Does Prison Work?

Last Updated on September 17, 2026 by Karl Thompson

It depends what you think prison is for. On the evidence:

1. As rehabilitation, not well. 67.1% of adults released from prison sentences of less than 12 months are proven to reoffend within a year.
2. Short sentences do worse than the alternatives. Offenders given short prison sentences reoffend more than similar offenders given community orders or suspended sentences.
3. As incapacitation, it works while people are inside – and some argue offenders deserve to be punished whether or not it reduces crime.
4. As crime control, the link is weak. England and Wales imprisons more people than most of western Europe, without clearly having less crime to show for it.

Whether prison works is a key debate within crime control, surveillance and punishment in the crime and deviance module, examined on AQA Paper 3. The aims of punishment it is judged against are explained in sociological perspectives on punishment.

The prison population in England and Wales

The number of people in jail has been increasing especially rapidly since Michael Howard, then Home Secretary, declared that ‘prison works’ in 1993 – a mantra adopted by successive governments. The prison population of England and Wales doubled in size between 1992 and 2012, and in 2024 it averaged around 87,300, the highest level recorded since 1900. That is about 170 prisoners for every 100,000 people aged 15 and over (House of Commons Library, 2026).

The population was 87,465 at the end of September 2025, and the Ministry of Justice projects it will reach around 100,600 by March 2030 – although those projections do not yet include the effects of the Sentencing Act 2026, which is designed to slow that growth. Between 1993 and 2016 the population grew by an average of 3.6% a year. (In Scotland and Northern Ireland the increase was considerably less during this period.)

This trend would suggest that we have truly entered the era of mass imprisonment (David Garland’s concept), but does prison actually work?

How England and Wales compares

Country	Prisoners per 100,000 population
United States (2021)	531
England and Wales (2024)	145
Scotland (2024)	145
France (2024)	111
Northern Ireland (2024)	96
Canada (2023)	90
Germany (2022)	67
Finland (2023)	51

Source: World Prison Brief, World Prison Population List (14th edition), as reported by the House of Commons Library (2026). International comparisons should be treated with caution, because countries count prisoners differently.

In England and Wales we lock up around 30% more people per head than France and more than twice as many as Germany, which are broadly comparable countries.

Is there a link between prison numbers and crime?

In 2019 the Prison Reform Trust compared prison numbers and crime rates across countries and found no clear link:

• England and Wales have seen a rising prison population and a rising then a rapidly falling crime rate.
• Finland has seen a declining prison population and a rising and then a gradually declining crime rate.
• Canada has seen a broadly level prison population and yet a relatively stable crime rate.

If locking more people up were the main driver of falling crime, you would expect Finland’s crime rate to have risen as its prison population fell. It didn’t. Right realists would reply that crime in England and Wales did fall as the prison population rose – but, as with most crime trends, other explanations are available.

Who is sent to prison, and for how long?

There is an apparent contradiction in the sentencing figures, and it is worth understanding because it trips up a lot of students.

Most people sent to prison get short sentences. In 2025, 58% of those sentenced to immediate custody got less than 12 months (Ministry of Justice, 2026).

But most people in prison on any given day are serving long sentences. In June 2025, 48% of sentenced prisoners were serving four years or more, 18% were serving indeterminate sentences, and only 7% were serving less than a year. Violence against the person was the largest single offence group (35% of prisoners), and sexual offences the second largest among adults (22%) (House of Commons Library, 2026).

The reason is simple: short-sentence prisoners come and go quickly, while long-sentence prisoners build up. So the prison population is not mainly made up of non-violent offenders, even though many of the people passing through prison each year are.

Sentences are getting longer. The custody rate for indictable offences was 33.5% in 2025, more than five percentage points higher than a decade earlier, and the average prison sentence was 19.9 months.

When this post was first written I wasn’t sure what lay behind the rise in long sentences – maybe violent, gang-related and terrorist crimes which involve harm rather than death? Something to research further! The House of Commons Library has since done some of that research: it puts the lengthening of sentences down partly to new offences being created, partly to higher maximum and minimum penalties for existing offences, and partly to changes in release dates.

Does prison reduce reoffending?

If your measure of success is rehabilitation and the prevention of reoffending then it appears not. For offenders released or sentenced between April and June 2024 (Ministry of Justice, published April 2026):

Group	Proven reoffending within one year
All offenders	29.6%
Adults released from custody or starting a court order	37.5%
Adults released from sentences of less than 12 months	67.1%
Adults released from sentences of 12 months or more	23.2%
Children released from custody	66.0%

Those who reoffended committed an average of 5.13 proven reoffences each, the highest in the time series.

NB These are the ones we know about, and this is only reoffending within one year. The actual reoffending rates are more than double these figures, and according to the National Audit Office reoffending costs us the equivalent of staging another Olympic Games every year.

To put these figures in context, if a school had between a third and two-thirds of its pupils leave with no GCSEs, OFSTED would be called in and the management sacked, yet for some reason we tolerate these levels of failure where prison is concerned.

HOWEVER, the Ministry of Justice warns that these rates should not be used on their own to compare sentences, because the people given different sentences are different – as the next section shows.

How do community sentences and suspended sentences compare to prison?

Community orders and suspended sentences seem to be more effective at preventing reoffending than short prison sentences. The obvious objection is that the people who avoid jail may have different circumstances and/or different characters to those who do go to jail – they might just be the kinds of people less likely to reoffend already!

The best British evidence tackles this problem directly. Ministry of Justice researchers matched offenders given short prison sentences with very similar offenders – on over 150 characteristics, including criminal history – who had instead received community orders or suspended sentences (Eaton and Mews, 2019). They found that:

• The one-year reoffending rate after a prison sentence of under 12 months was 4 percentage points higher than if a community or suspended sentence had been given instead.
• Short prison sentences led to around 65 more reoffences per 100 offenders.
• This held whether the prison sentence was under 3, 6 or 12 months.

The comparison matters, because offenders given short prison sentences had on average around 65 previous offences, against 33 for those on community orders and 37 for those on suspended sentences.

The researchers were careful about the limits of their findings. They could not rule out factors they hadn’t measured, over 80% of records had to be discarded to get full information, and the follow-up period started only when prisoners were released – which ignores the crime prevented while they were locked up.

The government has since changed the law in this direction. Since 22 March 2026, courts must suspend prison sentences of 12 months or less unless there are exceptional circumstances (Sentencing Act 2026, section 1).

Why prison might not work: life inside

When this post was first written in 2016, the evidence pointed to three possible reasons why prison doesn’t work.

Firstly, most prisoners (as in about two thirds) had no qualifications and many had the reading age of a 10 year old when they went into jail – and a lack of educational programmes in jail did little to correct this. Basically most prisoners are unemployable before they go inside, and they are doubly unemployable when they come out with a criminal record.

Secondly, many prisoners come from troubled and complex backgrounds – for example, 25% of prisoners had grown up in care and over 40% had no home to go back to when they were released.

Thirdly, at the same time as the prison population was doubling, the number of staff employed in the prison estate had been cut by 30% in the previous five years, with the prison budget being slashed by a quarter.

The result was overcrowding and terrible conditions. It was estimated that 1 in 5 prisoners spent 22 hours a day in their cells; violence and drugs were rife and suicide rates were at their highest for 25 years.

Overcrowding has not gone away. In December 2025, 56% of prisons in England and Wales held more prisoners than they were designed for, and HMP Leeds was at around 172% of its certified capacity (House of Commons Library, 2026).

This means that many jails simply aren’t the kind of environments which are conducive to rehabilitation – this has been the focus of many documentaries, including the BBC’s ‘Life in Wandsworth Prison‘.

This documentary demonstrates how under-staffing has resulted in a lack of care for prisoners, with many being locked down for 23 hours a day, with scant mental health care provision where required (which many prisoners need). In addition to this the documentary also shows how drugs are readily available in the jail, with weed being openly smoked in front of the guards, and it’s clear that many of the prisoners are victims of violence while inside.

It cost £56,618 to hold one prisoner for a year in 2024–25 (HMPPS, 2026) – maybe this money could be better spent on social schemes to prevent offending?

The case that prison does work

Right realists argue that prison works in two ways that reoffending figures miss.

The first is incapacitation. While someone is in prison, they cannot burgle your house. Charles Murray (1997) made this case in a pamphlet with the same title as this post. Even the Ministry of Justice researchers above note that their comparison ignores the crime prevented while offenders are inside.

The second is deterrence: the threat of prison may stop people offending in the first place, and this doesn’t show up in reoffending rates at all.

There is also the argument that some people just deserve to be punished whether they reoffend or not. That is retribution, and if retribution is the aim, prison works by definition.

Evaluating prison against the aims of punishment

✓ Prison works…	✗ Prison doesn’t work…
Incapacitation: offenders cannot commit crimes against the public while inside.	Most prison sentences are short, so the protection lasts only weeks or months.
Retribution: prison delivers proportionate punishment that satisfies the public’s sense of justice.	Retribution does nothing to reduce future crime, and victims often have no say in it.
Deterrence: the threat of prison may prevent offending that never shows up in the statistics.	67.1% of adults released from short sentences reoffend within a year, so it clearly fails to deter them.
Rehabilitation: some prisoners gain education and treatment inside.	Overcrowding, staff shortages and drugs make rehabilitation difficult (Goffman would say impossible).
Right realists link rising imprisonment to falling crime.	Matched comparisons show short sentences lead to more reoffending than community sentences.

Alternatives to prison

If prison doesn’t work, what might? The alternatives usually proposed are:

• Community sentences and suspended sentences, which the evidence above suggests are more effective than short prison terms for similar offenders.
• Restorative justice, which brings offenders and victims together and has a modest but consistent effect on reoffending.
• Early intervention – initially expensive, but cheaper than several years in and out of jail.

The original version of this post also suggested public shaming. The sociological version of that idea is Braithwaite’s reintegrative shaming, which shames the act but not the person – covered in the restorative justice post.

Overall these statistics suggest that while we like to lock people up in England and Wales, there is little evidence that doing so prevents crime.

How ‘does prison work?’ comes up in AQA Paper 3

In a 30-mark essay on punishment or crime control, the evidence here is your evaluation. The strongest point is the matched comparison of short prison sentences with community sentences, because it answers the obvious objection to the raw reoffending figures.

In the short-answer questions, you could be asked to outline reasons why prison may not reduce reoffending. Labelling, the mortification of the self, overcrowding and the lack of education are all good points, each with a brief development.

Link it to the perspectives. Right realists say prison works; Marxists ask who is sent there; Foucault asks why prison survives despite failing. See sociological perspectives on punishment, which also has a full essay plan on prison.

A level sociology revision bundle for sale

If you like this sort of thing, then you might like my Crime and Deviance Revision Bundle – which contains the following:

Crime and Deviance A-level sociology revision bundle

1. 12 exam practice questions including short answer, 10 mark and essay question exemplars.
2. 32 pages of revision notes covering the entire A-level sociology crime and deviance specification.
3. Seven colour mind maps covering sociological perspectives on crime and deviance.

Written specifically for the AQA sociology A-level specification.

Sources

• Eaton, G. and Mews, A. (2019) The impact of short custodial sentences, community orders and suspended sentence orders on reoffending. Ministry of Justice. PDF
• HMPPS (2026) Costs per place and costs per prisoner 2024–25. PDF
• House of Commons Library (2026) Prison population statistics, CBP04334. PDF
• Ministry of Justice (2026) Criminal Justice Statistics quarterly: December 2025. Link
• Ministry of Justice (2026) Proven reoffending statistics: April to June 2024. Link
• Murray, C. (1997) Does Prison Work? London: Institute of Economic Affairs.
• Prison Reform Trust (2019) Prison: the facts, Summer 2019 (Bromley Briefings).

Signposting and related posts

This post has been written for students studying crime and deviance as part of the AQA’s A-level sociology course, specification 7192, examined on Paper 3.

Before this, students usually study how criminals are punished in England and Wales and sociological perspectives on punishment. The main alternative to prison is covered in what is restorative justice and does it work?

Please click here to return to the homepage – ReviseSociology.com