September 2016 offender cohort.
4. Bedfordshire, Northamptonshire, Cambridgeshire & Hertfordshire; Durham Tees Valley; South Yorkshire; Staffordshire & West Midlands and West Yorkshire CRCs are currently in the nonpayment region based on their respective unadjusted binary rates for the October to December 2016 offender cohort.