Historic Redlining
Philadelphia example
The Census Home Owners’ Loan Corporation (HOLC) dataset, available on GitHub, is a valuable resource containing historical data about the process of redlining in the United States. Redlining is a discriminatory practice that involves denying loans or insurance to people living in specific neighborhoods, often based on race or ethnicity. The dataset provides details on the HOLC risk classifications, neighborhood boundaries, and descriptions for over 200 cities across the United States from the 1930s.
Historic redlining data refers to data from the Home Owners’ Loan Corporation (HOLC) that created residential security maps in the 1930s, which contributed to racial segregation and disinvestment in minority neighborhoods. One popular source for this data is the Mapping Inequality project (https://dsl.richmond.edu/panorama/redlining/).
In this example, we’ll download historic redlining data for Philadelphia in the form of a GeoJSON file and analyze the data in R and Python.
R: In R, we’ll use the ‘sf’ and ‘dplyr’ packages to read and process the GeoJSON data.
R code:
# Install and load necessary libraries]
library(knitr)
library(sf)
library(dplyr)
# Download historic redlining data for Philadelphia
url <- "https://dsl.richmond.edu/panorama/redlining/static/downloads/geojson/PAPhiladelphia1937.geojson"
philly_geojson <- read_sf(url)
# Count the number of areas per HOLC grade
grade_counts <- philly_geojson %>%
group_by(holc_grade) %>%
summarize(count = n())
kable(head(philly_geojson))
| name | holc_id | holc_grade | area_description_data | geometry |
|---|---|---|---|---|
| NA | A1 | A | { “1”: “Sub Phila A 1”, “2”: “Rolling”, “3”: “Highly desirable, well restricted residential area.”, “4”: “No sewers.”, “7”: ” 1934-36 55 very few rentals June 1937 \$8,600-100,000 70 “,”13”: “upward”, “14”: “\$15,000-25,000 houses predominate”, “15”: “David C. Snyder June 11 7”, “10a”: “good”, “10b”: “few units”, “10c”: ““,”11a”: “singles \$15000-\$20000”, “11b”: “30-40 houses”, “12a”: “yes”, “12b”: “yes”, “5a”: “Execs. - Professional”, “5b”: “over \$5,000”, “5c”: “None”, “5d”: “No”, “5e”: “desirables”, “5f”: “no”, “5g”: “slowly”, “6a”: “single detached”, “6b”: “stone-brick”, “6c”: “1-15”, “6d”: “good”, “8a”: “50”, “8b”: “95”, “8c”: “over 80”, “9a”: “fair”, “9b”: “singles \$10,000”, “9c”: “fair” } | MULTIPOLYGON (((-75.12338 4… |
| NA | A2 | A | { “1”: “Sub. Phila. Elkins Park A 2”, “2”: “Rolling”, “3”: “Good transportation - desirable residential area of good character.”, “4”: “None”, “7”: “10,000-up \$20,000 150 1934-46 6500-up 12,000 60 75 50 June 1937 \$7,500-up 15,000 75 100 67”, “13”: “static to upward”, “14”: “Many residents of Elkin Park commute to New York City.”, “15”: “David C. Snyder June 11, 7”, “10a”: “good”, “10b”: “anything”, “10c”: “good”, “11a”: “Singles \$12,000-\$15,000”, “11b”: “50 houses”, “12a”: “yes”, “12b”: “yes”, “5a”: “Execs - Professional”, “5b”: “over \$5,000”, “5c”: “no”, “5d”: “no”, “5e”: “desirables”, “5f”: “no”, “5g”: “slowly”, “6a”: “single- twins”, “6b”: “stone-brick-frame”, “6c”: “25 yrs.”, “6d”: “good”, “8a”: “60”, “8b”: “100”, “8c”: “over 80”, “9a”: “good”, “9b”: “Single \$13,000-15,000”, “9c”: “good” } | MULTIPOLYGON (((-75.12882 4… |
| NA | A3 | A | { “1”: “Philadelphia A 3”, “2”: “Rolling-level”, “3”: “Direct rail transportation- Homogeneous development of desirable residential area.”, “4”: “No”, “7”: ” very few rental units 1933-34 June 1937 \$8000.- \$25,000 “,”13”: “upward”, “14”: “Property around R.R. station not quite as high grade. Singles average \$20,000, Twins average \$10,000. High grade area recently developed. Not quite as accessible as west Germantown section.”, “15”: “Doherty June 3 7”, “10a”: “good”, “10b”: “limited number of units \$10,000-\$20,000”, “10c”: ““,”11a”: “Twins & singles”, “11b”: “100 houses”, “12a”: “yes”, “12b”: “yes”, “5a”: “Junior Executives- Professional men”, “5b”: “\$3000.-\$10,000”, “5c”: “no”, “5d”: “no”, “5e”: “no”, “5f”: “no”, “5g”: “moderately”, “6a”: “twins-singles”, “6b”: “stone & brick”, “6c”: “1-20 yrs.”, “6d”: “good”, “8a”: ““,”8b”: ““,”8c”: “over 80”, “9a”: “good”, “9b”: “for new houses \$10,000 twins”, “9c”: “good” } | MULTIPOLYGON (((-75.17329 4… |
| NA | A4 | A | { “1”: “Philadelphia A 4”, “2”: “Rolling”, “3”: “The finest residential section in the city. Very good schools- Near parks- electrified train transportation.”, “4”: “None”, “7”: “\$15,000- up \$40,000 \$100.00 \$150.00 1933-34 \$8,000- \$20,000 50% \$65.00 \$80.00 55% June 1937 \$10,000-\$100,000 25,000 60-65% \$75.-\$200. \$100.00 65%”, “13”: “static to upward”, “14”: “Shifting of population. People moving further out. Southern end of section has the smaller houses. Buyers still looking for bargains. Practically all available land is held at \$20,000 an acre. Direct roads to center of city. Many large estates in this section.”, “15”: “Doherty June 3 7”, “10a”: “good”, “10b”: “everything”, “10c”: “fair”, “11a”: “single detached”, “11b”: “25 units”, “12a”: “yes”, “12b”: “yes”, “5a”: “Executives- Professional men”, “5b”: “over \$6000.”, “5c”: “no”, “5d”: “no”, “5e”: “no”, “5f”: “no”, “5g”: “slowly”, “6a”: “One family detached”, “6b”: “stone”, “6c”: “1-20 yrs.”, “6d”: “very good”, “8a”: “80”, “8b”: “100”, “8c”: “over 80”, “9a”: “fair”, “9b”: ““,”9c”: “fair” } | MULTIPOLYGON (((-75.19831 4… |
| NA | A5 | A | { “1”: “Philadelphia A 5”, “2”: “Rolling”, “3”: “Good transportation- very desirable residential section- well restricted- wooded section”, “4”: “None”, “7”: “\$10,000.00 \$30,000 \$70 \$100 1933-34 \$5,000 \$12,000 40% \$40 \$60 60% June 1937 \$6,000-\$50,000 \$15,000 50% \$50-\$200 \$75 75%”, “13”: “static”, “14”: “German American population- Broad St. properties are not typical of section as a whole. Section is very desirable, but danger of Jewish encroachment is imminent. A fair first grade section.”, “15”: “H.B. Wilson June 14 7”, “10a”: “good”, “10b”: “anything”, “10c”: “good”, “11a”: “stone singles \$12,500”, “11b”: “10-12 units”, “12a”: “yes”, “12b”: “yes”, “5a”: “White collar class”, “5b”: “\$5000. & up”, “5c”: “none”, “5d”: “no”, “5e”: “desirables”, “5f”: “none”, “5g”: “slowly”, “6a”: “single- twins”, “6b”: “brick & stone”, “6c”: “1-20 yrs.”, “6d”: “good”, “8a”: “95”, “8b”: “100”, “8c”: “80”, “9a”: “fair”, “9b”: “single & twins \$9000”, “9c”: “fair” } | MULTIPOLYGON (((-75.1281 40… |
| NA | A6 | A | { “1”: “Philadelphia A 6”, “2”: “Level”, “3”: “Desirable industrial section- good transportation. Thrifty class of populace- 2 perks in section.”, “4”: “No”, “7”: “\$4200.-\$8500. \$6,500.00 \$35.-\$65. \$50.00 1933-34 \$2500.-\$4500. \$3,500.00 55% \$25.-\$40 \$32.00 15% June 1937 \$3000.-\$6000. \$4,500.00 70% \$28.-\$50. \$40.00 80%”, “13”: “Static to upward”, “14”: “Most of building in eastern end. High type inhabitants. A fair first grade section.”, “15”: “H.B. Wilson June 14 7”, “10a”: “good”, “10b”: “2 story row- \$40.”, “10c”: “good”, “11a”: “2 story row \$4500.”, “11b”: “100-125 units”, “12a”: “yes”, “12b”: “yes”, “5a”: “White collar-skilled mechanics”, “5b”: “\$1500- \$3000”, “5c”: “nominal”, “5d”: “no”, “5e”: “desirable”, “5f”: “small”, “5g”: “slowly”, “6a”: “2 story row”, “6b”: “brick”, “6c”: “1-25 yrs.”, “6d”: “good”, “8a”: “75”, “8b”: “100”, “8c”: “75”, “9a”: “good”, “9b”: “2 story row-\$4000-\$4500.”, “9c”: “fair” } | MULTIPOLYGON (((-75.11494 4… |
plot(grade_counts)
Python: In Python, we’ll use the ‘geopandas’ library to read and process the GeoJSON data.
Python code:
# Install necessary libraries
import geopandas as gpd
# Download historic redlining data for Philadelphia
url = "https://dsl.richmond.edu/panorama/redlining/static/downloads/geojson/PAPhiladelphia1937.geojson"
philly_geojson = gpd.read_file(url)
# Count the number of areas per HOLC grade
grade_counts = philly_geojson["holc_grade"].value_counts()
print(grade_counts)
B 28
D 26
C 18
A 10
Name: holc_grade, dtype: int64
In conclusion, both R and Python offer efficient ways to download and process historic redlining data in the form of GeoJSON files. The ‘sf’ package in R provides a simple way to read and manipulate spatial data, while the ‘geopandas’ library in Python offers similar functionality. The ‘dplyr’ package in R can be used for data manipulation and analysis, and Python’s built-in functions like value_counts() can be used for aggregating data. Depending on your preferred programming language and environment, both options can be effective for working with historic redlining data.
Per census track
Downloading and plotting the dataset in R:
R code:
# Load libraries
library(sf)
library(ggplot2)
# Download the dataset
url <- "https://raw.githubusercontent.com/americanpanorama/Census_HOLC_Research/main/2020_Census_Tracts/Tracts_2020_HOLC.geojson"
holc_data <- st_read(url)
Reading layer `Tracts_2020_HOLC' from data source
`https://raw.githubusercontent.com/americanpanorama/Census_HOLC_Research/main/2020_Census_Tracts/Tracts_2020_HOLC.geojson'
using driver `GeoJSON'
Simple feature collection with 42074 features and 9 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: -122.7675 ymin: 25.70537 xmax: -70.94938 ymax: 47.72251
Geodetic CRS: WGS 84
kable(head(holc_data))
| OBJECTID | GISJOIN | neighborho | SUM_Perc | FIRST_holc_grade | MAX_state | MAX_city | Shape_Length | Shape_Area | geometry |
|---|---|---|---|---|---|---|---|---|---|
| 1 | G0100730000100 | 180 | 0.0929688 | D | AL | Birmingham | 3336.386 | 709582.0 | MULTIPOLYGON (((-86.71213 3… |
| 2 | G0100730000100 | 181 | 0.1616926 | D | AL | Birmingham | 7731.381 | 1234059.9 | MULTIPOLYGON (((-86.72188 3… |
| 3 | G0100730000100 | 182 | 0.1703257 | D | AL | Birmingham | 6111.241 | 1300062.3 | MULTIPOLYGON (((-86.72644 3… |
| 4 | G0100730000100 | 185 | 0.2582595 | C | AL | Birmingham | 7905.510 | 1971178.6 | MULTIPOLYGON (((-86.71611 3… |
| 5 | G0100730000100 | 187 | 0.0419889 | B | AL | Birmingham | 3422.128 | 320481.6 | MULTIPOLYGON (((-86.70262 3… |
| 6 | G0100730000300 | 182 | 0.8398497 | D | AL | Birmingham | 6759.741 | 1745630.8 | MULTIPOLYGON (((-86.73525 3… |
# Plot the data
ggplot() +
geom_sf(data = holc_data) +
theme_minimal() +
ggtitle("2020 Census Tracts with HOLC Data")
Downloading and plotting the dataset in Python:
Python code:
# Import libraries
import geopandas as gpd
# Download the dataset
url = "https://raw.githubusercontent.com/americanpanorama/Census_HOLC_Research/main/2020_Census_Tracts/Tracts_2020_HOLC.geojson"
holc_data = gpd.read_file(url)
# Display the data as a table
print(holc_data)
# Plot the data
holc_data.plot()
plot.title("2020 Census Tracts with HOLC Data")
plot.show()
In both R and Python, the code downloads the dataset, reads the Shapefile, and plots the data using ggplot2 in R and geopandas in Python. Make sure to install the necessary packages before running the code.
Last update:
2024-05-09