TY - JOUR
T1 - Dealing with guns
T2 - investigating the spatial attraction of gun dealers on localized gun violence
AU - Griffiths, Elizabeth
AU - Xu, Jie
AU - Xie, Sandy
AU - Stansfield, Richard
AU - Semenza, Daniel
N1 - Publisher Copyright: © The Author(s) 2025.
PY - 2025
Y1 - 2025
N2 - Objectives: Test the spatial relationship between the locations of federally licensed firearms dealers (FFLs) and gun violence using shuttered FFLs as a counterfactual. Methods: Using a quasi-experimental design, we evaluate the proximity of gun violence to (a) dealers and (b) the spatial footprints of former dealers. The locations of violence incidents between 2019 and 2021 in Los Angeles, CA, Dallas, TX, and Philadelphia, PA are analyzed against dealer locations using street Network Cross K methods. Results: There is a robust attractive effect of dealers on gun violence within 3000 feet in two of the three cities, largely driven by the effect of dealers on violence in disadvantaged areas. The locations of former gun dealers did not uniformly fail to attract gun violence, however, as would be expected if FFLs were causally associated with the distribution of gun violence in cities. Conclusions: The findings on the spatial attraction of gun violence to the locations of former gun dealers are heterogeneous across sites, leading to mixed support for a causal spatial association between gun dealers and gun violence. Limitations of using former dealers as the counterfactual include defining former dealers and the small numbers of former dealers in cities.
AB - Objectives: Test the spatial relationship between the locations of federally licensed firearms dealers (FFLs) and gun violence using shuttered FFLs as a counterfactual. Methods: Using a quasi-experimental design, we evaluate the proximity of gun violence to (a) dealers and (b) the spatial footprints of former dealers. The locations of violence incidents between 2019 and 2021 in Los Angeles, CA, Dallas, TX, and Philadelphia, PA are analyzed against dealer locations using street Network Cross K methods. Results: There is a robust attractive effect of dealers on gun violence within 3000 feet in two of the three cities, largely driven by the effect of dealers on violence in disadvantaged areas. The locations of former gun dealers did not uniformly fail to attract gun violence, however, as would be expected if FFLs were causally associated with the distribution of gun violence in cities. Conclusions: The findings on the spatial attraction of gun violence to the locations of former gun dealers are heterogeneous across sites, leading to mixed support for a causal spatial association between gun dealers and gun violence. Limitations of using former dealers as the counterfactual include defining former dealers and the small numbers of former dealers in cities.
KW - Causal inference
KW - Counterfactuals
KW - Gun dealers
KW - Gun violence
KW - Place-based features
UR - http://www.scopus.com/inward/record.url?scp=85218684884&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85218684884&partnerID=8YFLogxK
U2 - 10.1007/s11292-025-09666-9
DO - 10.1007/s11292-025-09666-9
M3 - Article
SN - 1573-3750
JO - Journal of Experimental Criminology
JF - Journal of Experimental Criminology
ER -