Key omissions risk undermining the government’s bold sentencing goals

David Gauke’s long-awaited Sentencing Review, commissioned last October to consider major sentencing reforms to address the prison capacity crisis, makes dozens of recommendations. Many have been accepted by the government – but some crucial exceptions seriously weaken the final reforms.

The government has announced some positive and bold changes to sentencing

There are three big changes at the core of the government’s approach:

1. Earned early release for most prisoners and abolishing automatic early release
2. Emphasising community sentences over prison sentences where possible, including restricting the use of short prison sentences under a year
3. Harsher restrictions and more intensive supervision for people after release from prison or when serving a community sentence.

There is a lot to support here. There is good evidence from international examples supporting earned early release for good behaviour in prisons, ¹⁵ O’Hear M, 2015, ‘Let the Good Time Roll: Early Release for Good Behavior in Prison’, Faculty Publications, Marquette University Law School. https://scholarship.law.marquette.edu/cgi/viewcontent.cgi?article=1655&context=facpub  and releasing a significant proportion of prisoners after a third of their sentence (the new minimum which must be served in prison) should help to address long-term sentence inflation and ease capacity pressures on prisons.

Similarly, lots of evidence shows community sentences – which help offenders maintain jobs, housing and family relationships – are more effective than short prison sentences at reducing reoffending, including for prolific offenders. ¹⁶ Ministry of Justice, 2025, Reducing reoffending: A synthesis of evidence on the effectiveness of interventions, https://assets.publishing.service.gov.uk/media/680101e3da5bb2fc4a681fcb/Final_PDF_Reducing_Reoffending_-_Evidence_Synthesis.pdf  They also tend to address underlying causes of offending like drug and alcohol use and anger management problems more effectively than most prison sentences. Gauke’s review estimated that restricting short custodial sentences to exceptional circumstances would save about 2,000 places – and this is a benefit that would likely be felt quickly after implementation. 

Perhaps most critically, the government has also put some money where its mouth is, promising up to a 45% increase in funding for the probation service by the end of the spending review period to fund their increased workload. 

By rejecting some key recommendations, the government risks undermining its ultimate goals

But there are some crucial gaps in the government’s approach. The sentencing review recommended that earned early release be paired with automatic release after half or two-thirds of most offenders’ sentences – the status quo before the emergency early release measures implemented over the last couple of years. Gauke estimated this approach would save more than 4,500 prison spaces. But the government has chosen to scrap automatic release completely, meaning offenders who don’t earn early release through good behaviour could spend their entire sentence in custody. 

There are some inherent drawbacks to this approach. If someone spends their whole sentence in prison then there are no mechanisms for offering them support or keeping an eye on them once they are back in the community. But more critically in the current situation, scrapping automatic release makes it hard to know what impact the new earned release proposal will have on the total prison population.

Many prisoners are currently automatically released after 40% of their sentence, and others halfway through. If a large proportion serve significantly longer, that could outweigh the benefits of releasing some after only a third of their sentence. The current chaotic state of prisons means rehabilitative courses and employment in prison are seriously oversubscribed, and some offenders are already unable to secure release because they haven’t been able to get places on the courses they need to demonstrate good behaviour and rehabilitation. Depending on the criteria for earned early release, this could pose a serious problem for actually reducing the prison population.

The same is true of proposals to ‘toughen’ community sentences and supervision of prison leavers. Positive relationships between probation officers and the people they supervise are key to effectively reducing reoffending and managing risk in the community, ¹⁹ Ministry of Justice, 2025, Reducing reoffending: A synthesis of evidence on the effectiveness of interventions, https://assets.publishing.service.gov.uk/media/680101e3da5bb2fc4a681fcb/Final_PDF_Reducing_Reoffending_-_Evidence_Synthesis.pdf ^{, 20} HM Inspectorate of Probation, ‘Relationship-centred services’, retrieved 23 May 2025, https://hmiprobation.justiceinspectorates.gov.uk/our-research/evidence-base-relationship-centred-services/   but these will be hard to build and maintain if licence conditions are seen as unnecessarily punitive. At the same time, harsher restrictions will lead to more breaches of conditions and stressing the punitive role of supervision is likely to encourage more recalls in an already risk-averse and overstretched probation service. Again, Gauke combined his proposals on tough community supervision with major changes to recall, which he estimated would cut the prison population by a further 2,300 – but the government has given no sign it will adopt this.

Take together, it is not clear how much the government’s proposals will reduce the prison population, but the likely maximum is 5,000-6,000 spaces – meaning further reforms or emergency releases will be needed before 2028.

Lots of strong, evidence-based proposals have been left on the cutting room floor

Some of the most interesting proposals in Gauke’s report haven’t been picked up at all by the government. One that we have previously recommended is increasing the length of sentences that can be suspended, meaning the offender is not sent to prison unless they commit further offences or break their sentence conditions. Most of these offenders are non-violent and, as with community sentences, there is good evidence to show that suspended sentences reduce reoffending. ²³ Ministry of Justice, 2025, Reducing reoffending: A synthesis of evidence on the effectiveness of interventions, https://assets.publishing.service.gov.uk/media/680101e3da5bb2fc4a681fcb/Final_PDF_Reducing_Reoffending_-_Evidence_Synthesis.pdf  The review estimated this would save 1,300 spaces, and it could also encourage more guilty pleas – saving precious capacity in the Crown Court. 

Deferred sentences and ‘problem-solving’ or ‘close supervision’ courts also fall in this category. These provide opportunities for offenders to receive support and demonstrate good behaviour and a commitment not to reoffending before being sentenced. This can mean they avoid a prison sentence altogether, achieving better outcomes while saving public money. ²⁴ Centre for Justice Innovation, 2015, Problem-solving courts: An evidence review, https://justiceinnovation.org/sites/default/files/media/documents/2019-03/problem-solving-courts-an-evidence-review.pdf  

It seems likely that these measures may have been victims of politics trumping policy. In its framing of its reforms, the government has been eager to stress its ‘tough’ approach to community sentences and keeping people in prison until they earn their release. But if that leads it away from measures that will actually free up prison capacity and reduce reoffending, it risks undermining its ultimate goals.