British Prime Minister Andy Burnham bowed to pressure from the right-wing press and excluded serious sexual offenders from the early-release program. But this political concession did not add a single cell to the system. By the fall, the government will have to choose between an unpopular reform and the effective paralysis of the criminal justice system.
Britain’s prison system has reached a point at which the state can no longer fulfill all its promises simultaneously. It cannot keep every convicted offender behind bars for as long as the sentence requires. It cannot quickly build tens of thousands of new prison places. It does not have a probation service strong enough to supervise released prisoners safely. Nor is it prepared to admit publicly that the punitive criminal justice policies of recent decades have become an enormously expensive machine for producing repeat offenders.
Prime Minister Andy Burnham’s decision to suspend the accelerated-release program developed under Keir Starmer was therefore not a routine legislative adjustment. It marked a moment of political capitulation in the face of a conflict that cannot be resolved through declarations about protecting victims. The subsequent resignation of the reform’s architect, former prisons minister James Timpson, only underscored the depth of the rupture.
Burnham, who became prime minister on July 20, 2026, inherited more than overcrowded prisons from Starmer. He inherited a system operating without any strategic reserve, along with legislation that had become politically toxic even before its main provisions took effect. Just two weeks after entering Downing Street, Burnham was forced to announce that people convicted of rape, serious offenses against children, and organized sexual exploitation would not be eligible under the new scheme. Its implementation was postponed from September until October.
The move gave the prime minister a temporary reprieve. But it addressed none of the underlying causes of the crisis. The fall is still approaching. Available places are running out. The probation service is overloaded. The prison construction program is years behind schedule. And every new concession to public outrage reduces the number of prisoners eligible for release, bringing the system closer to the moment when emergency releases will become unavoidable.
The Month Burnham Does Not Have
The political logic behind Burnham’s decision is understandable. At the center of the controversy was the prospect of releasing people involved in crimes that cannot be explained to voters through the language of statistics and risk management. Particularly damaging was the case of police officer Andrew Harper, who was killed in 2019. The possibility that two men convicted for their involvement in his killing could be released early transformed a technical reform of sentencing policy into an emotional argument about how much the British state values a police officer’s life.
For the Conservatives, Reform UK, police unions, and the right-wing press, it was the perfect story. A complicated law could be reduced to a single sentence: dangerous criminals are being released early because the government failed to build enough prisons. In such a debate, the Ministry of Justice inevitably loses. It talks about percentages of sentences served, release licenses, electronic monitoring, and risk categories. Its opponents name the dead and put victims’ relatives in front of the cameras.
Burnham chose the politically rational course. He excluded the most inflammatory categories and delayed the scheme’s launch. But postponing it until October is an accounting maneuver, not a strategic solution. According to available estimates, without measures to reduce the prison population, the system could run out of available places as early as November. At the end of June, prisons in England and Wales held approximately 86,000 people, while the male prison estate was operating at roughly 98 percent of capacity. The government had planned to release as many as 6,000 prisoners to prevent overcrowding.
In other words, Burnham bought one month by sacrificing a substantial part of the system’s operational buffer. A month is not enough to build a prison, recruit and train thousands of probation officers, establish a comprehensive rehabilitation network, or sharply reduce the flow of new inmates. The government can only revise the lists of those eligible for release, strengthen electronic monitoring, and hope that prison population growth proves less severe than officials expect.
But hope is not a prison policy.
The Arithmetic of Disaster
The crisis did not suddenly emerge when Labour came to power. In August 2024, the prison population in England and Wales reached an all-time high. The authorities came so close to exhausting available capacity that they began releasing prisoners under emergency measures, moving detainees into police cells, and postponing certain court proceedings.
Between September and December 2024, 16,231 people were released under the SDS40 program. This did not mean that their punishment ended entirely. The point of conditional release was moved from halfway through the sentence to the 40 percent mark, after which the offender remained under supervision subject to license conditions. Yet the sheer scale of the operation revealed the true condition of the state. Prisons were not being emptied because rehabilitation had suddenly become a national priority. They were being emptied because available cells had nearly run out.
Even that failed to alter the long-term trajectory. In March 2026, 87,342 people were incarcerated. By June, the number had fallen to approximately 86,000, but the decline was temporary and had been achieved through a continuing acceleration of releases. The Ministry of Justice had previously forecast that the prison population could rise to between 98,000 and 103,600 by March 2030. Its central estimate was approximately 100,600.
This is the central paradox of British policy. The government continues to pass laws increasing sentences for serious crimes, hire more police officers, expand criminal prosecutions, and promise harsher punishment. But the infrastructure required to implement those decisions is expanding far more slowly.
Between May 2010 and September 2024, the net increase in prison capacity was just 1,005 places. New cells opened, but older facilities were simultaneously taken out of service. Over the same period, the prison population increased by approximately 3,000. Since October 2022, the male prison estate has routinely operated at between 98 and 99.7 percent of capacity. A fire, repairs, an outbreak of violence, a staffing shortage, or the closure of a prison wing could destabilize the entire system.
Such a model cannot be considered sustainable. It is management on the brink of failure.
Britain Spent Decades Buying Votes With Prison Sentences
The crisis is not merely the result of insufficient prison construction. For too long, Britain has treated imprisonment as universal proof of political resolve.
Since the 1990s, both Conservatives and Labour politicians have feared appearing soft on crime. Ministers competed to increase minimum sentences, restrict opportunities for early release, and create new criminal offenses. Over three decades, the prison population of England and Wales effectively doubled. Yet many of these political decisions were made without corresponding funding for new prison places, repairs to aging facilities, probation services, or rehabilitation programs.
England and Wales incarcerate approximately 136 people per 100,000 residents. The corresponding figure is about 69 in Germany and 55 in Norway. The gap with Italy and Spain is smaller than the most emotional critics of the British model suggest. Italy incarcerates roughly 110 people per 100,000 residents, while Spain incarcerates about 121. Nevertheless, England and Wales remain among the most incarceration-oriented jurisdictions in Western Europe.
The problem is not that serious criminals receive long sentences. Society has every right to demand the isolation of murderers, rapists, terrorists, and particularly dangerous repeat offenders. The problem arises when prison is used for people whose situation is almost certain to deteriorate after a short period of incarceration: people with addictions, the homeless, the mentally unstable, petty thieves, and offenders whose crimes are rooted in social breakdown.
A short prison sentence destroys employment, housing, and family ties, yet rarely provides enough time for treatment, education, or vocational training. After several months, the person emerges poorer, less stable, and more deeply connected to the criminal world. Formally, the state has punished him. In practice, it has financed the preparation of his next crime.
Timpson’s Reform Was Not an Amnesty but an Attempt to Redesign Punishment
The Sentencing Act of 2026 emerged in response to an independent review of the system. Its central concept was an earned-progression model under which release depended on the prisoner’s conduct and the nature of the sentence.
For standard determinate sentences, the earliest possible release point was moved to one-third of the sentence. For more serious categories, whose members had previously been released after serving two-thirds, the minimum threshold was set at one-half. Misconduct, violence, drug use, and serious disciplinary violations were intended to postpone the release date. Life sentences, extended sentences for dangerous offenders, and cases requiring a decision by the Parole Board were excluded from the reform.
The second element of the reform established a presumption against short prison sentences of up to 12 months. Courts retained the power to imprison someone who posed a threat to a specific victim, repeatedly violated court orders, or met other exceptional criteria. Under ordinary circumstances, however, judges were expected to use suspended sentences, community service, treatment, and intensive supervision.
The third element was the largest expansion of electronic monitoring in the country’s history. The government planned to place tens of thousands of people under electronic monitoring, combined with curfews, mandatory programs, and regular contact with probation officers.
In theory, this structure is more rational than mechanically releasing prisoners early. It links release to conduct, preserves extended supervision after release, and directs prison resources toward those who genuinely need to be isolated.
But the reform had a fundamental defect from the outset. It was designed as a long-term restructuring of the system, yet it had to be implemented as an emergency operation to free prison cells. As a result, the public did not see a new philosophy of punishment. It saw a list of criminals who would be released earlier than the courts had promised.
When a reform is simultaneously expected to conserve prison capacity, reduce reoffending, and protect victims, even the smallest mistake in selecting someone for release can destroy confidence in the entire model. One high-profile case is enough to erase thousands of successful releases from public view.
Timpson Left, but the Reasons for His Departure Remain
James Timpson’s appointment in 2024 was one of the Starmer government’s most unconventional decisions. He came not from the parliamentary establishment but from business. For many years, his family company, Timpson, had employed former prisoners, demonstrating that stable employment after release reduces the likelihood of reoffending.
Timpson openly argued that a substantial proportion of prisoners should not be in prison. In British politics, which had grown accustomed to measuring determination by the number of cells and the length of sentences, this bordered on heresy.
His resignation following the change of prime minister signaled the end of the reform’s technocratic phase. Timpson himself did not say that he had resigned because of disagreements, limiting his remarks to the honor he had felt in serving in government. Nevertheless, the timing of his departure, coinciding with the revision of the central reform program, inevitably created the impression of a political rupture.
Catherine McKinnell replaced him at the Ministry of Justice, while the department came under the leadership of the new lord chancellor and justice secretary, Alex Norris. Their problem is not limited to the substance of the law. They must restore the reform’s administrative coherence after the prime minister publicly cast doubt on its criteria.
Every probation officer who had been preparing a prisoner’s release must now determine whether it will happen, when it will happen, and under what legal mechanism. Every victim who previously received notification is awaiting a new decision. Every prison must recalculate its available capacity. A political pause in Downing Street is becoming operational chaos across the country.
The Probation Service Is the Weak Link Supporting the Entire Reform
A person can be released from a cell, but the state cannot be released from responsibility for that person’s conduct. The earlier a prisoner is released, the more heavily the system depends on probation.
This is where the most dangerous contradiction emerges. During the 2024–2025 fiscal year, Britain’s probation service met only seven of its 27 performance targets. Three years earlier, it had met approximately half. By March 2025, the number of prisoners returned to custody for violating license conditions had reached 13,583. That represented 15 percent of the entire prison population and an increase of 49 percent compared with June 2021.
By March 2026, the situation had become even more alarming. Over the previous 12 months, 51,419 people had been recalled to prison. During the first quarter of 2026, the number of people recalled exceeded the number released for the first time. The system began operating like a revolving door. The government releases prisoners to free capacity, while the probation service returns a comparable number to prison for breaching their conditions.
This does not necessarily mean that everyone recalled to prison committed another serious crime. A person may be recalled for violating a curfew, failing to report, contacting a prohibited individual, using drugs, or otherwise breaching license conditions. From the standpoint of prison capacity, however, the result is the same. The person occupies a cell again.
During 2023–2024, 770 people under supervision were charged with new serious offenses, an increase of 55 percent compared with 2020–2021. Cases like these become political bombs. They allow opponents of reform to argue that the probation service cannot guarantee public safety.
The government promised to increase investment in probation gradually by as much as £700 million. But money does not instantly become qualified personnel. Staff must be recruited, trained, and retained. An electronic ankle monitor is not enough to manage a single complex case. Effective supervision requires housing, addiction treatment, psychiatric care, employment, and continuous risk assessment.
Without these elements, early release simply transfers the problem from prison to the streets and then back to prison.
A £59,000 Prison Place That Produces Another Crime
Maintaining a single prison place costs British taxpayers approximately £59,000 a year. In exchange for that money, the state should do more than lock a cell door. It should also reduce the likelihood that a prisoner will create another victim after release. Yet in many institutions, it fails to perform this second task.
According to official statistics, the overall proven reoffending rate among offenders stood at 28.9 percent in early 2024. Among adults released after serving sentences of less than 12 months, it reached 67.1 percent. This means that approximately two out of every three former short-term prisoners committed another offense during the designated follow-up period.
Such a rate cannot be explained solely by the moral character of offenders. It reflects an institutional model. People are placed in overcrowded facilities where they spend most of the day confined to their cells, receive limited access to education and treatment, encounter drugs and violence, and are then released without stable housing or employment.
The prison inspectorate found that 28 of the 38 men’s and women’s prisons it inspected were performing poorly or insufficiently well in providing constructive activities. Overcrowding, boredom, broken furniture, infestations, and prolonged confinement in cells increased the demand for drugs.
The state pays £59,000 a year for an environment that, in many cases, increases the likelihood of another crime. It then spends more money on police, courts, transportation, and the reincarceration of the same person. Such a system may appear tough, but economically it is irrational.
Drugs Have Become the Internal Currency of Prisons
British prisons ceased long ago to be spaces isolated from the illegal market. They have become concentrated versions of it.
According to the National Audit Office, approximately 40,000 prisoners, roughly half of the prison population, had an identified drug problem in April 2025. Among adult male prisoners surveyed, 39 percent said illegal substances were easy to obtain. Drug discoveries increased by 25 percent during 2024–2025, while reported drone sightings rose by 43 percent.
Drones have transformed the prison economy. They allow synthetic drugs, phones, weapons, and other prohibited items to be delivered directly to prison windows. Inside, this creates debt, extortion, control by criminal groups, and pressure on prisoners’ relatives to make payments.
Between December 2022 and December 2024, the Prisons and Probation Ombudsman investigated 833 deaths in custody, 136 of which were drug-related. At the same time, 35 percent of scheduled addiction treatment appointments in 2024–2025 were missed, while 24 percent of those requiring assistance did not receive an initial assessment within three weeks of their medical examination.
This reveals the limits of the rhetoric of retribution. A prisoner may be deprived of liberty, but if the state does not control the internal drug market, effective authority over part of the institution passes to criminal networks.
A prison in which illegal substances are easier to obtain than treatment does not rehabilitate an offender. It reinforces addiction, debt relationships, and criminal connections.
Victims Have a Right to Be Heard, but They Should Not Determine Criminal Justice Policy
Critics of the right-wing campaign make a mistake when they portray the families of those killed as obstacles to rational reform. Victims have the right to know when an offender will be released, to receive protection, to participate in discussions about release conditions, and to demand that the state not expose them to renewed danger.
But there is a fundamental difference between giving victims a role in the process and granting them the power to determine the overall model of punishment.
Criminal justice policy must take risk, evidence, effectiveness, and proportionality into account. The pain of one family cannot answer the question of what to do with 86,000 prisoners, tens of thousands of people with addictions, and a system that is running out of cells. Just as a minister’s opinion cannot negate a victim’s suffering.
Burnham tried to reconcile the irreconcilable: preserve the law while excluding the categories that provoke the greatest public outrage. This is a reasonable political defense, but it creates a new risk. The more exceptions there are, the fewer prison places are freed. The fewer places are freed, the closer the system moves toward emergency measures. And emergency releases are almost always less selective and less carefully prepared than a planned reform.
The right may win the argument over individual offenders. But if that victory forces the government to abandon a controlled reduction of the prison population, the result will not be tougher punishment. It will be a less controlled system.
Women’s Prisons: Punishment That Extends to Children
The ineffectiveness of a one-size-fits-all approach to imprisonment is particularly evident in the treatment of women.
In March 2026, prisons in England and Wales held 3,461 women. They represented only a small share of the overall prison population, but the consequences of incarcerating women often extend far beyond the cell. Children lose their primary caregiver, families lose their homes, and social services inherit new crisis cases.
A significant proportion of women are held on remand. Some are ultimately not given custodial sentences. Many female prisoners are survivors of domestic and sexual violence, struggle with addiction, or have severe mental health problems. In December 2024, approximately one-quarter of women in prison were awaiting trial and facing proceedings that might not result in imprisonment.
This does not mean that women should not be held accountable for their crimes. The issue is choosing a punishment that protects society while minimizing collateral harm. For a nonviolent offense, community service, mandatory treatment, secure housing, and strict supervision may be less expensive, more demanding in terms of daily oversight, and more effective than a short prison sentence.
Prison sometimes punishes not only the guilty woman but also her children, who committed no crime. The state must account for that cost rather than presenting it as an unavoidable side effect of justice.
Norway Is Not a Paradise, but Its Results Cannot Be Dismissed With Ridicule
British opponents of reform traditionally mock Norwegian prisons for their comfort, private rooms, and respectful treatment of inmates. Behind this criticism lies an old assumption: punishment is considered genuine only when it includes humiliation.
The Norwegian model is based on the opposite principle. The punishment is the loss of liberty. Everything else should be organized so that a person can return to society less dangerous than before incarceration. The system therefore supports family ties, vocational training, therapy, and connections to the local labor market.
Norway incarcerates approximately 55 people per 100,000 residents, compared with 136 in England and Wales. But simply copying Norwegian prisons would not guarantee success in Britain. The countries differ in population size, patterns of crime, social policy, housing markets, and the capacity of local government services.
The central Norwegian lesson is not the architecture of its cells. It is consistency. When the state declares rehabilitation to be an objective, it funds education, treatment, personnel, housing, and post-release support. Britain often adopts a single element, such as an electronic ankle monitor, an educational program, or a support center, without building a complete chain around it.
Britain cannot achieve Norwegian reoffending rates while retaining British overcrowding, staffing shortages, and housing insecurity among former prisoners.
Construction Is Five Years Behind Schedule and Billions Over Budget
Supporters of a hard-line approach offer an apparently simple solution: do not release prisoners; build more prisons. But the arithmetic of construction is as unforgiving as the demographic figures.
In 2021, the authorities promised to create 20,000 additional prison places by the middle of the decade. By September 2024, only 6,518 places had been added, including 278 created not through new cells but by increasing the density of inmates in private prisons. Completion of the program was postponed until 2031, approximately five years later than originally planned.
The program was initially estimated to cost about £5.2 billion. The forecast subsequently rose to between £9.4 billion and £10.1 billion. The increase amounted to at least £4.2 billion, or approximately 80 percent. The causes included unrealistic timetables, inflation, planning problems, weak risk management, and the collapse of the major contractor ISG.
At the same time, the backlog of repairs in older facilities continued to grow. By September 2024, it was estimated at £1.8 billion, compared with £900 million in 2020. The Ministry of Justice calculated that bringing the entire prison estate to even a satisfactory condition would require approximately £2.8 billion.
The new government reaffirmed its intention to invest approximately £7 billion and create 14,000 additional places by 2031. But this program cannot resolve the crisis of October or November 2026. A prison cannot be built through a parliamentary declaration. Even temporary modular units require sites, utilities, staff, security arrangements, and permits.
The slogan “build, do not release” is therefore not a plan. It is a way of avoiding the question of what must be done in the coming months.
The Right Won the Narrative but May Inherit Responsibility for the Chaos
The tactical winner of the conflict was the British right. The Conservatives and Reform UK forced the new prime minister to retreat almost immediately after taking office. They demonstrated to voters that pressure works and that Labour’s reform had failed to take sufficient account of victims.
But this victory has a limited shelf life. If prisons become overcrowded in the fall, courts begin postponing cases, police are forced to hold convicted offenders in temporary detention cells, and the government returns to emergency releases, the right will have to explain what alternative it offered.
Harsher sentences without funding for additional prison places brought the system to its present condition. Much of the failure to expand capacity occurred under Conservative governments. The National Audit Office linked the crisis to the inability of previous administrations to align criminal justice policy, demand for prison places, and infrastructure budgets.
The danger is even greater for Burnham. He entered Downing Street as a politician who promised pragmatism, administrative competence, and attention to the regions. If a prison collapse becomes his first major crisis, his image as a capable administrator will be destroyed.
He has already assumed responsibility for revising the reform. Starmer and Timpson can no longer bear all the blame. Once Burnham intervened publicly, the program became his, even though its main provisions had been designed before he took office.
Four Scenarios for the Fall
The most likely scenario is a partial launch of the reform in October with an expanded list of exclusions. The government will preserve release after one-third or one-half of a sentence for less dangerous categories, strengthen electronic monitoring, and announce additional protections for victims. This would free several thousand places but place even greater pressure on the probation service.
The second scenario is another postponement. Politically, it may appear safer, but in practical terms it would bring prisons closer to exhausting their capacity. The authorities would then have to use police cells, accelerate selected releases through administrative decisions, or restrict the admission of new prisoners. Such an approach would be less transparent than a planned reform.
The third scenario is a complete rejection of the earned-progression system in favor of prison construction. This would win Burnham temporary approval from right-wing voters, but it would not physically eliminate the shortage of places before 2031. The government would still be forced to use the same early-release measures, merely under a different name.
The fourth and most rational course would be to preserve the reform while dividing prisoners into risk categories, strengthening individualized assessments, guaranteeing that victims receive timely information, urgently funding probation services, and expanding specialized courts that impose intensive supervision. At the same time, the government would need to limit short prison sentences for nonviolent offenses and direct people with addictions into mandatory treatment and closely supervised community programs.
The rational course, however, is also the most politically difficult. It requires acknowledging that punishment cannot be measured solely by the number of months spent behind bars, and that public safety is not synonymous with overcrowded prisons.
The Central Cell of the British System Is in Downing Street
The debate over prison reform is wrongly presented as a choice between compassion for offenders and respect for victims. The real choice is different: between a system that manages risk and one that responds to every scandal by creating another exception.
Burnham is right to exclude from accelerated release those categories for which the cost of an error would be particularly high. But he will be wrong if he decides that this concession has resolved the problem. Prisons remain overcrowded. Drugs remain readily available. The probation service cannot cope with its workload. Construction is behind schedule. Reoffending among short-term prisoners remains catastrophic.
The right won a political episode, but it has not proposed a model capable of simultaneously keeping dangerous offenders behind bars, reducing repeat crime, and preventing the state from physically running out of cells.
Timpson lost the battle within the government because he tried to speak about evidence in a political environment that demands symbolic retribution. Burnham gained several weeks because he listened to public anger. But in the fall, anger will not substitute for an available cell, a trained probation officer, or a functioning treatment program.
Britain’s prison crisis was not caused by excessive leniency. It was caused by decades of performative toughness unsupported by funding, infrastructure, or sound judgment.
And if Burnham ultimately subordinates criminal justice policy to newspaper headlines, the principal casualty will be neither the reputation of one prime minister nor the career of one minister. It will be the British state’s ability to enforce its own sentences.