Comprehending the link between historic policies and Cities

Connecting current environmental and demographic inequalities to 1930s redlining policies

Kasturi Acharya & Himali Chingale, Fall 2022

States where redlining effect is prevalent in the United States

The Redlining effect

A variety of political and cultural factors over the past century have drawn distinct boundaries between racial groups in the United States. These boundaries, which effectively restricted people from different demographic groups to specific areas of the city, were created by racist attitudes toward Black citizens and non-Whites and exhibited through urban planning, housing rules, discriminatory banking, and other activities. These trends were widely masked under the well-known policies of HOLC, the New Deal era, and the Federal Housing Association and were especially prevalent in the following cities.

One of them is Chicago, where racial and ethnic segregation has occurred along neighborhood lines and in accordance with the physical characteristics of the built and natural environments.

Second, in the early twentieth century, San Francisco's central and southeast neighborhoods were redlined, or labeled as high risk, making its residents less likely to be approved for government-backed mortgage loans than residents of other districts.

Background

1930 Home Owners Loan Association (HOLC)

This policy aimed at providing financial assistance to low and moderate-income households through which it embodied the idea of homogenous neighborhoods in the United States.

1934 - HOLC as Federal Housing Act

The HOLC under FHA started issuing federally insured home loans to middle income Americans.

1951 - HOLC ceases operation

The HOLC collected mortgage payments between 1936 and 1951, closed its doors, and gave the U.S. Treasury a tiny excess as payment

1968 - Fair Housing Act

Prohibited racial discrimination pertaining to the sale or rental of property or obtaining a mortgage

1977 community Reinvestment

Established with the goal to lessen some of the effects of unfair lending practices.

Present day

Basis of the hypothesis of this study

Understanding the HOLC Grading

Established in the 1930s to the 1960s, redlining prevented African Americans from obtaining homes, home loans, and home repairs, and it is still a significant factor in the wealth gap between black and white families. Between 1935 and 1940, the federal agency Home Owners' Loan Corporation used Redlining to legitimize racial discrimination across the United States.

The agency created color-coded graded maps for American cities, indicating risk levels for long-term real estate investment or mortgage security. The HOLC included these maps as "residential security maps" in the FHA Underwriting Handbook to help the government decide which neighborhoods would make good investments and which should be avoided when issuing mortgages.

Grade A (Best)- Green-colored areas, representing in-demand, up-and-coming neighborhoods populated by "professional men" or "homogeneous group".

Grade B- (Blue) These areas had "reached their peak," but they were thought to be stable due to the low risk of "infiltration."

Grade C- (Yellow)- generally communities bordered by black neighborhood with a risk of infiltration

Grade D- (Red) Already infiltrated areas.

The Issue

The consequences of redlining extend beyond the individual families who were denied loans because of the racial makeup of their neighborhoods. Many neighborhoods labeled "Yellow" or "Red" by the HOLC in the 1930s are still underdeveloped and underserved in comparison to nearby "Green" and "Blue" neighborhoods with predominantly White populations. Blocks in these areas are often empty or lined with vacant buildings. They frequently lack basic services such as banking and healthcare, as well as fewer job opportunities and transportation options.

Hypothesis

The Arc GIS analysis here attempts to impersonate the effects of the 75-year-old HOLC policy on the modern-day cities of San Francisco and Chicago by understanding the relationship of the following parameters - urban heat island effect, Toxic Release Inventory, Total Population, Median Household Income, Racial Diversity, and Asthma prevalence index in relation to the HOLC census tracts of the two cities by means of the "hotspot analysis " and "overlay analysis". The comparison would provide us with a visual and analytical reference as to whether or not these cities in the United States continue to bear the burden of this historic policy even after 50 years of its repeal.

Data Sources

The data for this project comprised shapefiles and tabular data sourced primarily from the city data portal, living atlas and EPA site stated as follows:

  • Shapefile
  • Tabular and Point data

Toxic Release Inventory data for the United States -  https://www.epa.gov/toxics-release-inventory-tri-program 

Methodology

The following chart explains the details steps of the analysis conducted for the above stated parameter followed by explaining the three main steps of the whole analysis.

Methodology flowchart created on Miro

Raster Function

Raster functions apply processing directly to the pixels of imagery and raster datasets, as opposed to geoprocessing tools, which create a new raster on disk. while Functions can be combined to create function chains, which can then be saved as raster function templates using the Function Editor. Raster function templates can be used and distributed in a variety of ways. In our project, we created a raster function for Landsat imagery to merge the temperature bands and create an urban heat island effect

Enrich Tool

Enriches your data by adding demographic and landscape facts about the people and places that surround or are inside your data locations. In our study we used this tool to get the total population data, median income and racial diversity index for the city of Chicago and San Francisco

Optimize Hot Spot Analysis

This tool identifies statistically significant spatial clusters of high and low values (hot spots) (cold spots). It automatically aggregates incident data, determines an appropriate scale of analysis, and accounts for multiple testing and spatial dependence. We used this tool to compare the spatial concentration of different parameters in both cities to the HOLC redlining index.

Standardize field

Converts field values to values on a specified scale to standardize them. The z-score, minimum-maximum, absolute maximum, and robust standardization are all methods of standardization. Before performing the overlay analysis, we used this tool to standardize the values of the Z min and max score of the hot spot analysis, as well as the redlining percentage.

Analysis

Based on the background study and analyzing the jarring effects of Governmental policies this study focuses on assessing the effects of historical redlining on minority demographics in teh cities of Chicago and San Francisco

Chicago | HOLC Boundaries

Chicago is 60,000 hectares in size and is located on Lake Michigan's southwestern shore ( Fifth largest water body in the world) The city is bisected by the Chicago and Calumet rivers. Chicago, as a multicultural city that thrives on the harmony and diversity of its neighborhoods, embodies and reflects the values of America's heartland—integrity, hard work, and community—in the social fabric of its 77 distinct neighborhoods. Chicago is a leader in many public and policy arenas, but it is hampered by the historic policies of its past.

San Francisco | HOLC

The San Francisco County Redlining Map from 1937 is similar to Map 2. From Area A (good/adequate) to Area D (inadequate/risks), the redlining areas are color-coded. At the time, Area D primarily included Black Americans, Asians, and foreign laborers like Italians.The 2019 U.S. Census data on the various racial/ethnic groups, homeownership, and income was obtained in order to properly examine the consequences of historical redlining.

Urban Heat Island

City of Chicago map with reference to Urban Heat island and HOLC Grading

The maps in this section compare the urban heat island effect to the HOLC grade based on the Census tract. The temperatures calculated using Multispectral Landsat imagery for the month of JULY 2022 are used to represent the warmest and coolest parts of Chicago and San Francisco. In both cities, we notice that the grade "A" and "B" neighborhoods are among the coolest indicating a probable high tree coverage. The amount of tree cover and vegetation affects the relative temperature. However, we can see that there are more Grade "D" areas in the Chicago Downtown area with higher warmer regions than cooler temperature zones, compared to the San Francisco area, where even tracts classified as C or D have a similar amount of cooler zones, indicating greater tree coverage or pervious surface/heat island resisting methods being in place than Chicago Downtown.

City of San Francisco map with reference to Urban Heat island and HOLC Grading

Redlining Index and HOLC Grade

Redlining Index, HOLC Grade hotspot analysis and overlay analysis, Chicago

The maps here demonstrate the redlining index, HOLC Grade hotspot, and overlay analysis for the city of Chicago. Negative indexing (-1 to 1) was used for redlining the index map so that it co-relates with the hotspot analysis and gets the cumulative overlay map. The redlining index and the hotspot analysis map both represent the concentration of the percentage as well as the hotspot around the Chicago downtown area and a similar pattern is visible in overlay analysis where the yellow grade (-0.639 to 0.01) indicates the cumulative concentration of both the factors.

Redlining Index, HOLC Grade hotspot analysis and overlay analysis, San Francisco.

The maps here demonstrate the redlining index, HOLC Grade hotspot, and overlay analysis for the city of San Francisco. Negative indexing (-1 to 1) was used for redlining the index map so that it co-relates with the hotspot analysis and gets the cumulative overlay map. The redlining index and the hotspot analysis map represent the concentration of the percentage as well as the hotspot around the northeast side of the city and a similar pattern is visible in overlay analysis where the yellow grade (-0.20 to 1.3) indicates the cumulative concentration of both the factors. However, one peculiarity in the redlining map observed on the northwest side is that certain b grade census tracts are coded as dark red indicating that the percentage of the b grade are either higher or moved into the c and d grade category.

Demographics

01 / 08

Chicago | Total Population

The map below depicts the population density in Chicago. Population density is higher in the colors highlighted in bright red zones. Although the graphic shows that the Grade "c" has the most people in Chicago, the map shows a different picture, in which the Grade "a" only has four census tracts but accounts for 19414 people, indicating higher desirability.

HOLC grade count per census tract

San Francisco | Total Population

The map on the right depicts the population density in San Francisco (SF). Population density is higher in the colors highlighted in bright red zones. Although the graphic shows that the Grade "D" has the most people in SF the map shows a different picture, in which the Grade "a" and "b" even though account for less tracts have comparatively higher population.

HOLC grade count per census tract

Chicago | Racial Diversity

The map here compares the HOLC Grade to the diversity index prevalent in the City of Chicago. The dark blue zones indicate a census tract with a diverse population, whereas the lighter blue zones indicate a concentration of one type of population in the area. The bar graph below attests the analytical value to the importance of the colored zones, with the zones graded "C" and "D" having comparatively higher racial diversity in the City of Chicago.

San Francisco | Racial Diversity

The map on the right compares the HOLC Grade to the diversity index prevalent in the City of San Francisco. The dark blue zones indicate a census tract with a diverse population, whereas the lighter blue zones indicate a concentration of one type of population in the area. The bar graph below attests analytical value to the importance of the colored zones, with the zones graded "C" and "D" having comparatively higher racial diversity in the city of San Francisco as well.

Chicago | Median Income

The median income map and its relationship to the HOLC Grade paint a different picture for Chicago. Whereas the population is higher in the "D" Grade zones in terms of HOLC count, the median income is inversely related, with residents of "A" zones receiving the highest median income of 140,530$ and other zones falling somewhere in the similar range of 58000$ to 50000$. This could be an indicator of the zones' affordability.

San Francisco | Median Income

The Median income map here shows the relationship to the HOLC Grades and paints a different picture for SF as well. Whereas the population is higher in the "D" Grade zones in terms of HOLC count, the median income is inversely related, with residents of "A" zones receiving the highest median income of 250k and other zones falling somewhere in the similar range of 120k to 150k. One peculiar distinction here is that the median income of grades C and D is higher than grade B whereas we noticed a linear inversion pattern in Chicago.

Chicago | Asthma index

The map shows that asthma is more prevalent in the Grade "D" and "C" zones, which are also indicated in the bar below. Although the map shows a higher concentration of asthma index in grades D and C, the census tract counts of 4 and 66, respectively show that grades A and B are equally affected.

San Francisco | Asthma Index

The map shows that asthma is more prevalent in the Grade "D" and "C" zones in SF. Although the map shows fairly equal grading in the city the asthma index in grades D and C ranges from 8 to 16 percent whereas in Grade "A" and "B" zones it is generally noticed under 12. Denoting higher risk in Grade C and D zones of SF

Overlay analysis of Total population, Median Income, Racial diversity index, and asthma prevalence index for Chicago

Overlay analysis of Total population, Median Income, Racial diversity index, and asthma prevalence index for San Francisco

The maps here are a cumulative analysis of the hotspot analysis ( provided in the appendix) and the redlining analysis (shown in the previous) section. these maps provide evidence of whether or not the data are correlated. The first set of maps is for the city of Chicago wherein the census tract highlighted in yellow leaning toward the 1 index aligns with the hotspot analysis and the data obtained from the symbolized maps. The map also highlights that apart from the asthma prevalence index (-0.2 to 1) all other parameters total population (0.6 to 1), the median income ( -0.8 to 1), racial diversity (-0.29 to 1) are concentrated in the North or north west side of Chicago.

The second set of maps is for the city of San Francisco, where the census tract highlighted in yellow leaning toward the 1 index corresponds to the hotspot analysis and data from the symbolized maps. The first thing we notice about the San Francisco map is that there isn't as much clustering of each parameter as we see in the city of Chicago. While the total population (-0.58 to 1) and median income (-0.15 to 1) are concentrated on the city's southwest side (yellow code), the Diversity index appears to be more prevalent along the city's west and southeast edges/census tracts ( although resembling the pattern of Chicago but not as concentrated). One astonishing fact is that, unlike Chicago, the asthma prevalence index (-0.7 to 1) is concentrated in areas with a high population and median income.

Density Mapping

Kernel Density | Chicago

This analysis helps clcaulates a magnitude-per-unit area from point or polyline features by fitting a smoothly tapered surface to each point or polyline with a kernel function. The concentration of points is in the Downton and in the area closer to the lake, which is also graded as HOLC grade "D" in the redlining map, providing a density index for analyzing the grading composition.

Kernel Density | San Francisco

The concentration of points in San Francisco is largely on the east side, which is also graded as HOLC grade "D" in the redlining map, providing a density index for analyzing the grading composition. The D grade here functions as an apex from which the grades from C to A are zone moving outwards toward the ocean.

TRI Index | Chicago

The TRI index is a resource for learning about toxic chemical releases and pollution prevention activities reported by industrial and federal facilities. TRI data assists communities, governments, businesses, and others in making informed decisions. The map below depicts the promotional symbols for TRI locations in Chicago. The larger symbols indicate that a since site has more than one toxic release index (e.g., stack air, fugitive air, etc.). As we can see, the concentration of such sites is higher in the city's south and west sides, indicating an increased health risk index.

TRI Index by Industry

TRI Index | San Francisco

While The TRI Map of Chicago demonstrates such a large amount of TRI concentration the city of San Francisco only indicates the presence of Three. However, we can notice on the map that all of these three location are closer or within the tracts graded as D or C under the HOLC grading, thus indicating increased health risk for the population in this area.

Conclusion

Thus, based on the analysis above and our research on the segregation and redlining policies of the twentieth-century United States. Although the policy was repealed about 50 years ago, the consequences of it continue to be felt by residents of both cities. While Grade D and C are more prevalent in Chicago, the City of San Francisco addresses a balanced mix of all zones. The four demographic parameters compared here are median income, total population, racial diversity, and asthma. The concentration of affluent/desirable indicators, such as higher median income, population density, racial diversity, is found in the Grade A and B zones, whereas the health indicator of asthma prevalence index are found in higher value the Grade D and C zones. Also based on the Z score and overlay analysis these parameters reflect the same trend hence proving our hypothesis true. As planner and policy makers we should provide due consideration of these studies and indices while proposing a future policy or planning derivatives and move towards mending the hardships inflicted on the minority groups especially during the 20th century.

Limitations

Although these parameters paint a good picture of how these factors are interrelated we intentionally started by considering more environmental hazards but due to issues in visualizing the data we switched it to a combination of health and demographic study.

The first limitation occurred while creating a PM 2.5 and AQI value map from a point source data; while we were able to achieve an interpolated output, the point count was insufficient to convert the data into a vector file and use it for a hot spot analysis to standardize the results.

Secondly, we faced issues in using the pivot table for indexing the redline percentages. We began with using the tabular intersection method which gave us the percentages but when executed in excel the indexing didn't go as planned and hence we went ahead with the standardize field tool from the ArcGIS pro tool

Finally, the Multispectral Landsat Imagery raster function did not work as planned and had to be redone several times to finally get the desired output; however, because the output was not an integer type, it could not be converted into a raster file, and all other methods of doing so were beyond the scope of our study, so we simply used an image overlay to analyze the relationship between the urban heat island and the HOLC Grade.

Appendix

Hot spot analysis for Total population, racial diversity, median income and asthma for Chicago

Hot spot analysis for Total population, racial diversity, median income and asthma for San Francisco

Image Sources

Credits

UPP 462_FALL 2022_KASTURI ACHARYA & HIMALI CHINGALE

States where redlining effect is prevalent in the United States

Methodology flowchart created on Miro

City of Chicago map with reference to Urban Heat island and HOLC Grading

City of San Francisco map with reference to Urban Heat island and HOLC Grading

HOLC grade count per census tract

HOLC grade count per census tract

TRI Index by Industry