Data-Smart City Solutions

Search

By Chris Bousquet

Gentrification—demographic and physical changes in neighborhoods that bring in wealthier residents, greater investment, and more development—has become a buzzword in urban planning.  As traditionally low-income neighborhoods across the U.S. gentrify, social justice advocates have become increasingly concerned about displacement, the dislocation of low-income residents due to prohibitive prices. As a result, policymakers and urban planners have begun to consider strategies to combat the byproducts of gentrification in recently-developed or developing neighborhoods, such as providing low-cost amenities and rent controlled or low-income housing.

The first step in addressing gentrification is understanding where it has happened and where it is likely to happen in the future. A number of cities have found mapping to be a powerful tool for observing gentrification trends, allowing them to intervene before low-income residents are seriously affected. Cities have created maps using data mostly from public sources both to better understand historical trends in gentrification and displacement and predict the next areas where low-income residents are likely to lose their homes. While each model is unique, all display methodologies that are applicable across cities. For a factor by factor overview of models in seven U.S. cities, see

Los Angeles i-Team’s Indices of Neighborhood Change and Displacement Pressure

Established in the summer of 2015, with support from Bloomberg Philanthropies, Mayor Eric Garcetti’s Los Angeles Innovation Team (i-team) sought to understand changing neighborhoods and prevent the displacement of businesses and residents.

In order to gain a preliminary understanding of those neighborhoods in need of intervention, the i-team set out to map gentrification and assess its potential effects across Los Angeles. In 2016, the city published the Los Angeles Index of Neighborhood Change, a map that allows users to explore the degree to which zip codes in Los Angeles experienced gentrification between 2000 and 2014. Using publicly-available data, pulled primarily from Census and American Community Survey (ACS) databases, the city calculated index scores for each zip code based on six demographic measures indicative of gentrification, including changes in income, education, race, rent, and household size.

The Los Angeles Index of Neighborhood Change

This index serves as an exploratory tool, providing residents and municipal employees alike with an accessible platform for information on gentrification. The visualization empowers users to conduct their own analyses in order to understand geospatial trends in gentrification that can inform interventions.

The i-team also created the Los Angeles Index of Displacement Pressure, which repurposed much of the data from the neighborhood change project in addition to information from ESRI Community Analyst in order to predict neighborhoods where displacement was likely to occur. The mapping effort combined factors that have been historically indicative of displacement, such as transportation investment, percent of rent-burdened households, and availability of affordable housing. After weighting each factor based on its predictive power, the i-team assigned and mapped risk scores for every neighborhood.

The Los Angeles Index of Displacement Pressure

By specifically identifying neighborhoods at risk of displacing residents, this model provides a more prescriptive roadmap than the neighborhood change map. While maps of historical change detail trends that residents and policymakers may be able to extrapolate into the future, displacement risk maps directly forecast the future, guiding interventions. Moreover, these maps furnish residents with a view into the future of their neighborhoods, providing a political call to action in neighborhoods likely to displace residents.

The city then combined these two indices into a single interactive tool, which, paired with city administrative data, has enabled the Los Angeles i-team to more effectively select locations for its projects. For example, the i-team used the indices to assist with site selection for its Rent Stabilization Ordinance (RSO) outreach campaign, a marketing and awareness initiative to help tenants and landlords better understand and access city resources that can help longtime residents stay in their homes during periods of rising prices. Using the city’s mapping tools, the i-team selected a number of transit shelters, bus stops, subway stations, and city facilities where representatives would offer guidance to residents. Moreover, the i-team identified a number of neighborhoods with concentrated RSO units for direct outreach, issuing guidebooks and door hangers to residents’ doorsteps. Other projects, including the Secondary Unit Prototyping project—an initiative to build prototypes of legal second units on sites with single-family homes—and the Civic Youth project—which teaches LA youth ages 16-21 skills to become civic leaders—also relied on the indices for site selection.

Urban Displacement Project Los Angeles Map of Neighborhood Change

The Urban Displacement Project, an initiative of UCLA and University of California, Berkeley, in conjunction with the State of California’s Air Resources Board (ARB), also pursued a project in 2016 to track neighborhood change in Los Angeles. Using Census data, Geolytics’ neighborhood change database, and LA’s open data, their study sought to map trends in gentrification as well as upscaling—changes in economic and demographic characteristics in all Census tracts—from 1990-2013. While the gentrification analysis focused on indicators like changes in rental markets, racial and ethnic composition, and economic health of low-income tracts, the upscaling analysis looked at factors like household income and education for all tracts. The resulting map categorizes neighborhoods based on four categories: gentrified 1990-2000 only, gentrified 2000-2013 only, gentrified both decades, and did not gentrify.

Urban Displacement Project Los Angeles Map of Neighborhood Change

Unlike the map created by LA’s i-team, the Urban Displacement Project’s initiative draws its own insights from the data, providing additional narrative context. The study found that areas around transit stations were moving towards upscaling more so than other neighborhoods, that populations in these areas were becoming whiter, more educated, and wealthier, and that rents were increasing. These areas also lost disadvantaged populations, including residents with less than a high school diploma and low-income households. By offering its own analysis of LA’s gentrification data, the Urban Displacement Project’s map is more accessible to residents, providing citizens with direct insights into the factors that may cause displacement in their neighborhoods, while still offering a useful GIS tool to policymakers.

Portland’s Susceptibility to Gentrification Model

Creating its first maps of gentrification in 2013, Portland was one of the first U.S. cities to use data visualizations in order to better understand neighborhood change and displacement. For the first of three mapping layers, the city’s Bureau of Planning and Sustainability (BPS) used mostly publicly-available Census and ACS data on housing tenure and price changes, income level, educational attainment, and racial/ethnic population changes in order to assess gentrification risk in each Census tract. BPS then added second and third layers that used data on current and planned public investment, as well as information on proximity to transit, respectively.

Based on this data, BPS assigned neighborhoods to one of six categories: susceptible to gentrification, early gentrification with little demographic change, early gentrification with demographic change, dynamic gentrification, late-stage gentrification, and continued loss.

The Portland Susceptibility to Gentrification Model

This categorization was meant to be both descriptive and loosely predictive; BPS’ intent was to develop a metric for starting policy analysis of the city’s neighborhoods that would motivate follow-up in neighborhoods in need. The accompanying report calls for “data drilldowns” in neighborhoods identified by the typology as at risk of gentrification. While the map allows policymakers to more easily visualize those areas requiring intervention, governments need to gather more detailed data on a neighborhood’s demographic profile, housing vulnerability, and commercial activity and ownership in order to more effectively plan the next steps.

In 2015, Portland did just this with its N/NE Neighborhood Housing Plan. After using the city’s map of neighborhood change to identify North and Northeast Portland as areas experiencing significant gentrification and displacement pressure, the Portland Housing Bureau (PHB) organized community forums and outreach efforts in an effort to better understand the types of housing assistance that would be most helpful. Based on seven months of community engagement, the city developed a five-year plan for addressing the consequences of gentrification in the area, providing grants and loans for home repairs and mortgages, creating new affordable homes, and acquiring more land to be used for permanent affordable housing.

Seattle Displacement Risk Analysis

As a part of its 2015 Growth and Equity Analysis, Seattle’s Department of Planning & Development (DPD) analyzed displacement risk across the city. Similarly to Portland’s model, Seattle’s project layered neighborhood information on three factors: vulnerability of residents to rent increases and discrimination, proximity to amenities, and development capacity based on zoning rules. In order to measure these factors, DPD used public data primarily from the Census and ACS. The vulnerability category included data on education levels, demographics, and housing cost burden; the amenities category measured proximity to transit, core businesses, and civic infrastructure; and the development capacity category looked at zoning rules and median rent. Based on this data, Seattle assigned and mapped risk scores to each city neighborhood.

Seattle's Displacement Risk Analysis

Seattle’s project is unique in that it paired displacement risk data with an access to opportunity index, which identifies the level of economic opportunity in each neighborhood based on public data on education, access to jobs, transit availability, presence of civic infrastructure, and proximity to health facilities and healthy food. Combining displacement and opportunity ratings, Seattle categorized neighborhoods in four strata: high displacement risk/low access to opportunity, high displacement risk/high access to opportunity, low displacement risk/low access to opportunity, and low displacement risk/high access to opportunity.

Seattle's Access to Opportunity Index

Seattle’s project drew its own insights from the mapping effort. DPD’s report notes that displacement risk is greatest in neighborhoods of color and that the determinants of social, physical, and economic well-being are not equitably distributed.

The analysis also includes more policy guidance than any other mapping initiative. DPD conducted a policy impact assessment for each of four potential courses of action: continuing current trends, guiding growth to urban centers, guiding growth to urban villages near light rail, and guiding growth to urban villages near transit. The city assigned estimates of housing and job growth to each possible strategy, then compared those estimates against each neighborhood’s rating for displacement risk and access to opportunity. While not issuing any direct prescriptions, the report concludes that guiding growth to urban centers would likely cause the least displacement.

Boston’s Displacement-Risk Map

Because many cities have experienced similar challenges with gentrification and displacement, one city’s map can provide other cities with opportunities for replication. The City of Boston’s Department of Neighborhood Development took advantage of this opportunity, designing its 2017 Displacement-Risk Map based on Seattle’s Displacement Risk Analysis and Portland’s Susceptibility to Gentrification Model. Like both Seattle and Portland’s models, Boston’s analysis used mostly Census and ACS data on three factors indicative of displacement risk: vulnerability of renters including race, poverty, housing cost burden, and education; proximity to amenities like rapid transit, core businesses, and high income areas; and market changes such as decreases in affordable housing, rent appreciation, and commercial development.  

According to Amelia Najjar, Research and Development Analyst in Boston’s Department of Neighborhood Development and one of the creators of the city’s displacement map, “Seattle’s project was the starting point. We had Boston employees speaking with Seattle employees and asked Seattle to us send their data, indicators and methodology.” Boston mimicked Seattle’s approach of designing a map that tracked displacement risk, rather than historical neighborhood change, used parts of Seattle’s weighting scheme, and then “made changes based on what made sense for Boston,” said Najjar. Boston’s project illustrates the possibility for replication across gentrification and displacement mapping projects, even between communities with  drastic geographic and demographic differences.  

However, Boston took a unique multi-layer approach to creating its final map of displacement risk. The city first completed a tier 1 analysis that identified neighborhoods where more than 95 percent of housing units are deed restricted affordable housing. The model deemed these neighborhoods not at-risk of displacement, and removed them before beginning the tier 2 analysis, which layered the other three factors. The Department of Neighborhood Development assigned each indicator a weight based on historical correlation with displacement, layered them all in GIS, then produced a final risk score for each Census tract in the city.

By integrating these two tiers into its mapping effort, Boston was able to produce a more accurate picture of displacement risk. Before introducing the Tier 1 analysis, designers noticed that a number of block groups that consisted almost entirely of public housing units were appearing as at high risk of displacement. While such a neighborhood—which may have many low-income residents and sit near a transit line—might initially appear prone to displacement, existing availability of affordable housing can mitigate displacement risk. With information on affordable housing, the city can identify those neighborhoods where interventions are already succeeding in preventing the dislocation of low-income residents, and those where intervention is most needed.

Urban Displacement Project San Francisco Bay Area Displacement Risk Analysis

A 2015 study of the San Francisco Bay Area by the Urban Displacement Project sought to map displacement risk, provide analysis of trends, and offer potential policy solutions. The study drew primarily from Census and ACS data on housing, demographics, employment, transportation, land use, and policy as well as datasets on home sales from DataQuick and employment density from NETS. By pairing this government data with mobility data on low-income residents, the Urban Displacement Project produced a map that characterizes neighborhoods based on four levels of gentrification: not losing low-income housing, at risk of gentrification or displacement, undergoing displacement, and advanced gentrification.

Urban Displacement Project San Francisco Bay Area Displacement Risk Analysis

From this information, the Urban Displacement Project concluded that 48 percent of Census tracts and 53 percent of low-income households lived in neighborhoods at risk of or already experiencing displacement or gentrification pressures. These effects were not isolated to low-income neighborhoods, but rather many high-income neighborhoods that housed low-income households were also losing their low-income population. Moreover, transit investment, loss of market-rate affordable housing units, and spikes in housing prices were all identified as red flags for low-income housing displacement.

However, the study also found that the relationship between gentrification and displacement is more complicated than most might assume. In some cases, gentrification precedes displacement, but in others the relationship works in the opposite direction or the two are not correlated at all. Improved mobility, neighborhood revitalization, and other amenities can benefit locals if they bring in additional job and low-income housing opportunities. The study points to neighborhoods like San Francisco’s Chinatown, East Palo Alto, and Marin County as examples of neighborhoods that have avoided displacement through subsidized housing, tenant protections, and strong community organizing.

In providing analysis around displacement in the Bay Area, this study is more directive than traditional displacement risk maps. The map not only directs residents and policymakers to areas in need of intervention, but also identifies interventions that have been effective, providing an ostensible roadmap for creating economic growth in neighborhoods without displacing low-income residents.

The Association for Neighborhood and Housing Development’s Displacement Alert Project Map

Providing the most fine-grained analysis of displacement risk, the Association for Neighborhood and Housing Development (ANHD) created the Displacement Alert Project Map (or Dap.Map), which assesses displacement risk building-by-building in New York City. The analysis uses publicly-available data on 96,000 New York buildings and assigns each a risk score based on three factors: loss of rent-regulated units, rate of tenant turnover, and price the building sold for. After standardizing each factor on a scale of 100, ANHD added the three and assigned each building a combined risk score from 0 to 300, then displayed each score on an interactive map.

The Dap.Map

Using these factors, ANHD hopes to target the direct causes of displacement: higher prices based on speculative behavior and an influx of new residents. Armed with this easily-accessible information, residents, community groups, and the local government have the capacity to target displacement on a building-by-building level, expose abusive landlords, and fight back against the loss of affordable housing units.

Limitations

Importantly, displacement risk maps use proxies to estimate displacement trends and are therefore inexact measures of displacement. These tools use mobility data—information on changes in the demographic makeup in neighborhoods—as a proxy for displacement. If a neighborhood loses a portion of its low-income residents, the study will therefore assume that they moved due to displacement pressures. As a result, displacement risk maps may misinterpret economic mobility—low-income residents earning more and becoming mid- or high-income earners—as displacement.

Yet, cities can use data on general demographic trends in order to improve the accuracy of their displacement risk assessments. In their study of San Francisco, researchers from the Urban Displacement Project noted that in the Bay Area, Census tracts’ low-income population had grown overall. As a result, the study could fairly conclude that in neighborhoods experiencing losses in low-income households and stable population, low-income residents were likely moving out rather than earning more. While inexact, maps of displacement pressure remain useful guides for beginning to understand what factors cause displacement and how to mitigate harmful effects on low-income residents.

Neighborhood change and displacement risk maps are both useful resources for gaining an initial picture of housing trends, potential effects on low-income residents, and possible solutions. Once policymakers have identified areas where low-income residents appear to be at risk of losing their homes, they may organize street-level engagement campaigns like Portland’s N/NE Housing Plan in order to better understand the problems facing those neighborhoods. By then pairing data from gentrification tools with detailed resident-level knowledge from personal engagement, policymakers can better develop solutions tailored to individual neighborhoods.

Appendix: Comparison of Models

The table below outlines the factors included in each of the models described above, providing a starting point for cities interested in developing algorithms to map gentrification or displacement. The models differ greatly in both the number of factors considered and the fine-grainedness of the data used, and cities’ decisions of what typology to follow should depend on their goals and available data.

 

 

LA i-Team

UDP LA

Portland

Seattle

Boston

UDP SF

Dap.Map

Factors Included

 

 

 

 

 

 

 

Population, Population Density, or % Change in Population Density

 

X

 

 

 

X

 

% Affordable Housing

 

 

 

 

X

X

 

Projected Change Affordable Housing

X

 

 

 

 

 

 

% Residents that Rent

X

X

X

X

X

X

 

% Change in Renters

 

 

X

 

 

 

 

% White Residents

 

X

X

X

X

X

 

% Change in White Residents

X

X

X

 

 

 

 

% Residents Without College Degrees

 

X

X

X

X

X

 

% Change in Residents Without College Degrees

X

X

X

 

 

X

 

% Residents Without HS Education

 

X

 

 

 

 

 

% Change in Residents Without HS Education

 

X

 

 

 

 

 

% English Speaking

 

 

 

X

X

 

 

% Change in Household Size

X

 

 

 

X

 

 

Median Household Income

 

X

 

 

 

X

 

% Change in Median Household Income

X

X

X

 

 

X

 

% Low-Income

 

X

X

X

X

X

 

% Change in Low Income

X

 

 

 

 

X

 

% Cost-Burdened/Severely Cost-Burdened

X

X

 

X

X

 

 

% Change in Cost-Burdened

 

X

 

 

 

 

 

% Households Using Vouchers or Tax Credits

 

X

 

 

 

 

 

% Change in Households Using Vouchers or Tax Credits

 

X

 

 

 

 

 

Median Rent

 

X

 

X

X

 

 

% Change in Median Rent

X

X

 

 

X

X

 

% Change Property Sales

 

 

 

 

 

X

 

Median Home Value

 

 

X

 

 

 

 

% Change in Home Value

 

 

X

 

X

X

 

Projected Change in Home Value

X

 

 

 

 

 

 

Past Commercial Development

 

X

 

 

X

X

 

Potential Development Sites

 

 

 

X

X

 

 

Proximity to Bus

 

 

 

X

X

 

 

Proximity to Rapid Transit

X

 

 

X

X

 

 

Proximity to Businesses

 

 

 

X

X

 

 

Proximity to Civic Infrastructure

 

 

 

X

 

 

 

Proximity to Already-Gentrified or Affluent Neighborhood

X

 

 

X

X

 

 

Proximity to Job Center

 

 

 

X

 

 

 

Employment Density

 

X

 

 

 

X

 

Building Loss of Rent-Regulated Units

 

 

 

 

 

 

X

Building Rate of Tenant Turnover

 

 

 

 

 

 

X

Price of Building Sale

 

 

 

 

 

 

X

 
 

 

 

About the Author

Chris Bousquet

Chris Bousquet is a Research Assistant/Writer for Data-Smart City Solutions. Before joining the Ash Center, Chris worked at the Everson Museum in Syracuse, NY and wrote for DC Inno in Washington, D.C., where he covered tech policy, cybersecurity, and startups. Chris holds a bachelor’s degree from Hamilton College.

Connect

Comments

comments powered by Disqus