02 / The Atlas · Criminal justice
Arrests & ethnicity
Explore recorded arrests in England and Wales, 2020/21–2022/23. Compare years, police force areas and demographic groups with the definitions in view.
The Atlas / Crime & policing
Examine the arrest record.
Compare ethnicity, age, sex and police force area across three financial years.
April 2020 — March 2023
These are recorded arrests for notifiable offences. A person may appear more than once. An arrest does not establish guilt; these figures are not crimes committed or convictions.
Rates use the 2021 Census population for all three years. Unknown ethnicity has no published population rate.
Inspect the figures 2022/23 · detailed groups
One row per ethnic group. All-ethnicity totals appear above and are not added to group rows.
| Self-defined ethnicityEthnicity | Arrests | 2021 population2021 pop. | Arrests per 1,000Per 1,000 |
|---|---|---|---|
| White British | 405,964 | 44,355,034 | 9.2 |
| White Irish | 5,441 | 507,451 | 10.7 |
| Gypsy or Irish TravellerGypsy / Irish Traveller | 1,165 | 67,776 | 17.2 |
| Roma | 8 | 101,002 | 0.1 |
| Any Other White BackgroundOther White | 43,815 | 3,667,999 | 11.9 |
| Indian | 8,435 | 1,864,315 | 4.5 |
| Pakistani | 17,984 | 1,587,806 | 11.3 |
| Bangladeshi | 5,581 | 644,868 | 8.7 |
| Chinese | 1,240 | 445,613 | 2.8 |
| Any Other Asian BackgroundOther Asian | 13,156 | 972,784 | 13.5 |
| Black African | 19,465 | 1,488,404 | 13.1 |
| Black Caribbean | 14,164 | 623,132 | 22.7 |
| Any Other Black BackgroundOther Black | 15,614 | 297,796 | 52.4 |
| Mixed White and Black CaribbeanMixed: White/Black Caribbean | 9,213 | 513,047 | 18.0 |
| Mixed White and Black AfricanMixed: White/Black African | 2,597 | 249,597 | 10.4 |
| Mixed White and AsianMixed: White/Asian | 2,421 | 488,213 | 5.0 |
| Any Other Mixed/Multiple Ethnic BackgroundOther Mixed | 7,324 | 467,123 | 15.7 |
| Arab | 1,605 | 331,843 | 4.8 |
| Any Other Ethnic BackgroundOther ethnicity | 9,051 | 923,777 | 9.8 |
| Unknown | 84,736 | Unavailable | Unavailable |
On smaller screens, labels are shortened to keep the figures together. Full source names remain in the group selector, accessible labels and CSV download.
Unavailable means no published value for this selection. It never means zero. Population and rate columns are unavailable for age-specific or Other/Unknown sex selections.
Read the method, coverage and original sources ＋
What is being counted?
The Home Office records arrests for notifiable offences under its counting rules. A person can be recorded again after a separate arrest. Where an arrest involves several notifiable offences on one occasion, the principal offence rule applies. These observations do not count unique people, all crimes, charges or convictions.
How are the comparisons kept consistent?
Choose one year, police force area, sex and age category. The broad five groups and detailed nineteen groups are alternative classifications: parent categories are never added to their subgroups. Unknown ethnicity remains visible. The headline total is the source’s all-ethnicities row, rather than a sum of overlapping categories. Other and Unknown sex, and Unknown age, remain available because they form part of the total.
The source CSV stores 2020/21 sex categories in a legacy “gender” column, and the two later years in “sex”. The supplied spreadsheet omitted that legacy column. This display restores the categories from the original official CSV; it does not infer them from arrest numbers or use the spreadsheet’s cached dashboard.
What do the rates mean?
Only source-published rates are shown: arrests per 1,000 people in the corresponding ethnicity, sex and police force area in the 2021 Census. The denominator stays fixed across all three financial years. Rates are available only for all ages and All, Female or Male sex selections. They are not age-standardised; no rate is calculated for an age-specific selection, unknown ethnicity, or Other/Unknown sex.
The City of London has no published local arrest rate because its resident population does not reflect its substantial visiting population. Its arrest counts are included in England and Wales totals. People arrested in any area may live elsewhere. Resident population change after the census, differences in age structure, policing and missing ethnicity affect interpretation.
For small resident groups, a few additional arrests can change a rate sharply. Repeat arrests and arrests of non-residents mean the rate does not measure the share of that population arrested.
Where does coverage stop?
This historical extract covers England and Wales from April 2020 to March 2023. It contains 43 police force areas; Scotland and Northern Ireland are outside its coverage. An additional England and Wales excluding Lancashire aggregate is retained as a source comparator and is never combined with the national total. Roma counts are unavailable in 2020/21 and 2021/22. Changes in recording can interrupt subgroup comparisons.
March 2023 is the end of this extract, not the end of collection. The Home Office published later arrest and ethnicity statistics through March 2025 ↗ on 6 November 2025. Those later years have not been merged into this historical display. The Home Office publishes the arrests; ONS supplies the census population basis.
Inspect the evidence
Historical source published 3 July 2024; retrieved and checked 9 September 2026. This is a fixed, dated extract. The 2022/23 source incorporates a March 2024 correction to North Wales arrests. Read the original publication’s notes before reusing the figures.
Evidence record CT-EV-0005 ↗ · Ethnicity Facts and Figures: number of arrests ↗ · Original official CSV ↓ · Cartulary machine-readable extract ↗ · Home Office counting guidance ↗
Original CSV SHA-256: 61db99c554a75f8f8c030b74b9988f53261186ee2378283b85501d7e1c49e61b
Contains public sector information licensed under the Open Government Licence v3.0.
