36% vs 15%
Germany 2025: foreign citizens' share of police suspects and of residents
3.2×
Germany: foreign suspect rate compared with Germans, before adjusting for age or for visitors counted as suspects
4.3 → 8.6 → 20.7
Denmark 2025: guilty per 1,000 among Danish origin, non-Western immigrants and their Danish-born children
85% vs 88%
Finland 2025: people of Finnish background among suspects and among residents
How to read these figures
- These are raw rates. Crime peaks among young men, and many immigrant groups have more of them, so part of any gap is age and sex. Studies that adjust for age, income and education usually find the gap shrinks but does not disappear.
- They count people the police or courts recorded, not all crime. Reporting and policing can differ between groups.
- Suspect counts often include tourists, commuters and asylum seekers, while the population counts only registered residents. That pushes up foreign rates, most of all for small groups near a border. Each chart says whether it applies.
- Citizenship is not origin. Where only citizenship is published, naturalised immigrants and their children count as natives. Compare multiples within a country rather than raw rates between countries, since each counts crime differently.
Natives and foreigners in each country
The foreign share of people counted, against the foreign share of residents. The further apart the dots, the larger the gap.
Highest and lowest groups in each country
Every group an office publishes, ranked by rate. Switch country, offence or generation.
One origin across countries
The same origin group can sit far above natives in one country and close to them in another.
Compare any groups
Pick up to four groups from any country and see them across types of crime.
Change over time
Every country
Few countries outside Europe publish crime by origin. Almost all report foreign prisoners, so this is the widest view available.
United States: by race
The U.S. publishes no national crime series by birthplace or citizenship. Race is what exists. Rates are in the ranking above.
Sentenced prisoners (sentence > 1 year) under state or federal jurisdiction, by race and Hispanic origin. White and Black are non-Hispanic. Counts rounded to nearest 100 for race/ethnicity. Remaining “other/two or more races” are in the BJS total but not shown as a separate series here.
Have a correction, a better estimate, or additional data—at the global, national, regional, or city level? Help improve the data
Arrests for murder and nonnegligent manslaughter by race (FBI UCR Table 43). Same race/ethnicity caveats as total arrests.
Have a correction, a better estimate, or additional data—at the global, national, regional, or city level? Help improve the data
Who publishes what
Offices differ sharply. Denmark, Finland and the Netherlands publish by origin. Germany and Norway publish by citizenship for each nationality. Many others publish only foreign against national, or nothing. We never fill in figures. Updated 2026-10-31.
| Country | Status | Detail |
|---|---|---|
| Denmark | Full series ancestry | Statistics Denmark STRAFNA3, STRAFNA4 and STRFNA10: persons found guilty by country of origin, offence and generation, annually. Source |
| Norway | Full series citizenship | SSB StatBank 09421: persons charged by citizenship and offence, annually. Figures by immigrant background (including Norwegian-born to immigrant parents) come only in periodic one-off tables. Source |
| Netherlands | Full series origin | Statistics Netherlands (CBS) 85658NED: suspects per 10,000 by origin, generation, education and household income, annually. Turkey, Morocco, Suriname, the Dutch Caribbean and Indonesia are the named origins. Source |
| Finland | Full series citizenship and origin | Statistics Finland 13jg and 13z4: suspects per 10,000 by citizenship, origin, generation, background region and offence, annually. Source |
| Germany | Full series citizenship | BKA Police Crime Statistics table T62: suspects by every citizenship and offence, annually. Germany does not record migration background, so naturalised citizens count as Germans. Source |
| United States | Full series race | BJS publishes sentenced state/federal prisoners by race and Hispanic origin annually. FBI UCR Table 43 publishes arrests (and murder arrests) by race. There is no continuous federal convictions-by-immigrant-ancestry series comparable to Denmark’s STRAFNA9. Source |
| Sweden | Partial background reports | Brå publishes research reports on registered offending by native/non-native background (e.g. report 2021:9 with 2007–2018 offence shares). Not a continuous open StatBank series like Denmark’s STRAFNA9. Source |
| Austria | Partial citizenship | Interior ministry / Statistik Austria publish some suspect or prisoner figures by citizenship. Eurostat prisoners-by-citizenship covers Austria. Foreign and national totals are in the Eurostat comparison on this page. Source |
| Switzerland | Partial citizenship and residence | Federal Statistical Office: police defendants split into Swiss citizens, permanent foreign residents, the asylum population and other foreigners, annually. Convictions by nationality appear in separate releases. Source |
| France | Partial citizenship | France does not publish crime by ethnicity, ancestry or immigrant background. It does report suspects, convictions and prisoners by citizenship (foreign or French) through Eurostat. Source |
| United Kingdom | Partial nationality | Ministry of Justice publishes prisoners by nationality each quarter for England and Wales, and offender ethnicity in census categories. There is no matching rate by nationality. Source |
| New Zealand | Partial ethnicity | Ministry of Justice / Stats NZ publish offender ethnicity; not immigrant ancestry. Source |
| Italy | Partial citizenship | ISTAT / Interior Ministry publish some foreign-national offender and prisoner statistics (citizenship). Foreign and national totals are in the Eurostat comparison on this page. Source |
| Spain | Partial citizenship | INE / Interior publish foreign-national prisoner and some offender statistics. Foreign and national totals are in the Eurostat comparison on this page. Source |
| Belgium | Limited / reports citizenship | Some judicial / prison statistics by nationality; no Denmark-style ancestry conviction series. Foreign and national totals are in the Eurostat comparison on this page. |
| Australia | Limited / reports reports | Some state-level or one-off reports; no national ABS convictions-by-ancestry open series like Denmark. Source |
| Canada | Not published | Statistics Canada does not publish a national convictions-by-immigrant-status or ancestry time series comparable to Denmark. Source |
What this page measures
- Intentional Homicide Rateper 100,000
- Intentional homicides per 100,000 people. Source: World Bank Open Data.
- Foreign Citizenship Share of Prisoners% of prisoners
- Share of persons held in prison who hold foreign citizenship (%). Eurostat crim_pris_ctz. Citizenship is not the same as immigrant ancestry or migration background. Source: Eurostat.
- Homicide is the only crime counted consistently enough for international comparison; other offence counts reflect reporting and policing practice as much as underlying crime.
- Citizenship is not ancestry or migration background, and prison populations reflect sentencing and remand policy as well as offending.
- Rates for national groups are not adjusted for age, sex or income unless a chart says so.
Definitions for every indicator are listed in the glossary, and collection methods in the methodology.
Sources & citation
- World Bank Open Data — World Development Indicators and population statistics. Licence: CC BY-4.0.
- Eurostat — European statistical office — crime, population and social statistics for EU/EFTA and candidate countries. Licence: Eurostat terms (CC BY 4.0).
birthrate.io, "Crime by origin and citizenship". https://www.birthrate.io/crime (accessed 15 November 2026). Underlying data: World Bank Open Data; Eurostat.
Please cite the original statistical agency for the underlying figures. Charts and tables on this page may be reused with attribution.