{
  "schema_version": "immigrationdb-research-article-v1",
  "slug": "unemployment-recorded-crime",
  "url": "https://immigrationdb.com/research/unemployment-recorded-crime/",
  "title": "Does Higher Unemployment Mean More Recorded Crime?",
  "headline": "The unemployment–crime link changes sharply by country",
  "question": "Inside a country, do cities with higher unemployment also record more offences per 100,000 residents?",
  "summary": "Does unemployment cause crime? City-level data cannot answer causation: within 11 countries, unemployment–crime correlations range from 0.004 to 0.683.",
  "published": "2026-08-06",
  "updated": "2026-08-06",
  "author": "Maksim Streltcov",
  "license": "https://immigrationdb.com/legal/terms/",
  "sections": [
    "What we measured",
    "What the data shows",
    "Where it breaks down",
    "What this means if you are moving",
    "Limits",
    "Method",
    "Sources"
  ],
  "faq": [
    {
      "q": "Does unemployment cause crime?",
      "a": "This analysis cannot answer causation. It compares city-level unemployment and recorded-crime ranks at one point or period, without measuring policing, reporting, deprivation, urban form or other possible influences. It also cannot say anything about an individual unemployed person’s behaviour; doing so would be the ecological fallacy."
    },
    {
      "q": "Where is the unemployment–crime relationship strongest?",
      "a": "The strongest reportable Spearman coefficients are Sweden at +0.683, the Netherlands at +0.647 and Belgium at +0.646. Denmark follows at +0.421. These are within-country rankings under each country’s own unemployment and recorded-crime definitions, not evidence that their national levels are higher or directly comparable."
    },
    {
      "q": "Where is the coefficient closest to zero?",
      "a": "The United Kingdom is +0.004 across 25 places, Spain +0.013 across 425, Canada +0.021 across 37, Lithuania +0.029 across 30 and Portugal +0.044 across 280. These are sample coefficients near zero, not proof that the underlying relationship is exactly zero."
    },
    {
      "q": "Does city population explain the unemployment–crime result?",
      "a": "No single size explanation fits all countries. Lithuania has a +0.638 population–crime coefficient but only +0.029 for unemployment; Belgium and the Netherlands are positive on both; Portugal is slightly positive for unemployment and negative for population. Population is a competing correlate, not a demonstrated mechanism."
    },
    {
      "q": "How should movers use these figures?",
      "a": "Use the pairing only as a country-specific screening clue. Check each shortlisted city’s observation year, geographic grain, offence definition and unemployment definition, then compare places covered by the same source system. Direct neighbourhood, offence-category and labour-market evidence is more useful than assuming one national coefficient applies everywhere."
    }
  ],
  "countries": [
    "belgium",
    "canada",
    "denmark",
    "germany",
    "lithuania",
    "netherlands",
    "portugal",
    "spain",
    "sweden",
    "united-kingdom",
    "united-states"
  ],
  "answer": "City-level data cannot show that unemployment causes crime. Across 7,310 places in the 11 country samples with at least 20 observations, every Spearman coefficient is positive, but the range runs from +0.004 in the United Kingdom to +0.683 in Sweden . Spain , Canada, Lithuania and Portugal are also near zero. A shared sign is not a shared relationship."
}