What Swedish register‑based studies compare crime rates by immigrants' country or region of origin and adjust for socio‑economic factors?

1. Brå’s canonical 2005 register analysis and its socioeconomic adjustment

The Swedish Crime Prevention Agency (Brå) 2005 report remains the reference point: it used register data to show over-representation of people with foreign background in criminal statistics but explicitly demonstrated that taking account of socioeconomic variables—age, gender, education and income—substantially reduces the crime-rate gap between immigrants and natives ^{[1] [2]}.

2. Long-run, register-based conviction trends (1973–2017) that stratify by origin and social background

A comprehensive register study of convictions from 1973 to 2017 presents results by immigrant background and gender and uses parental income as a socioeconomic indicator; it reports that overrepresentation persists but that levels have fallen for some groups in recent decades and that socioeconomic background is an important explanatory variable in multivariate analyses ^{[3] [5]}.

3. Studies that compare country/region of origin and test socioeconomic explanations

Nordic cross-country and Swedish-origin subgroup studies show heterogeneity by origin: immigrants from Western Europe, North America and Oceania generally fare better socioeconomically and have lower crime rates than some other origin groups, and a small set of Swedish/Nordic register studies explicitly adjust for employment, income and education and find that those adjustments reduce measured over-representation considerably ^{[6] [7]}.

4. Self‑report time‑trend work and the limits of register-only inference

Time‑trend analyses of self‑reported offending (1999–2017) document declining youth crime across backgrounds but are primarily descriptive and do not adjust for explanatory socioeconomic variables, underscoring the methodological boundary between register-based causal inference and descriptive trend work ^{[8] [9]}.

5. Recent municipal and meso‑level register studies that reframe origin prevalence

Newer municipal‑level register studies link changes in violent crime to local correlates and immigrant-origin prevalence, finding that municipalities with rising violence also show other social and economic correlates and that immigrant prevalence alone does not explain increases; these analyses emphasize population prevalence of origin groups and local socioeconomic context rather than a simple origin→crime causal story ^{[10] [4] [11]}.

6. What adjustments do— and do not—resolve: consensus, contested points, and political uses

Across the literature the recurring finding is that socioeconomic and demographic controls reduce but do not always eliminate overrepresentation by immigrant-origin groups; some register studies still report roughly twofold higher conviction risk for persons of foreign background even after controls, and authors stress heterogeneous mechanisms (segregation, family-level factors, labour market exclusion) and the need for more granular causal work ^{[5] [2] [7]}. Political debates in Sweden have amplified descriptive statistics (crime, origin) in election cycles and media coverage, which can obscure methodological caveats and the difference between correlation and causation ^[1].

7. Gaps, methodological caveats and next steps for researchers and readers

Several reviewers and research pieces note that no single register study answers all causal questions—whether pre-migration factors, selection, or post-migration socioeconomic disadvantage drive differences—and call for more fine-grained register work combining origin, parental background, housing segregation and exact socioeconomic trajectories; some recent work explicitly warns that prevalence of origin groups must be modelled at municipality and family levels to avoid ecological misinterpretation ^{[12] [11] [7]}.