{"id":108253,"date":"2022-02-24T10:38:56","date_gmt":"2022-02-24T15:38:56","guid":{"rendered":"https:\/\/ncrc.org\/?p=108253"},"modified":"2022-07-08T14:42:53","modified_gmt":"2022-07-08T18:42:53","slug":"redlining-score","status":"publish","type":"post","link":"https:\/\/ncrc.org\/redlining-score\/","title":{"rendered":"Tracing the Legacy of Redlining: A New Method for Tracking the Origins of Housing Segregation"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t\t
<\/div>\n\t\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Tracing the Legacy of Redlining: <\/h1>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

A New Method for Tracking the Origins of Housing Segregation<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

February 2022<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Photo: <\/i>\u00a9Brian Jackson – stock.adobe.com<\/i><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Helen C.S. Meier<\/strong>, MPH, PhD., Assistant Research Scientist, University of Michigan<\/em>
Bruce C. Mitchell<\/strong>, PhD., Senior Researcher, NCRC<\/em><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\n\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\tCONTENTS<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

\u2014 Key Takeaways<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
\n\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Redlining was one form of federal policy which segregated communities and concentrated disadvantage. The maps are the most comprehensive existing archive of data available to tracing the historical legacy of residential segregation.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
\n\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Eighty years after the HOLC formally drew its redlining maps, those neighborhoods are still high minority population areas with the highest rates of vacancy in their metropolitan areas.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
\n\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Our synthesis of historic redlining scores from the HOLC with Census 2010 and 2020 tract boundaries provides a new tool for identifying the degree to which mid-20th century redlining practices continue to shape housing segregation and concentrated disadvantage today. This updates a\u00a02018 report<\/a>\u00a0and\u00a0offers a new dataset<\/a> for the historic redlining score, with the Census 2010 and 2020 tract boundaries.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t

\u2014 Key Findings<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Persisting segregation<\/strong>\u00a0– 8.25 million people live in neighborhoods identified by the federal government 80 years ago as \u201chazardous\u201d and redlined since that time, per 2020 Census data. More than three-quarters of these Americans identify as belonging to a minority group.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t

\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

The concentration of Black and Hispanic residents in redlined areas is high<\/strong>\u00a0– The average redlined neighborhood is predominantly Black (32%) and Hispanic (30%).<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t

\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Greater levels of vacancy or abandonment in redlined neighborhoods<\/strong>\u00a0– Levels of housing vacancy\u00a0are\u00a0much higher in \u201chazardous\u201d redlined areas at nearly 10%, contrasted with only about 6% in areas of cities that were not graded by the federal government. This suggests that there are greater levels of abandonment in redlined areas.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\n\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\tDownload Report<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t\n\t\n<\/svg>\t\t<\/div>\n\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Introduction<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

A fundamental element in the social dimension of US urban residential patterns is racial segregation (Massey & Denton 1998). One practice\u2014that of redlining\u2014was essential to establishing the racially segregated neighborhood structure which continues to persist in many cities (Rothstein 2017). During the 1930s through the postwar period the construction of homes across newly developing residential tracts, expansion of the federal highway system, and availability of low interest\/low down payment mortgages underwritten by the Federal Housing Administration (FHA) and the Department of Veterans Affairs (VA) contributed to rapid suburbanization of the White middle class (Jackson 1987). Redlining – the denial of access to financial services such as mortgage loans or insurance for people living in minority neighborhoods, was a common practice during this important era in the development of the US urban system (Aaronson et al. 2017). Redlining was government sanctioned. Mortgage loans insured by the FHA were subject to redlining practices until 1962 when President Kennedy banned the practice by federal agencies (Executive Order No.11,063, 1962). Private lenders continued to redline areas without penalty until the Fair Housing Act of 1968 made it illegal (Department of Housing and Urban Development n.d.).<\/p>

Redlining was one practice among several that created a system of inequitable access to credit, reinforcing economic and residential segregation (Shapiro 2004). While mortgage credit flowed to White families\u2014allowing them to purchase new homes and build equity\u2014families living in redlined areas were denied the same opportunity. People of color could not purchase properties in the newly developed suburbs, nor did they have sufficient access to credit to reinvest in their neighborhoods. The resulting low rates of homeownership are one factor that has continued to exacerbate the wealth divide between White and minority households (Bhutta\u00a0et al. 2020). Decades of discriminatory lending practices by the FHA and VA cemented the segregated residential pattern of structural racism in US urban areas.<\/p>

The most comprehensive source of maps delineating which neighborhoods would have been redlined are the Home Owners\u2019 Loan Corporation (HOLC) residential security maps (Nelson et al. n.d.). The HOLC maps were created between 1935 and 1940 and documented neighborhoods in over 200 cities. As they created these maps, HOLC examiners relied upon information gathered from local real estate and lending professionals to complete standardized forms that comprehensively evaluated not only the site and location, structural condition, and past valuations of properties, but also the social and demographic make-up of neighborhoods (Winling & Michney 2021). The examiners used a four-tiered grading system in order to classify neighborhoods: a grade of \u201cA\u201d was reserved for the \u201cbest\u201d neighborhoods, which were colored in green on the map\u2014while the lowest graded areas received a \u201cD,\u201d or \u201chazardous,\u201d rating and were shaded red. Neighborhoods with a population that was not of native-born European background were generally downgraded, and almost invariably a neighborhood with any African American presence received the lowest grade of \u201cD\u201d<\/p>

While the HOLC itself refinanced loans in both White and Black neighborhoods in the period 1933-1936, before the residential security maps were completed, the FHA did practice redlining (Hillier 2003). An\u00a0in depth\u00a0study of FHA lending patterns and the HOLC maps in Baltimore, Greensboro, and Peoria found that \u201chazardous\u201d graded areas received only 1.6%, 2.2%, and 1.2% respectively of all loans the FHA made in these cities between 1935 and 1940 (Fishback et al. 2021). High graded areas with \u201cBest\u201d and \u201cStill Desirable\u201d ratings received 67.3%, 53.1%, 73.9% of the FHA loans made during that period. Mortgage lending by the FHA favored new developments or neighborhoods that were graded highly on the HOLC maps and avoided \u201chazardous\u201d neighborhoods.<\/p>

Redlining denied access to financial services in neighborhoods, concentrating disadvantages and resulting in higher measures of social vulnerability today (Richardson et al. 2020). A key question in understanding the legacy of redlining in US neighborhoods is: Who lived in these neighborhoods in 2020? A spatial mismatch between historic neighborhood boundaries and current census tracts\u2014the level at which many sources of data (i.e., demographic, health, etc.) is available\u2014is a barrier to using HOLC maps in demographic research. To overcome this spatial mismatch and facilitate research on historic redlining policies, we used ArcGIS to create weighted redlining scores for 2010 and 2020 census tract boundaries, representing the degree of redlining within a tract. We then used the 2020 US Census Bureau data to determine:<\/p>

  1. How many people today live in the HOLC \u201cD,\u201d or \u201chazardous,\u201d graded areas?<\/li>
  2. What is the demographic profile of \u201chazardous\u201d areas contrasted with better graded \u201cbest\u201d areas?<\/li>
  3. How many housing units are in \u201chazardous\u201d areas?<\/li>
  4. What is the rate of vacancy for housing in \u201chazardous\u201d areas?<\/li>
  5. Are there significant differences in the race\/ethnic composition of \u201chazardous\u201d areas compared to non-hazardous areas?<\/li><\/ol>

    This new report updates our findings from the 2018 report and introduces an updated historic redlining score method which can be used with the Census 2020 tract boundary files.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

    \n\t\t\t\t\t\t
    \n\t\t\t\t\t
    \n\t\t\t
    \n\t\t\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t\t\t
    <\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t

    Methods<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t

    Creating a historic redlining score<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t

    Data from the residential security maps created by the HOLC\u2019s City Survey program is hosted by the University of Richmond Digital Scholarship Lab as part of their \u201cMapping Inequality: Redlining in New Deal America<\/em>\u201d project (Nelson et al. n.d.). These maps are available as\u00a0georectified\u00a0shapefiles, suitable for projection and use in geographic information systems (GIS).\u00a0We overlaid the HOLC maps with 2020 census tract boundary files for 146 metropolitan areas across the US using ArcGIS and determined the proportion of HOLC residential security grades contained within the boundaries. We assigned a numerical value to each HOLC risk category as follows: 1 for \u201cA\u201d grade, 2 for \u201cB\u201d grade, 3 for \u201cC\u201d grade, and 4 for \u201cD\u201d grade. We calculated a historic redlining score (HRS) from the summed proportion of HOLC residential security grades multiplied by the corresponding weighting factor based on area within each census tract. A higher HRS means greater redlining of the census tract, with a value of 4.0 corresponding to complete redlining (Figure 1).<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

    \n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"Method\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t

    Assessment of 2020 Census and HRS<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t

    The continuous HRS, assessing the degree of redlining, were linked to the available Census 2020 P1 file, with population data and H1, with data on housing (Census Bureau 2021). The census files are currently limited to demographic counts, with the housing data consisting of the number of units and their occupancy status. Data to assess the number of rentals and homeownership were not available. Next we analyzed the distribution of the HRS for both 2010 and 2020. The resulting continuous distributions were multimodal, with clustering at the 1.0, 2.0, 3.0, and 4.0 intervals. This clustering was the result of the influence of tracts with \u201cpure\u201d HOLC grading on the distributed values. Consequently, dividing the distribution into equal intervals with ranges of: 1.00-1.749; 1.75-2.49; 2.50-3.249; 3.25-4.00 allowed for an analysis of tracts ranging from the least to the most impacted by redlining. Tracts in metro areas not meeting the 20% threshold for HOLC-graded areas were assigned a \u201c0\u201d value and assessed separately as \u201cungraded\u201d tracts. Figure 2 provides an example of the HOLC grading and the resulting intervals for Detroit, Michigan.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

    \n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"Detroit\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t

    Descriptive statistics were calculated for the ungraded and the HOLC-graded tracts, as well as the intervals. Finally, associations between historic redlining score and 2020 demographics were assessed by calculating Pearson\u2019s correlation coefficient for binary associations and by an ANOVA to determine the statistical significance of the differences in mean values.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

    \n\t\t\t\t
    \n\t\t\t\t\t\t\t
    <\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t

    Results<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t

    Descriptive statistics for ungraded tracts and for the 13,424 tracts with an assigned HRS are presented in Table 1. Comparisons of the ungraded and HOLC-graded areas indicated that the ungraded areas had higher average population, higher percentage of non-Hispanic White, more housing units, and much lower rates of vacancy. The mean value of all 2020 census tracts with HOLC grading and a historic redlining score was 2.95, which interpolated to a \u201cC\u201d grade. On average, tracts with HOLC grades (downtown cores of metropolitan areas) had a higher proportion of minority residents (69%) than areas ungraded by HOLC (commercial areas and undeveloped areas or farm land).<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

    \n\t\t\t\t
    \n\t\t\t\t\t