Payment by results and all that jazz
This is the sixth and final in a series of posts exploring the Ministry of Justice’s plans to re-design its Transforming Rehabilitation project. The MoJ says it wants our views on how best to re-design probation and asks 17 key questions in its consultation document, “Strengthening probation, building confidence”.
This week’s post examines the final, and perhaps most important, question asked by the MoJ – how to drive performance improvements.
Please do take issue with me and set out your views and thoughts in the comments section below.
Question 17: What should our key measures of success be for probation providers, and how can we effectively encourage the right focus on those outcomes and on the quality of services?
When commentators questioned Chris Grayling’s original privatisation of the probation system, his counter-argument was that the new private providers would be subject to payment by results based contracts; in short:
“If they don’t do the job, we won’t pay them.”
However, that didn’t prove to be the case; probation performance has been poor, but on the whole CRC providers have been paid; partly in acknowledgement that the original contracts were under-funded and that CRCs did not receive the amount o f business they had been promised.
The latest PbR probation performance figures, which cover the CRCs’ first year of operation were analysed for this blog by Jack Cattell of Get the Data. He concluded:
The macro trend across all CRCs was for fewer re-offenders but those that did were likely to commit more re-offences than previously.
This meant that the CRCs recorded a small overall increase in the number of reoffences compared to the contract baseline year of 2011.
However, over the course of TR, there has been an emerging academic and practitioner consensus (with which I wholeheartedly agree) that reoffending rates (both “binary” – did someone reoffend? – and “frequency” – how many offences did they commit?) are poor indicators of probation performance. Overlying crime trends (fewer crimes, more committed online and undetected) and local police priorities and diminishing resources are likely to be more powerful influencers of official reconviction data.
In common sense terms, this is plain to see from the fact that during the large scale upheaval and crash in morale which characterised the last two years of the former probation system and the first year of TR, reoffending rates still fell.
One of the main drivers of the current re-configuration of the probation system was the MoJ’s awareness not only that the current system is drastically under-funded, but that the contract conditions in coming years (where CRCs’ income would increasingly be dependent on their reoffending outcomes) were almost inevitably going to cause several providers to fail.
Changes in the payment mechanism
The MoJ is clear that it wants to change the way CRCs are funded and to devise new performance incentives. Interestingly, it says it wants to focus on key desistance factors such as housing, jobs and drug & alcohol treatment, even though all these are beyond the influence of CRCs are have all themselves experienced large cuts in government expenditure through the years of austerity.
One of the options the MoJ is considering is setting a “Guaranteed Maximum Price with Target Cost” (already a new acronym “GMPTC”). If this option is adopted, then bidders will submit a target cost and target price (including profit) for given volume bands of different activities. The MoJ will pay actual costs up to the maximum price with profit going up and down inversely to actual costs. One of the intriguing implications of this approach is that, in direct contrast to the original TR contracts, it requires open-book accounting and reviews and justifications of costs. The MoJ provided an illustration of the GMPTC approach in its engagement events:
To my mind, the GMPTC option is a clear indication of the difficulties facing the MoJ. My first reaction is that this is an exceedingly complex system, vulnerable to gaming, and certain to use up significant resources at HMPPS and providers which could be better spent on service delivery.
Conclusion
My initial purpose in this series of posts was to raise key issues and stimulate positive ideas about how to re-design the probation service for the better.
I have to say that I am disappointed at my inability to suggest positive ways forward.
I think I have a clear-headed view of the political realities which mean that a “mixed-economy” is inevitable (it’s unlikely that a Conservative Government would admit that a privatisation project has failed and Chris Grayling still sits at the Cabinet table). Nevertheless, I find it difficult to make positive, concrete suggestions to the well-acknowledged difficulties besetting the modern probation service.
For me, the root of all these problems is the design of a split service. Everyone affected by probation — courts, PCCs, partners, victims and offenders themselves — sees probation as a single service whose aim is to protect the public and help offenders desist from crime. The National Probation Service and Community Rehabilitation Companies don’t think like this, focusing only on their own remit and responsibilities.
This fragmentation is the root cause of many, indeed most, of the key questions raised by the MoJ be they sentencer confidence in community sentences, better through-the-gate work or common professional training.
It is for the elected government of the day to decide whether probation should be a public or private enterprise but for the life of me, I find it hard to see how the current two-tier model (which the current consultation has re-confirmed) will succeed.
Countdown to consultation deadline
Whatever your views on the next version of probation, do find the time to respond to the consultation which closes on 21 September. You can submit your response here.
3 Responses
The problem with GMPTC and most payment models is that the supplier has an incentive to deliver the minimum service specification at minimum cost (and therefore maximum profit). The result is a service of little value as service users soon work out that the objective is to achieve the tick list – not to provide meaningful help and a relationship of trust which is arguably what matters most. An alternative approach (like that adopted for the Secure Schools proposal) is for Government to say what their budget is (either in cash or unit cost terms) and for bidders to compete on the strength of their proposed solutions and on measures of quality. This works best when bidders are all not for profit. But the approach could allow bidders to specify how much of the price will be a return to investors and then compete on quality with what is left. The experience of TR is that notional savings from bidders offering lower prices in a competition prove illusory. Low price has meant low value and Government ends up having to put in a reasonable level of resourcing regardless.
Thanks Rod. I’m in agreement with you, it would be fairer to set a budget and say what can you do for this amount? Makes for a more level playing field, also makes it fairer to apply sanctions if providers over-promise. We shall have to see how the government respond to the views they have gathered in consultation.
How can you improve on perfection, probation is perfect already.