Redlining, Reinvestment, and Racial Segregation: A Bayesian Spatial Analysis of Mortgage Lending Trajectories and Firearm-Related Violence

Background : In the United States, firearm-related violence disproportionately impacts low-income, racially segregated communities more than affluent, predominantly White neighborhoods. This trend stems from historical disinvestment, discriminatory lending practices, and persistent structural inequalities. Housing policies have enforced racial segregation, limiting wealth accumulation in low-income communities. This study examines the relationship between historical and contemporary lending discrimination in mortgage originations and firearm-related violence in Chicago, Illinois. By analyzing investment and disinvestment patterns, we assess how housing discrimination continues to influence the risk of victimization in various social contexts. Methods : Redlining scores were derived from the 1930s Homeowners’ Loan Corporation (HOLC) grades, while contemporary lending indicators were obtained from the 2019 Home Mortgage Disclosure Act (HMDA). We classified neighborhoods into four lending trajectories—sustained disinvestment, disinvestment, growing investment, and high investment—based on historical redlining and contemporary mortgage lending patterns. Sustained disinvestment reflects historical redlining and ongoing lending discrimination, while growing investment targets areas that were historically redlined but are now experiencing increased lending activity. Bayesian spatial models examined firearm-related homicide risk across lending trajectories, adjusting for area deprivation index (ADI) and racial segregation. Results : In unadjusted models, sustained disinvestment (Relative Risk [RR] = 2.230, 95% CrI: [1.352, 3.681]) was associated with increased firearm-related homicide risk, while growing investment (RR = 0.782, 95% CrI: [0.452, 1.359]) and high investment (RR = 0.146, 95% CrI: [0.054, 0.397]) were associated with lower risk. After adjusting for ADI and racial segregation, the effect of sustained disinvestment attenuated (RR = 1.714, 95% CrI: [1.054, 2.791]), suggesting partial mediation. However, growing investment increased by 155% (RR = 1.987, 95% CrI: [1.144, 3.458]), indicating suppression, indicating that ADI and segregation initially masked its association with firearm homicide risk. Conclusion : Findings highlight the need for policies that address the long-term effects of lending discrimination. Reverse redlining—where financial institutions target minority communities with high-cost loans—further exacerbates existing inequities. Additionally, neighborhood deprivation and segregation shape firearm-related violence risk, underscoring the broader consequences of systemic housing discrimination.