Identifying high-risk firearms dealers

Identifying high-risk firearms dealers

A machine learning study of rapidly diverted firearm sales in CA

  • by Laqueur HS, Smirniotis C

Quick Summary

  • Firearms dealers can facilitate the diversion of guns to the criminal market through practices such as selling to straw purchasers or failing to conduct required background checks.

About the Paper

FINDINGS IN BRIEF

A visual abstract for this paper is available in our visual abstract archive.
  • Many highest-risk dealers consistently sold crime guns across multiple years.
  • Our machine learning models accurately identified many high-risk firearms dealers and outperformed simpler prediction approaches.

IMPLICATIONS AND CONCLUSIONS

  • Using machine learning with detailed purchase and recovery records could efficiently identify high-risk retailers for targeted enforcement to disrupt the flow of crime guns.
  • Opportunities for policy and enforcement action based on our findings include targeting enforcement on the small number of dealers that repeatedly sell numerous short time-to-crime guns and enhancing reporting or licensing requirements and oversight of pawn dealers, as pawnbrokers are more likely to sell crime guns.
  • Future research should validate risk predictions with inspection data and assess whether high counts or high proportions better signal illegal activity.

METHODS

With firearm purchase and crime gun recovery data from California (2010-2021), we trained machine learning models to identify dealers who sold the largest number and highest fraction of guns recovered in crimes within 1 year of sale. Both are well-established indicators of potential illegal activity by dealers or traffickers.

Citation

Laqueur HS, Smirniotis C. Identifying high‐risk firearms dealers: A machine learning study of rapidly diverted firearm sales in California. Criminology & Public Policy. 2025 Aug;24(3):363-403.

Learn More

Primary Category

Firearm Acquisition and Ownership

Tags

Research