Migrant crime in Britain: what prison figures show and what remains unmeasured

Foreign nationals made up about 12% of the prison population in England and Wales on September 30, 2026. The Ministry of Justice counted 10,217 people: 6,439 sentenced prisoners, 3,381 on remand and 397 held for non-criminal reasons. Those figures establish a substantial demand on the justice system. They do not establish the share of all crime committed by migrants. Ministry of Justice, published October 29.

The distinction matters because debate about migrant crime often combines several questions. How many non-British citizens are convicted? Which offenses are involved? Did they arrive on a work visa, seek asylum or enter without permission? How many are removed after serving sentences? The available statistics answer some of these questions much better than others.

An October 21 response from the Office for National Statistics makes a central gap clear. ONS does not hold the criminal-justice outcome data needed to produce the requested analysis linking immigration and asylum status to those outcomes. It also says it has not undertaken that linkage. That is a statement about ONS’s holdings and work, rather than proof that every relevant record is absent across government. ONS freedom-of-information response.

What the latest prison snapshot establishes#

The September figures cover England and Wales, not all four UK nations. The total prison population was 87,712. Foreign nationals were around 5% fewer than a year earlier, according to the same release. A description of the latest snapshot as an unchecked rise would therefore be wrong for this particular measure and comparison period. Official prison statistics.

Source: Ministry of Justice, England and Wales. A remand prisoner may be awaiting trial or may have been convicted but not yet sentenced; the whole category cannot be described as convicted offenders.

Sentenced

6,439

Remand

3,381

Non-criminal

397

Source: Ministry of Justice

Show the numbers

Custody category	People
Sentenced	6,439
Remand	3,381
Non-criminal	397

A prison snapshot gives greater weight to people held for a long time. Someone serving a long sentence can appear in many successive snapshots; someone given a short sentence may enter and leave between them. The stock of prisoners therefore answers a different question from the number of people convicted or admitted during a year. Neither should be silently relabeled as the number of crimes committed.

Nationality, birth and immigration status describe different things#

Nationality is a legal status. Country of birth records where someone was born. A person born overseas may be British, while a non-British citizen may have lived in the UK for decades. ONS guidance explicitly warns that ethnicity is not a measure of migration. A finding about men of Pakistani heritage, for example, cannot establish that they recently migrated or claimed asylum. Our grooming-gangs report examines the local findings and their limits. ONS definitions and reporting guidance.

The same distinction applies to a British person born overseas and to someone born in Britain who holds another nationality. Switching definitions halfway through a comparison changes who is counted. A newspaper headline about immigrants may therefore describe a dataset that actually counts passport or citizenship records. Readers need to know which measure the source used.

Immigration status is another variable. A nationality label does not identify a work visa, refugee status, a pending asylum claim, a visitor or someone without permission to remain. A valid comparison of those routes requires records that identify the route itself. Inferring asylum status from the prevalence of asylum applications among people of the same nationality is a substitution, not a measurement.

Why population estimates matter as much as offender counts#

A crime rate needs both a numerator and a denominator: the relevant incidents or people, and the population exposed over the same period. An accurate prison count does not fix an inaccurate population estimate. If a migrant group has grown since the census, dividing a recent count by the old population can exaggerate its apparent rate.

This was a problem with a 2025 Conservative claim that Channel migrants were 24 times more likely to go to prison than British citizens. Full Fact found that the published account appeared to apply nationality-wide custody figures to small-boat arrivals and compare 2025 prison data with 2021 population estimates. It judged the figure unreliable. Full Fact also recorded that Chris Philp’s office disputed its assessment and did not issue a correction. Full Fact’s review and recorded response.

The criticism does not establish an alternative offending rate for small-boat arrivals. It identifies why that calculation cannot establish one. Arrivals from the same country through different routes may have different backgrounds and circumstances; they cannot be assumed to be interchangeable without evidence.

Age and sex also matter. Comparing one population with many young men to another with more children or older adults can produce a misleading impression about nationality’s independent effect. A useful release should show the raw figures and explain any adjustment, so readers can assess both the actual caseload and the demographic comparison.

What research says about differences between groups#

Oxford’s Migration Observatory reported that non-UK citizens accounted for 13.1% of people receiving cautions or convictions in 2024. Its August 2026 briefing found substantial variation by nationality and offense type, and said age-adjusted comparisons differed from headline comparisons. It also stressed missing immigration-status information and uncertain population estimates. These findings describe justice-system outcomes; they are not a complete count of offending. Migration Observatory briefing.

There is a further distinction between measuring who was convicted and estimating whether immigration changed crime. A person can be convicted of an offense without that observation establishing how the national crime rate would have developed under a different migration policy. That second question requires a comparison designed to examine cause and effect.

A peer-reviewed study by Brian Bell, Francesco Fasani and Stephen Machin examined two earlier migration waves. It found an increase in property crime associated with the asylum wave of the late 1990s and early 2000s, a different effect for workers arriving after EU enlargement, and no observable violent-crime effect for either wave. The authors emphasized differences in labor-market opportunities. Published study.

That is evidence about particular periods and policy settings. It should not be presented as a direct measurement of people arriving by small boat in 2026. Nor does an area-level effect identify the nationality of every offender. Earlier research can help frame new questions without settling a question its data did not observe.

National crime trends cannot supply the missing link#

The latest Crime Survey for England and Wales estimated 9.7 million headline crime incidents in the year ending June 2026, with no statistically significant change from the previous year. Its headline measure includes offenses such as theft, fraud and violence, while sexual victimization and domestic abuse estimates are reported separately. ONS crime bulletin.

Police records and a victim survey measure different things. The survey captures experiences whether or not they were reported to police, but has limits on the populations and offenses it covers. Police figures cover a wider range of offenses, but depend on reporting and recording practices. ONS warns against treating changes in recording as direct changes in underlying crime.

Neither a national fall nor a national rise identifies a migration effect on its own. Crime can change alongside migration for several reasons. A persuasive claim must show why competing explanations do not account for the pattern, rather than place two lines on a chart and assume one caused the other.

The data are improving, but important questions remain#

The Home Office says upgraded collection systems allowed new breakdowns of foreign-offender returns and detention outcomes to be published in May and August 2026. It also says further work is needed before additional breakdowns of offenders living in the community can be released. That is progress in monitoring the immigration-enforcement system, rather than a complete national migrant-offending series. Our returns and community-monitoring report explains what those enforcement measures count. Home Office statistical update.

The strongest case for fuller publication is practical: agencies and the public need to know where serious harm occurs, whether systems identify offenders and which interventions work. That requires data detailed enough to distinguish offenses, outcomes and arrival routes. Publication should also make unknown or missing fields visible, so gaps cannot disappear inside a precise-looking total.

The strongest caution concerns what comparisons can support. Better records would still need matching population estimates, dates and geography. Linking administrative systems would also require checks that the records refer to the same people. A large dataset can create false precision if those checks are skipped.

For now, the September prison count is firm evidence of the number of foreign nationals held on that date. Individual convictions can establish serious crimes by identified offenders. The broader rate of crime by immigration route remains a different question, and the sources reviewed here do not supply a comprehensive answer. Reporting that limit should accompany scrutiny of offending, rather than substitute for it. The same need to match a crime claim to its population appears in our examination of Japan’s immigration debate. That comparison concerns methods, not a claim that the countries have the same crime patterns.

Methodology: Primary releases and original research were checked November 5, 2026. Prison figures cover England and Wales on September 30; the crime survey covers the year ending June 2026. No nationality-wide rate is applied to an asylum or small-boat cohort. This article does not estimate unreported crime by immigration status.

Photo: Tbmurray, Old Bailey (2012), CC BY 3.0. Cropped for display. Archival image.

The distinction between allegations, findings and missing evidence follows 567politics’ sourcing and attribution standards.

Differing perspectives

What would make reporting on migration and crime more useful?

Each position is summarized as its supporters make it, with equal weight and in alphabetical order. A summary is not an endorsement; the evidence for each is weighed in the story above.

• Match definitions and account for demographic differences

  Who holds it: University of Oxford’s Migration Observatory

  The Observatory stresses that nationality, immigration status, population estimates and justice-system outcomes measure different things. It argues that differences in age, sex and socioeconomic circumstances matter when comparing groups. From this view, fuller publication is useful when its limits are explicit and comparisons do not assign nationality-wide findings to an unmeasured immigration route.

  In their words: Migration Observatory briefing, August 18, 2026 ↗

• Publish fuller, reliable administrative breakdowns

  Who holds it: Home Office statistical development program

  The Home Office describes fuller reporting as a way to meet public and parliamentary demand for information about foreign offenders and enforcement outcomes. It has upgraded collection systems and added breakdowns during 2026. This approach favors improving and publishing records as their quality becomes sufficient for statistical use.

  In their words: Home Office update on foreign-offender statistics ↗

{{cite news |title=Migrant crime in Britain: what prison figures show and what remains unmeasured |url=https://567politics.com/news/uk-migrant-crime-prison-nationality-data-evidence-november-2026 |work=567politics |author=567 Politics staff |date=2026-11-06 |language=en}}

{
  "id": "https://567politics.com/news/uk-migrant-crime-prison-nationality-data-evidence-november-2026",
  "type": "article-newspaper",
  "title": "Migrant crime in Britain: what prison figures show and what remains unmeasured",
  "URL": "https://567politics.com/news/uk-migrant-crime-prison-nationality-data-evidence-november-2026",
  "container-title": "567politics",
  "language": "en",
  "author": [
    {
      "literal": "567 Politics staff"
    }
  ],
  "issued": {
    "date-parts": [
      [
        2026,
        11,
        6
      ]
    ]
  }
}