Homelessness Is a Housing and Racism Problem: Measuring Structural Racism and Its Association with Homelessness Disparities

Background and Purpose: In the United States, Black people experience disparate rates of homelessness. Attention to scholarship on race and inequality has positioned researchers studying racial disparities in homelessness to understand these inequities as caused by historical and ongoing racism, discrimination, and exclusion across systems—including land use and housing, wealth, income and employment, and criminal justice. Moreover, it positions us to understand homelessness as fueled by structural racism—the interconnectedness across these systems, as well as their historical and cultural roots. This study explores geographic variation in Black to White disparities in homelessness rates and examines how local measures of structural racism are associated with these disparities.

Methods: To conduct the analyses in this study, I used area-weighting techniques to construct a unique dataset of 260 urban and suburban communities—with estimates of homelessness from HUD’s Point-in-Time count (2016-2020), housing, economic, and demographic characteristics from the American Community Survey (2015-2019), and a multidimensional measure of structural racism that draws on data from multiple domains of Black to White disparities (Siegel et al., 2022). I use R to generate descriptive statistics at the community-level and conduct linear regression (with logged dependent variables and robust standard errors clustered at the state level) to examine the associations between of community-level characteristics and community-level homelessness disparities.

Results: For all total homelessness, Black rates are 5 times higher than White rates, and the family disparity is nearly 11. Northeast and Midwest areas have the highest Black to White disparities in homelessness. Black to White disparities in homelessness are positively associated with a measure of structural racism that accounts for housing, economic, education, and criminal justice inequities as well as latent interactions between these domains (the structural racism factor score on its own explains one third of the variance in disparities). In multivariate models adjusting for covariates, disparities in homelessness were also uniquely associated with Black to White inequities in criminal justice (the ratio of the percent of Black people in jail/detention to the percent of White people in jail/detention), housing (renter share ratio), and economic disparity (poverty rate ratio). In models using standardized z-score coefficients, the strongest predictor was the jailing disparity.

Conclusions and Implications: I discuss the findings in the context of racial discrimination across multiple systems and argue that efforts to address homelessness will fall short without attention to redressing structural racism. In the presentation, I will discuss how I have engaged with practice and policy networks to share my findings and inform policy.