Analysis · Brå 2021:9 as primary source
Foreign background overrepresentation in serious crime
Brå's register study 2021:9 is the most recent comprehensive review of crime suspicions by background category. Figures are standardised for age and sex. The baseline is native-born with two native-born parents = 1.0×.
How to read the headline figures
The Brå figures concern people registered as suspects in 2007–2018, not convictions. Ratios use native-born people with two native-born parents as the reference and adjust for age and sex. Education, income and neighbourhood are not included in this first adjustment; group differences do not establish causation.
Fatal shootings use a separate research-based estimate, and 85–90 percent is a share of suspects rather than an over-risk. Measures with different definitions should not be treated as one series.
×2.5
Murder & manslaughter — foreign-born
Source: Brå 2021:9
×5.0
Rape — foreign-born
Source: Brå 2021:9
×3.2
Robbery — foreign-born
Source: Brå 2021:9
85–90 %
Gang murders — suspects with foreign background (estimate)
Source: Swedish Police NOA · Sturup et al.
Chart 1 — Overrisk by offence
Overrisk relative to native background (= 1.0×). Source: Brå 2021:9 with fatal shootings from Sturup et al. and the Swedish Police lägesbild.
Murder, manslaughter, fatal assault
Source: Brå 2021:9, kap. 4 (mord/dråp) ↗Native background×1,0One foreign parent×1,4Two foreign parents (2nd gen.)×3,2Foreign-born×2,5- Native background×1,0One foreign parent×1,5Two foreign parents (2nd gen.)×3,3Foreign-born×5,0
- Native background×1,0One foreign parent×1,7Two foreign parents (2nd gen.)×3,3Foreign-born×3,2
- Native background×1,0One foreign parent×1,5Two foreign parents (2nd gen.)×2,4Foreign-born×1,9
Fatal shootings — estimate(estimate)
Source: Sturup m.fl. (2019) · Polisens NOA lägesbild (forskningsbaserad skattning) ↗Native background×1,0One foreign parent×2,0Two foreign parents (2nd gen.)×5,4Foreign-born×4,1
Switch definition — what counts as "foreign background"?
Sweden's definition is narrower than many neighbours'. Under Denmark's definition the share of serious-violence suspects with "foreign background" would be higher — not because the crimes change, but because more people are counted.
Change country's definition
The share of suspects classified as 'foreign background' depends on how the term is defined. Only the Swedish Brå definition has published shares — other countries' definitions are never recomputed on Swedish data.
Murder & manslaughter
60%
Rape
58%
Sverige 🇸🇪: Official SCB/Brå definition — foreign-born OR Swedish-born with both parents foreign-born.
Region breakdown — foreign-born
Overrisk for murder/manslaughter among foreign-born by region of origin, relative to native background. Source: Brå 2021:9, table 5.4.
- Nordic (excl. Sweden)×1.5
- EU-15 (excl. Nordic)×0.9
- Eastern Europe (incl. Balkans)×2.1
- MENA×3.0
- Sub-Saharan Africa×3.9
- Asia (other)×1.1
Time series — murder & manslaughter 2002–2018
Share of suspects with foreign background (SE definition). The series ends in 2018 — Brå has not published a comparable figure for the years after. Source shown per data point.
- 200245%Brå 2005:17
- 200548%Brå 2005:17
- 201052%Brå 2011:21
- 201556%Brå 2011:21 (framskrivning) + Brå 2021:9
- 201858%Brå 2021:9
Method & sources
- • Primary source: Brå 2021:9. Register data for the entire population 2007–2018.
- • Standardisation: the Brå ratios adjust for age and sex. This step does not control for education, income or neighbourhood.
- • Baseline: native-born with two native-born parents = 1.0×.
- • Fatal shootings: Sturup et al. (2019) and Swedish Police NOA 2024. Brå has not published a separate overrisk for fatal shootings — the research-based estimate is shown here.
- • Modelled figures: clearly marked in the definition switcher. Foreign definitions applied to Swedish Brå data.
- • Dark figures: suspicion statistics reflect reports and policing — not all crimes. Read overrisks as direction and magnitude, not exact numbers.
Concrete cases behind the numbers
Court case archive
The statistics consist of thousands of individual cases. We document the most notable — final convictions only.
To the archive