“After accounting for age, sex, location, and socioeconomic conditions, foreign origin does not independently predict crime and the apparent crime gap disappears.”
@library_of.alexandriaReel analyzed by Clear
“Good caution about causation, but the zero-gap claim goes too far.”
Police statistics count suspects, not convictions, and raw comparisons can be distorted by age, sex, residence status, non-resident suspects, policing, and where people live. Germany’s ifo analysis found no statistical relationship between changes in a district’s foreigner share and its local crime rate from 2018 through 2023, including for violent crime. ([ifo.de](https://www.ifo.de/en/press-release/2025-02-18/more-foreigners-do-not-increase-germanys-crime-rate?utm_source=openai))
The post treats adjustment as making the relevant gap zero. Denmark’s 2025 official report does not show that: after adjustment for age and socioeconomic background, male MENAPT immigrants had an index of 177 and male MENAPT descendants about 270 for all convictions in 2023. Its age-adjusted violent-crime indexes also remained elevated. ([integrationsbarometer.dk](https://integrationsbarometer.dk/tal-og-analyser/filer-tal-og-analyser/arkiv/copy_of_NotatvedrrendekriminalitetblandtindvandrereogefterkommerefraMENAPTlandene.pdf/at_download/file)) Germany’s 2025 police data likewise show higher recorded violent-crime suspect rates for some resident nationality groups even after age and sex are tabulated; those figures still cannot establish cause. ([dserver.bundestag.de](https://dserver.bundestag.de/btd/21/060/2106098.pdf))
This is material because a reasonable viewer could leave believing that properly adjusted evidence shows no remaining group differences at all. The cited Danish evidence instead finds remaining differences, while not establishing that origin itself is the cause.
Why Clear says this
The central causal caution has a sound basis: aggregate overrepresentation is not proof that migration raises crime, and the German district study supports that limited conclusion. But the post uses that caution to imply that social status fully explains observed differences. Its Denmark example directly fails to support that stronger framing, because sizable adjusted differences remain. The appropriate conclusion is narrower: causation cannot be inferred from raw group rates, and the size and causes of residual differences remain unresolved.
Evidence
- Germany’s PKS 2025 recorded about 5.5 million offenses and reported that non-German suspects remained overrepresented in violent crime; these are recorded suspects, not proof of an innate or causal group propensity. ([bundesregierung.de](https://www.bundesregierung.de/breg-de/service/newsletter-und-abos/newsletter-verbraucherschutz/polizeiliche-kriminalstatistik-2025-2422020))
- The ifo district analysis excluded immigration-law offenses and found no statistical correlation between increases in a district’s foreigner or refugee share and local crime during 2018–2023, but it also notes limits from crime-location data, non-resident suspects, repeat counting across districts, and unavailable linked individual data. ([ifo.de](https://www.ifo.de/en/press-release/2025-02-18/more-foreigners-do-not-increase-germanys-crime-rate?utm_source=openai))
- Denmark’s official report uses convictions and provides both age-adjusted and age-plus-socioeconomic indexes. It reports that adjusted differences persist rather than disappear. ([integrationsbarometer.dk](https://integrationsbarometer.dk/tal-og-analyser/filer-tal-og-analyser/arkiv/copy_of_NotatvedrrendekriminalitetblandtindvandrereogefterkommerefraMENAPTlandene.pdf/at_download/file))