Mortgage Redlining, Homeownership and Implications for Social, Physical and Economic Health

Purpose

The aim of this study was to examine the relationship between a census tract’s status under federal racially-based lending restrictions before the 1950s and the level of home mortgage lending in that neighborhood in 2012 in an urban area in the U.S.

Background

Homeownership remains the primary method by which people build wealth (Rappaport, 2010; Fontes & Kelly, 2013).   However, before the 1950s, federal policies restricted the federal government from insuring home loans in predominantly racial/ethnic minority areas.  Each area received a grade from A to D, and people in areas with a grade of D could not obtain loans.  This practice, known as redlining, has resulted in continued residential segregation and the proliferation of urban poverty (Crossney & Bartelt, 2005). 

Not only does homeownership help people build wealth, but it has a significant effect on people’s health and longevity.  Recent studies show that people who own their own homes have a lower incidence of strokes (Jackson, Jones & Mishra, 2014) and a greater sense of control over their lives, which can lead to better physical and mental health outcomes (Manturuk, 2012).  Further, evidence suggests that the longer someone lives in a home that they own, the less likely they are to need to receive institutional care as they age (Rouwendal & Thomese, 2013). 

Hypothesis

A census tract's  mortgage lending grade prior to the 1950s is significantly associated with modern-day mortgage lending levels in that census tract.

Study Design

I used a random sample of mortgage loans from 2012 Home Mortgage Disclosure Act (HMDA) data to determine the number of purchase and refinance loans made in a census tract (n=738) using 2010 US Census Data to determine the racial/ethnic composition, unemployment rate, homeownership rate, average educational attainment and median income of the census tract.  I also used spatially referenced data in ArcGIS to determine the grade of the modern-day census tract.

After prescreening my data for necessary assumptions, I created an OLS regression model to test the relationship between a census tract’s grade from the federal government and the number of purchase and refinance mortgage loans made in that census tract in 2012.

Results

Controlling for census tract's predominant race/ethnicity, homeownership rate, unemployment rate, income level and educational attainment, I found that a census tract’s grade had a significant effect (unstandardized B=-21.904, p<0.001; R²=0.637) on the number of mortgage loans made in that census tract in 2012.  Specifically, the worse a grade for a tract was, the fewer mortgage loans were made in that tract in 2012.  This effect indicates that researchers should consider the history of mortgage lending in an area when studying current issues of social and economic inequality.

Implications

This study provided evidence that historical redlining still impacts mortgage lending in modern times.  Since homeownership can be an essential part of positive social and health outcomes, researchers and practitioners should examine potential historic institutional barriers to homeownership in their communities to better understand the communities in which their clients live when planning interventions to facilitate long, healthy lives for clients.