What empirical studies show about sanctuary policies’ impact on crime rates and deportations since 2015?

1. What the strongest county‑level analyses find about deportations

The most-cited empirical work exploiting staggered rollouts of sanctuary policies finds a sizable fall in deportations of people fingerprinted by local law enforcement—roughly one‑third fewer deportations relative to counterfactual trends in treated counties during the 2010–2015 window—while deportations of people with violent convictions were not appreciably reduced in that period ^{[1] [2] [3] [6]}. Congressional summaries and briefings echo that finding: sanctuary policies were effective at limiting deportations of nonviolent offenders while not significantly changing the pace of deportations for violent offenders ^{[7] [6]}.

2. What the data say about crime rates

Across multiple designs and datasets, studies report no detectable increase in violent or property crime after adoption of sanctuary policies; several papers even find neutral or negative effects on crime, consistent with a growing “null or negative” pattern in the small empirical literature ^{[1] [8] [4]}. The National Academy of Sciences–style analysis that combines FBI Uniform Crime Reports with ICE deportation data finds no rise in crime following sanctuary adoption between 2010 and 2015 ^{[1] [3]}, and other reviews and syntheses reach similar conclusions that sanctuary policies do not threaten public safety ^{[9] [10]}.

3. Mechanisms, counterarguments and scholarly caveats

Scholars propose two competing mechanisms: critics argue lower local cooperation lowers the cost of offending and could attract criminals, while proponents point to a “trust spiral” where immigrant communities cooperate more with police—raising reporting and community cohesion and thereby reducing crime ^{[5] [11]}. Methodological caveats recur in the literature: selection into sanctuary status (cities that adopt these policies differ in observable and unobservable ways), limited crime‑type coverage (most work focuses on homicide, robbery, property crime), and temporal windows concentrated around 2010–2015 limit generalizability to other periods and newer policy designs ^{[5] [4] [6]}.

4. Where consensus ends and uncertainty begins

The consensus from rigorous analyses through the mid‑2010s is clear that sanctuary policies reduce certain deportations and do not increase aggregate crime rates; nonetheless, researchers warn that few studies cover long post‑adoption horizons, heterogeneity across locales and crimes remains underexplored, and selection and measurement issues could moderate effects in specific places or times ^{[1] [4] [5]}. Policy advocates and local officials point to community safety and reporting benefits as intentional design goals and empirical support ^{[9] [10]}, while critics and some policy commentaries continue to assert risks—an assertion the empirical work has not borne out at scale in the datasets analyzed to date ^{[11] [7]}.

5. Bottom line for policymakers and journalists

The best-identified empirical evidence available through studies that analyze policy rollouts between roughly 2010–2015 shows sanctuary policies materially reduce deportations of nonviolent arrestees but do not increase crime rates—and may in some contexts strengthen the association between immigrant presence and lower crime—yet further research is required to test newer policy variants, to address selection bias more fully, and to break effects down by crime type and later time periods ^{[1] [8] [5] [6]}.