This dataset represents the geospatial extent as polygons and the corresponding attribution for census block groups that meet the definition of low-income communities according to the Virginia 2020 Environmental Justice Act: “Low-income community” definition: “’Low-income community’ means any census block group in which 30 percent or more of the population is composed of people with low income.”

The referenced “low income” definition is also provided below: “Low income” definition: “’Low income’ means having an annual household income equal to or less than the greater of (i) an amount equal to 80 percent of the median income of the area in which the household is located, as reported by the Department of Housing and Urban Development, and (ii) 200 percent of the Federal Poverty Level.”

This dataset and map service provides information on the U.S. Housing and Urban Development's (HUD) low to moderate income areas. The term Low to Moderate Income, often referred to as low-mod, has a specific programmatic context within the Community Development Block Grant (CDBG) program. Over a 1, 2, or 3-year period, as selected by the grantee, not less than 70 percent of CDBG funds must be used for activities that benefit low- and moderate-income persons. HUD uses special tabulations of Census data to determine areas where at least 51% of households have incomes at or below 80% of the area median income (AMI). This dataset and map service contains the following layer.

This dataset represents the geospatial extent as polygons and the corresponding attribution for census block groups that meet the definition of low-income communities according to the Virginia 2020 Environmental Justice Act: “Low-income community” definition: “’Low-income community’ means any census block group in which 30 percent or more of the population is composed of people with low income.”

The referenced “low income” definition is also provided below: “Low income” definition: “’Low income’ means having an annual household income equal to or less than the greater of (i) an amount equal to 80 percent of the median income of the area in which the household is located, as reported by the Department of Housing and Urban Development, and (ii) 200 percent of the Federal Poverty Level.”

Urban Displacement Project’s (UDP) Estimated Displacement Risk (EDR) model for California identifies varying levels of displacement risk for low-income renter households in all census tracts in the state from 2015 to 2019(1). The model uses machine learning to determine which variables are most strongly related to displacement at the household level and to predict tract-level displacement risk statewide while controlling for region. UDP defines displacement risk as a census tract with characteristics which, according to the model, are strongly correlated with more low-income population loss than gain. In other words, the model estimates that more low-income households are leaving these neighborhoods than moving in.This map is a conservative estimate of low-income loss and should be considered a tool to help identify housing vulnerability. Displacement may occur because of either investment, disinvestment, or disaster-driven forces. Because this risk assessment does not identify the causes of displacement, UDP does not recommend that the tool be used to assess vulnerability to investment such as new housing construction or infrastructure improvements. HCD recommends combining this map with on-the-ground accounts of displacement, as well as other related data such as overcrowding, cost burden, and income diversity to achieve a full understanding of displacement risk.If you see a tract or area that does not seem right, please fill out this form to help UDP ground-truth the method and improve their model.How should I read the displacement map layers?The AFFH Data Viewer includes three separate displacement layers that were generated by the EDR model. The “50-80% AMI” layer shows the level of displacement risk for low-income (LI) households specifically. Since UDP has reason to believe that the data may not accurately capture extremely low-income (ELI) households due to the difficulty in counting this population, UDP combined ELI and very low-income (VLI) household predictions into one group—the “0-50% AMI” layer—by opting for the more “extreme” displacement scenario (e.g., if a tract was categorized as “Elevated” for VLI households but “Extreme” for ELI households, UDP assigned the tract to the “Extreme” category for the 0-50% layer). For these two layers, tracts are assigned to one of the following categories, with darker red colors representing higher displacement risk and lighter orange colors representing less risk:• Low Data Quality: the tract has less than 500 total households and/or the census margins of error were greater than 15% of the estimate (shaded gray).• Lower Displacement Risk: the model estimates that the loss of low-income households is less than the gain in low-income households. However, some of these areas may have small pockets of displacement within their boundaries. • At Risk of Displacement: the model estimates there is potential displacement or risk of displacement of the given population in these tracts.• Elevated Displacement: the model estimates there is a small amount of displacement (e.g., 10%) of the given population.• High Displacement: the model estimates there is a relatively high amount of displacement (e.g., 20%) of the given population.• Extreme Displacement: the model estimates there is an extreme level of displacement (e.g., greater than 20%) of the given population. The “Overall Displacement” layer shows the number of income groups experiencing any displacement risk. For example, in the dark red tracts (“2 income groups”), the model estimates displacement (Elevated, High, or Extreme) for both of the two income groups. In the light orange tracts categorized as “At Risk of Displacement”, one or all three income groups had to have been categorized as “At Risk of Displacement”. Light yellow tracts in the “Overall Displacement” layer are not experiencing UDP’s definition of displacement according to the model. Some of these yellow tracts may be majority low-income experiencing small to significant growth in this population while in other cases they may be high-income and exclusive (and therefore have few low-income residents to begin with). One major limitation to the model is that the migration data UDP uses likely does not capture some vulnerable populations, such as undocumented households. This means that some yellow tracts may be experiencing high rates of displacement among these types of households. MethodologyThe EDR is a first-of-its-kind model that uses machine learning and household level data to predict displacement. To create the EDR, UDP first joined household-level data from Data Axle (formerly Infogroup) with tract-level data from the 2014 and 2019 5-year American Community Survey; Affirmatively Furthering Fair Housing (AFFH) data from various sources compiled by California Housing and Community Development; Longitudinal Employer-Household Dynamics (LEHD) Origin-Destination Employment Statistics (LODES) data; and the Environmental Protection Agency’s Smart Location Database.UDP then used a machine learning model to determine which variables are most strongly related to displacement at the household level and to predict tract-level displacement risk statewide while controlling for region. UDP modeled displacement risk as the net migration rate of three separate renter households income categories: extremely low-income (ELI), very low-income (VLI), and low-income (LI). These households have incomes between 0-30% of the Area Median Income (AMI), 30-50% AMI, and 50-80% AMI, respectively. Tracts that have a predicted net loss within these groups are considered to experience displacement in three degrees: elevated, high, and extreme. UDP also includes a “At Risk of Displacement” category in tracts that might be experiencing displacement.What are the main limitations of this map?1. Because the map uses 2019 data, it does not reflect more recent trends. The pandemic, which started in 2020, has exacerbated income inequality and increased housing costs, meaning that UDP’s map likely underestimates current displacement risk throughout the state.2. The model examines displacement risk for renters only, and does not account for the fact that many homeowners are also facing housing and gentrification pressures. As a result, the map generally only highlights areas with relatively high renter populations, and neighborhoods with higher homeownership rates that are known to be experiencing gentrification and displacement are not as prominent as one might expect.3. The model does not incorporate data on new housing construction or infrastructure projects. The map therefore does not capture the potential impacts of these developments on displacement risk; it only accounts for other characteristics such as demographics and some features of the built environment. Two of UDP’s other studies—on new housing construction and green infrastructure—explore the relationships between these factors and displacement.Variable ImportanceFigures 1, 2, and 3 show the most important variables for each of the three models—ELI, VLI, and LI. The horizontal bars show the importance of each variable in predicting displacement for the respective group. All three models share a similar order of variable importance with median rent, percent non-white, rent gap (i.e., rental market pressure calculated using the difference between nearby and local rents), percent renters, percent high-income households, and percent of low-income households driving much of the displacement estimation. Other important variables include building types as well as economic and socio-demographic characteristics. For a full list of the variables included in the final models, ranked by descending order of importance, and their definitions see all three tabs of this spreadsheet. “Importance” is defined in two ways: 1. % Inclusion: The average proportion of times this variable was included in the model’s decision tree as the most important or driving factor.2. MeanRank: The average rank of importance for each variable across the numerous model runs where higher numbers mean higher ranking. Figures 1 through 3 below show each of the model variable rankings ordered by importance. The red lines represent Jenks Breaks, which are designed to sort values into their most “natural” clusters. Variable importance for each model shows a substantial drop-off after about 10 variables, meaning a relatively small number of variables account for a large amount of the predictive power in UDP’s displacement model.Figure 1. Variable Importance for Low Income HouseholdsFor a description of each variable and its source, see this spreadsheet.Figure 2. Variable Importance for Very Low Income HouseholdsFor a description of each variable and its source, see this spreadsheet. Figure 3. Variable Importance for Extremely Low Income HouseholdsFor a description of each variable and its source, see this spreadsheet.Source: Chapple, K., & Thomas, T., and Zuk, M. (2022). Urban Displacement Project website. Berkeley, CA: Urban Displacement Project.(1) UDP used this time-frame because (a) the 2020 census had a large non-response rate and it implemented a new statistical modification that obscures and misrepresents racial and economic characteristics at the census tract level and (b) pandemic mobility trends are still in flux and UDP believes 2019 is more representative of “normal” or non-pandemic displacement trends.

AMI acute myocardial infarction, AHA American Heart Association, ESC European Society of Cardiology.

This dataset represents the geospatial extent as polygons and the corresponding attribution for census block groups that meet the definition of low-income communities according to the Virginia 2020 Environmental Justice Act: “Low-income community” definition: “’Low-income community’ means any census block group in which 30 percent or more of the population is composed of people with low income.”

The referenced “low income” definition is also provided below: “Low income” definition: “’Low income’ means having an annual household income equal to or less than the greater of (i) an amount equal to 80 percent of the median income of the area in which the household is located, as reported by the Department of Housing and Urban Development, and (ii) 200 percent of the Federal Poverty Level.”Click Here to view Data Fact Sheet.

Urban Displacement Project’s (UDP) Estimated Displacement Risk (EDR) model for California identifies varying levels of displacement risk for low-income renter households in all census tracts in the state from 2015 to 2019(1). The model uses machine learning to determine which variables are most strongly related to displacement at the household level and to predict tract-level displacement risk statewide while controlling for region. UDP defines displacement risk as a census tract with characteristics which, according to the model, are strongly correlated with more low-income population loss than gain. In other words, the model estimates that more low-income households are leaving these neighborhoods than moving in.This map is a conservative estimate of low-income loss and should be considered a tool to help identify housing vulnerability. Displacement may occur because of either investment, disinvestment, or disaster-driven forces. Because this risk assessment does not identify the causes of displacement, UDP does not recommend that the tool be used to assess vulnerability to investment such as new housing construction or infrastructure improvements. HCD recommends combining this map with on-the-ground accounts of displacement, as well as other related data such as overcrowding, cost burden, and income diversity to achieve a full understanding of displacement risk.If you see a tract or area that does not seem right, please fill out this form to help UDP ground-truth the method and improve their model.How should I read the displacement map layers?The AFFH Data Viewer includes three separate displacement layers that were generated by the EDR model. The “50-80% AMI” layer shows the level of displacement risk for low-income (LI) households specifically. Since UDP has reason to believe that the data may not accurately capture extremely low-income (ELI) households due to the difficulty in counting this population, UDP combined ELI and very low-income (VLI) household predictions into one group—the “0-50% AMI” layer—by opting for the more “extreme” displacement scenario (e.g., if a tract was categorized as “Elevated” for VLI households but “Extreme” for ELI households, UDP assigned the tract to the “Extreme” category for the 0-50% layer). For these two layers, tracts are assigned to one of the following categories, with darker red colors representing higher displacement risk and lighter orange colors representing less risk:• Low Data Quality: the tract has less than 500 total households and/or the census margins of error were greater than 15% of the estimate (shaded gray).• Lower Displacement Risk: the model estimates that the loss of low-income households is less than the gain in low-income households. However, some of these areas may have small pockets of displacement within their boundaries. • At Risk of Displacement: the model estimates there is potential displacement or risk of displacement of the given population in these tracts.• Elevated Displacement: the model estimates there is a small amount of displacement (e.g., 10%) of the given population.• High Displacement: the model estimates there is a relatively high amount of displacement (e.g., 20%) of the given population.• Extreme Displacement: the model estimates there is an extreme level of displacement (e.g., greater than 20%) of the given population. The “Overall Displacement” layer shows the number of income groups experiencing any displacement risk. For example, in the dark red tracts (“2 income groups”), the model estimates displacement (Elevated, High, or Extreme) for both of the two income groups. In the light orange tracts categorized as “At Risk of Displacement”, one or all three income groups had to have been categorized as “At Risk of Displacement”. Light yellow tracts in the “Overall Displacement” layer are not experiencing UDP’s definition of displacement according to the model. Some of these yellow tracts may be majority low-income experiencing small to significant growth in this population while in other cases they may be high-income and exclusive (and therefore have few low-income residents to begin with). One major limitation to the model is that the migration data UDP uses likely does not capture some vulnerable populations, such as undocumented households. This means that some yellow tracts may be experiencing high rates of displacement among these types of households. MethodologyThe EDR is a first-of-its-kind model that uses machine learning and household level data to predict displacement. To create the EDR, UDP first joined household-level data from Data Axle (formerly Infogroup) with tract-level data from the 2014 and 2019 5-year American Community Survey; Affirmatively Furthering Fair Housing (AFFH) data from various sources compiled by California Housing and Community Development; Longitudinal Employer-Household Dynamics (LEHD) Origin-Destination Employment Statistics (LODES) data; and the Environmental Protection Agency’s Smart Location Database.UDP then used a machine learning model to determine which variables are most strongly related to displacement at the household level and to predict tract-level displacement risk statewide while controlling for region. UDP modeled displacement risk as the net migration rate of three separate renter households income categories: extremely low-income (ELI), very low-income (VLI), and low-income (LI). These households have incomes between 0-30% of the Area Median Income (AMI), 30-50% AMI, and 50-80% AMI, respectively. Tracts that have a predicted net loss within these groups are considered to experience displacement in three degrees: elevated, high, and extreme. UDP also includes a “At Risk of Displacement” category in tracts that might be experiencing displacement.What are the main limitations of this map?1. Because the map uses 2019 data, it does not reflect more recent trends. The pandemic, which started in 2020, has exacerbated income inequality and increased housing costs, meaning that UDP’s map likely underestimates current displacement risk throughout the state.2. The model examines displacement risk for renters only, and does not account for the fact that many homeowners are also facing housing and gentrification pressures. As a result, the map generally only highlights areas with relatively high renter populations, and neighborhoods with higher homeownership rates that are known to be experiencing gentrification and displacement are not as prominent as one might expect.3. The model does not incorporate data on new housing construction or infrastructure projects. The map therefore does not capture the potential impacts of these developments on displacement risk; it only accounts for other characteristics such as demographics and some features of the built environment. Two of UDP’s other studies—on new housing construction and green infrastructure—explore the relationships between these factors and displacement.Variable ImportanceFigures 1, 2, and 3 show the most important variables for each of the three models—ELI, VLI, and LI. The horizontal bars show the importance of each variable in predicting displacement for the respective group. All three models share a similar order of variable importance with median rent, percent non-white, rent gap (i.e., rental market pressure calculated using the difference between nearby and local rents), percent renters, percent high-income households, and percent of low-income households driving much of the displacement estimation. Other important variables include building types as well as economic and socio-demographic characteristics. For a full list of the variables included in the final models, ranked by descending order of importance, and their definitions see all three tabs of this spreadsheet. “Importance” is defined in two ways: 1. % Inclusion: The average proportion of times this variable was included in the model’s decision tree as the most important or driving factor.2. MeanRank: The average rank of importance for each variable across the numerous model runs where higher numbers mean higher ranking. Figures 1 through 3 below show each of the model variable rankings ordered by importance. The red lines represent Jenks Breaks, which are designed to sort values into their most “natural” clusters. Variable importance for each model shows a substantial drop-off after about 10 variables, meaning a relatively small number of variables account for a large amount of the predictive power in UDP’s displacement model.Figure 1. Variable Importance for Low Income HouseholdsFor a description of each variable and its source, see this spreadsheet.Figure 2. Variable Importance for Very Low Income HouseholdsFor a description of each variable and its source, see this spreadsheet. Figure 3. Variable Importance for Extremely Low Income HouseholdsFor a description of each variable and its source, see this spreadsheet.Source: Chapple, K., & Thomas, T., and Zuk, M. (2022). Urban Displacement Project website. Berkeley, CA: Urban Displacement Project.(1) UDP used this time-frame because (a) the 2020 census had a large non-response rate and it implemented a new statistical modification that obscures and misrepresents racial and economic characteristics at the census tract level and (b) pandemic mobility trends are still in flux and UDP believes 2019 is more representative of “normal” or non-pandemic displacement trends.