Predictors of ATF inspections of FFLs and subsequent violations.

Background: The Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) conducts inspections of Federal Firearms Licensees (FFLs) to ensure compliance with federal laws in an effort to keep the public safe from firearm violence. However, the criteria for which FFLs are inspected lacks transparency, potentially signalling inefficiencies or biases.

Methods: We extracted data on FFL inspections and violations (2016-2017) requiring remedial action from the Gun Store Transparency Project database and the ATF's FFLs listings. Using a generalised linear mixed model with a random effect for states and robust SE, we estimated the census tract and state level predictors of 1) the likelihood of the ATF inspecting an FFL and 2) among those inspected, the likelihood of a violation.

Results: Of the 93,420 FFLs, 22 921 were inspected and 2392 had violations requiring remedial action. Increased likelihood of inspection was statistically significantly associated with higher levels of poverty, lower median income, greater non-White population percentages, and older census tract median ages. The sole predictor of an FFL receiving a violation requiring remedial action was the state's firearm law permissiveness.

Discussion: ATF inspections disproportionately targeted communities with higher poverty, lower median income, larger non-White populations, and older demographics. This indicates potential racial and socioeconomic biases, diverting attention from jurisdictions with more lenient firearm laws where violations were more likely to occur.

Conclusion: The ATF needs to reassess the influence of potential biases on its strategies for selecting FFLs for inspections and instead focus on risk-based assessments in order to promote public safety.