Lawful Handgun Licensing, Population Density, Poverty, Police Staffing, And Personal Violent Crime: A Two-Cohort Comparison of U.S. Jurisdictions (2023-2024)

Objective. Prior research on gun ownership and violent crime has yielded inconsistent conclusions, largely because nearly all studies rely on proxy measures of prevalence - surveys, state-level aggregates, or policy classifications - that cannot isolate handgun ownership, cannot confirm lawful versus illicit possession, and are vulnerable to confounding. For example, Kleck (1997) and Cook & Ludwig (2006) highlighted measurement error as the main barrier to inference, Siegel et al. (2014) reported positive associations using household survey proxies, and Donohue et al. (2019) found liberalized right-to-carry regimes associated with higher violent crime, but with treatment defined as legal regime type rather than actual prevalence. As the National Research Council concluded in its landmark 2004 review, available studies were too compromised by proxy limitations and unmeasured confounders to provide reliable evidence of causal effects.Design. This study is the first to eliminate that uncertainty by restricting analysis to U.S. jurisdictions that publish auditable, administrative counts of active resident handgun licenses. We constructed two cohorts: A (higher licensing prevalence, mean ≈ 10.7% of residents) and B (lower prevalence, ≤1.6%). We then compared two-year (2023-2024) totals for homicide, rape, robbery, and aggravated assault per 100,000 residents, with covariates for poverty (%), sworn law-enforcement staffing (LEO per 100,000) and population density. We estimated cohort means and rate ratios, Pearson correlations with Fisher-z intervals, OLS slopes of violent rate on licensing share, and Poisson GLMs with log(population) offsets to report percent change per +1-percentage point increase in licensed share.Results. Mean violent crime was ≈ 542 per 100,000 in Cohort A versus ≈ 1,238 per 100,000 in Cohort B (rate ratio ≈ 2.28, 95% CI 1.52-3.43). Each +1-percentage point increase in licensing share corresponded to ≈ 7.6% lower violent incidence (IRR 0.924; 95% CI 0.890-0.958), or ≈ 60-70 fewer incidents per 100,000 depending on model specification. Adjusting for poverty or LEO staffing attenuated but did not eliminate the association. Population density was negatively correlated. Sensitivity analyses excluding outliers (Flint, MI and Washington, DC) yielded consistent results.Conclusions. This study is the first to demonstrate, using clean administrative licensing exposure measures, that higher lawful handgun ownership prevalence is strongly and inversely associated with personal violent crime. The association is robust across models and covariates, survives outlier tests, and directly addresses the measurement weaknesses that have limited prior scholarship. Unlike advocacy-sponsored work, this study is independent, fully transparent, and reproducible, providing the strongest evidence to date that lawful handgun prevalence aligns with substantially lower violent crime rates. We emphasize correlation rather than causation, highlight demographic, lifestyle and illicit-market limitations, and provide a reproducibility appendix with administrative URLs to support independent audit.