EARLY PRISONER RELEASES: HOW MANY BREACH THEIR CONDITIONS AND WHAT HAPPENS NEXT?
CATCH A THIEF UK | NATIONAL NEWS INVESTIGATION
Thousands of licence recalls are recorded across England and Wales each quarter. As the Government introduces a new early-release model, questions remain over electronic tagging, probation capacity, public protection and the information available to measure the policy's success.
Investigation 2 of 12 | 17 October 2026
The question Britain deserves answered
When a prisoner is released early, what happens if they fail to follow the rules?
It is a straightforward question, but answering it properly requires more than counting arrests, highlighting individual cases or quoting statistics without explaining what they mean.
A person may breach a licence condition without committing a new criminal offence. Someone may be recalled to prison because they have failed to comply with supervision requirements. An electronic tag may fail or not be fitted when expected. And a person accused of a further offence has not necessarily been convicted.
These distinctions matter because the public needs accurate information about the effectiveness of the early-release policy, not statistics that combine different events and imply they mean the same thing.
Following the first two release tranches under the new progression model in England and Wales on 1 and 13 October 2026, questions have been raised about electronic tagging, probation workloads and the response to breaches.
Catch a Thief UK has examined Ministry of Justice statistics, official Government guidance, independent analysis and recent reporting to establish what is known, what remains unconfirmed and what information the Government should publish.
Our central finding is this: official figures show substantial numbers of licence recalls across the wider prison system, but the sources reviewed do not establish a complete national total of breaches, recalls or reoffences specifically attributable to the October 2026 release tranches.
That gap matters. Without clearly defined figures for the people affected by the new policy, the public cannot fairly judge its results.
1. What counts as a breach?
A licence is the set of conditions a person must follow after being released from prison while still serving their sentence in the community.
Depending on the person's sentence and circumstances, conditions may include attending probation appointments, living at an approved address, complying with restrictions on movements or activities, and following electronic monitoring requirements.
Four outcomes need to be kept separate.
A licence breach
This occurs when someone fails to comply with a condition of their release. The circumstances and seriousness can vary. A missed appointment is not the same as an allegation of a serious new offence, although repeated or concerning non-compliance may prompt further action.
A recall to prison
Recall is a legal process through which someone released on licence is returned to custody under the applicable arrangements. It may follow non-compliance, concerns about risk or alleged further offending. A recall does not necessarily mean a new criminal offence has been proved.
Further offending
This concerns a person committing another offence after release. An arrest, allegation or charge is not the same as a conviction. Reliable reporting must identify which measure is being used and the period covered.
An electronic monitoring failure
A tag may not be fitted as expected, may malfunction or may fail to transmit a signal. That can create a monitoring problem, but it does not automatically establish that the person has breached a condition or committed an offence.
The distinctions are important for public safety and fairness. They also prevent misleading comparisons between official datasets.
2. The official figures: 13,193 licence recalls in three months
Ministry of Justice statistics for January to March 2026 recorded, 3,193 licence recalls in England and Wales, an increase of 31% compared with the same quarter in 2025.
This is a substantial number and deserves scrutiny. However, it is essential to understand what it represents.
The figures cover the wider prison and probation system. They pre-date the October 2026 progression model and therefore cannot be presented as the number of people released under that model who subsequently breached their conditions.
The Ministry has linked the wider recall trend to several factors, including changes to recall arrangements, earlier release schemes and other developments affecting the system. The statistics do not attribute the entire increase to a single policy.
The official breakdown of recorded reasons for recall also provides useful context:
- Around 73% involved non-compliance with licence conditions.
- Around 38% involved failure to keep in touch.
- Around 24% involved failure to reside.
- Around 22% involved a charge of further offending.
These categories overlap because a recall can have more than one recorded reason. They must not be added together as if each recall had only one cause.
Nor should a recall involving a charge of further offending be described as a conviction for a new crime.
The figures demonstrate that licence compliance and recall are already significant issues within the wider system. They do not tell us how many people released under the new October scheme will breach their conditions.
The question for the Government: How soon will comparable, cohort-specific figures be published so the public can assess the new policy separately from the wider recall system?
3. A further concern: thousands recalled but not returned to custody
The Ministry of Justice also recorded 3,137 people who had not returned to custody following recall at the end of June 2026.
This is a historical, system-wide figure covering people recalled over a period of years. It is not a count of people released under the October 2026 scheme.
The official data included 489 people originally serving sentences for violence against the person and 87 originally serving sentences for sexual offences.
Those figures need careful interpretation. The wider total includes people whose circumstances may differ significantly, including individuals believed to be dead or living abroad where this had not been formally confirmed.
It would therefore be inaccurate to suggest that all 3,137 people were actively evading authorities in England and Wales, or that all presented the same level of risk.
Nevertheless, the figures raise legitimate questions about how recall cases are tracked, how quickly people are located and returned when required, and how outstanding cases are prioritised.
They also reinforce the need to distinguish the historical recall system from the new release policy.
The question: What procedures ensure that people recalled to custody are located and returned as quickly as the circumstances and law require, and how will performance be measured?
4. What has been reported since the October releases?
On 13 October, The Guardian* reported concerns raised by MPs about electronic tagging following the second tranche of early releases.
The report said some people released early were not being tagged as expected. MPs questioned whether monitoring safeguards were being delivered as promised.
The justice minister acknowledged operational challenges, while stating that the decision about whether to tag an individual ultimately involved operational judgement.
Serco, the company responsible for electronic monitoring, disputed allegations that it had failed to meet its contractual requirements. Its position was that it was attempting to fit devices to those required to wear them and that relevant authorities were informed where fitting was not possible or conditions were breached.
The report also included concerns from Metropolitan Police Commissioner Sir Mark Rowley about what happened after people were arrested for alleged tagging-condition breaches. According to his account, more than 95% of the cases he described resulted in the person being bailed back into the community on the same conditions.
This is an attributed statement about the cases he described. It is not a national rate for all tagging breaches, all people released early or all reoffending.
The report raises serious questions about the coordination between monitoring providers, probation, police and the courts. However, it does not establish a complete national count of October-scheme breaches or prove that every person who was not tagged as expected committed an offence.
The question: When a tag is required, what happens if fitting is delayed, and who is responsible for assessing and managing any resulting risk?
5. Electronic monitoring: a wider problem that cannot be ignored
The concerns reported in October sit against a broader background of electronic-monitoring problems.
An Institute for Government analysis, reported by The Guardian on 12 October, found that more than 5,000 people who should have been electronically monitored were not tagged between January and April 2026. It also identified almost 1,600 cases in which tags failed to transmit a signal for at least two weeks.
These figures relate to the wider monitoring system, not exclusively to people released under the October progression model.
They should not be used as the number of early-release prisoners who were untagged. But they are relevant to the question of whether electronic monitoring is being delivered reliably across the system.
Electronic monitoring can support the enforcement of certain conditions, including curfews and location restrictions, depending on the equipment and arrangements in place. It cannot prevent every offence, replace professional risk assessment or guarantee that a person will comply.
A tag that is not fitted, is faulty or fails to transmit information may limit the usefulness of the monitoring arrangement. What matters is not only whether a technical problem occurs, but how quickly it is identified, who receives the alert and what action follows.
The public deserves clear information about how many people require monitoring, how many devices are fitted on time, how often failures occur and how unresolved cases are managed.
The question: Can the Government demonstrate that electronic monitoring is reliable enough to support the safeguards attached to early release?
6. What does the new release model require?
The Government's progression model changes the release point for certain eligible prisoners serving determinate sentences. It does not apply to every prisoner, and specific offences and sentence categories are excluded.
The Government's guidance describes a progression from custody into a period of intensive community supervision, followed by a final period on licence under the applicable arrangements.
The model places significant importance on the services responsible for supervising people after release.
Licence conditions may include requirements to attend appointments, live at a specified address or comply with monitoring and other restrictions. The conditions applied to an individual depend on the relevant legal framework and their circumstances.
The Government has also announced enhanced licence restrictions, including exclusion zones and restrictions involving certain locations, venues or events where appropriate.
But the existence of a condition is not the same as evidence that it is being monitored effectively. The public needs to know whether the arrangements are implemented on time, whether breaches are identified and whether appropriate action follows.
7. Probation under pressure: who manages the additional workload?
Probation staff supervise people serving community sentences and those released from prison on licence. Their responsibilities can include assessing risk, monitoring compliance, supporting rehabilitation and responding when concerns arise.
The Institute for Government's Performance Tracker 2026, published on 12 October, highlighted ongoing pressures on probation capacity. The new release model adds to the importance of having sufficient experienced staff, manageable caseloads and effective coordination between agencies.
The additional releases are not the only demand on probation. The service already manages people subject to community sentences, people released under existing arrangements and cases requiring different levels of supervision.
Probation union Napo has also raised concerns about pay and workloads. As at 17 October, its planned industrial-action ballot was due to begin on 19 October. A ballot is not a confirmed strike, and any eventual impact would depend on its outcome and the form of any action.
These concerns do not prove that the early-release policy has failed. They do underline why the Government should explain how the service will manage the workload and maintain appropriate safeguards.
The question: Is there sufficient probation capacity to supervise people safely, and what evidence will demonstrate that the necessary checks and interventions are happening?
8. What happens when someone breaches their licence?
The response depends on the nature of the breach, the person's circumstances, the applicable licence conditions and the assessment of risk.
Possible responses include further contact or enforcement by probation, action relating to electronic monitoring, referral to police where appropriate, or recall to custody under the applicable legal arrangements.
A missed appointment does not automatically produce the same response as a serious threat, prohibited contact or suspected new offence. Authorities must assess the circumstances and act proportionately while managing risks to the public and victims.
That does not mean breaches should be ignored. It means the response should reflect the nature of the concern and the legal powers available.
The public deserves to know how quickly breaches are identified, how decisions are made and whether agencies have enough capacity to respond when action is required.
Important questions include:
- How many breaches are detected each week?
- How quickly are serious breaches assessed?
- How many lead to recall, police action or another formal response?
- How many are associated with alleged further offending?
- What happens when an electronic monitoring alert cannot be resolved?
- Are there delays caused by staffing shortages, technical problems or administrative backlogs?
Without those figures, it is difficult to judge whether the safeguards are working consistently.
9. How many October early-release prisoners have reoffended?
This is arguably the question most likely to concern the public, but it is also the question most vulnerable to misleading headlines.
As of 17 October 2026, the sources reviewed for this investigation do not establish a complete, verified national total of further offences committed by people released specifically under the October progression-model tranches.
That does not mean that no one has reoffended. It means a reliable national figure for this specific group has not been established in the material reviewed.
It is also too early to use the absence of a published national total as evidence that the scheme has succeeded or failed.
To measure reoffending properly, analysts need to define the group being studied, establish the follow-up period and distinguish between allegations, arrests, charges and convictions. They must also account for the time it takes for cases to progress through the courts.
A person released yesterday cannot meaningfully be compared with someone who has been living in the community for a year. A fair assessment needs to consider the length of time each person has been observed.
The Government should publish cohort-specific figures as they become available, with clear definitions and an explanation of their limitations.
10. What about victims and their families?
Victims may have legitimate concerns about changes to release dates, licence conditions and the enforcement of restrictions.
The Government has published guidance explaining the progression model, how victims can be informed and what support is available. It has also announced a victim support helpline.
Victims should be able to find out what information they are entitled to receive, whom they can contact and how concerns will be handled. At the same time, personal information about offenders and victims must be managed lawfully and sensitively.
The existence of a helpline or a formal notification process does not automatically establish that every victim has received the information and support they need.
The Government should publish evidence about how these arrangements are working, including whether notifications are timely and how reported concerns are addressed.
The question: Are victims receiving the information and support they need, and how is the Government measuring the effectiveness of these arrangements?
11. What should the Government publish?
Catch a Thief UK believes that the public should be able to assess the policy using a clear, regularly updated set of figures.
We are calling for the Government to publish, where legally and operationally appropriate:
Actual release numbers: How many people were released under each tranche, compared with the original estimates.
Licence compliance: How many people breached conditions, with clear definitions of what counts as a breach.
Recall decisions: How many were recalled to prison, why they were recalled and how quickly action was taken.
Further offending: how many were arrested, charged and convicted of further offences, with those outcomes kept separate.
Electronic monitoring: how many people required tags, how many were fitted on time, and how often devices failed or alerts remained unresolved.
Victim notification: Whether eligible victims were informed in accordance with the relevant arrangements and how concerns were handled.
Probation capacity: Staffing, workload, case management and service-performance data.
Follow-up periods: How long people have been observed after release, so comparisons are fair and meaningful.
The figures should be broken down by release cohort and published with explanatory notes. Where the data are not yet mature enough to draw conclusions, that should be stated plainly.
12. How will we know whether the scheme has succeeded?
No single statistic can answer this question.
A high number of recalls does not automatically prove that early release has failed: recalls can indicate that enforcement mechanisms are identifying non-compliance. But a high number may also raise questions about release planning, supervision or support.
Likewise, a low number of recorded breaches does not automatically prove that the policy is working. It could reflect compliance, but it could also be affected by how effectively breaches are detected and recorded.
The strongest evaluation would examine several outcomes together: public safety, licence compliance, further offending, victims' experiences, probation performance, prison capacity and overall cost.
It should also compare outcomes with a clearly defined baseline and allow enough time for meaningful follow-up.
The public should not be asked to accept a declaration of success based solely on the number of people released, or a declaration of failure based on isolated cases.
The policy should be judged by transparent evidence.
Catch a Thief UK's conclusion
People have a right to ask what happens when prisoners are released early and whether the safeguards designed to protect the public are working.
The available evidence raises important questions about licence compliance, electronic monitoring and probation capacity. But the evidence must be interpreted carefully.
The 13,193 recalls recorded in January–March 2026 relate to the wider system, not the October release tranches. The 3,137 people recorded as not returned to custody following recall at the end of June are also a historical, system-wide total. Wider electronic-monitoring failures warrant scrutiny, but they are not a count of October-scheme breaches.
Without a complete national count of breaches, recalls and reoffending specifically attributable to the October releases, it would be premature to declare the new scheme either a success or a failure.
The Government should publish the information needed to assess its performance, explain any operational failures and demonstrate how risks are being managed.
Victims deserve clear information and appropriate support. Probation staff need the capacity to carry out their responsibilities. The public deserves evidence that the system is working as intended.
This debate should be guided by facts, not assumptions.
Catch a Thief UK will continue following the evidence as further official figures and developments become available.
HAVE YOUR SAY
Do you believe the Government is providing enough information about what happens after early release?
Should it publish regular figures showing licence breaches, recalls, electronic monitoring failures and further convictions for each release cohort?
What safeguards would you want to see to protect victims and the wider public?
Share your views respectfully and help keep the debate focused on evidence and accountability.
Investigation 2 of 12 | Catch a Thief UK
DETECT. REPORT. DETER. RECOVER.
SOURCES AND FURTHER READING
Editorial note: This article reflects sources available as of 17 October 2026. National recall statistics cover the wider system and must not be attributed to the October progression-model releases. The figures for people not returned to custody following recall are historical and system-wide. Reported tagging concerns are attributed to the relevant reporting and parties. A complete national total of breaches or reoffences specifically attributable to the October tranches was not established in the sources reviewed.
Comments