The first part of our analysis looks at which neighborhoods in Dallas have high concentrations of residents who are vulnerable to displacement in the face of rising housing costs. Following the Austin study, we utilized a set of five key indicators to identify groups of residents who, according to social science research, are least able to absorb rising housing costs and who have limited options following displacement. These groups are: low-income households, people of color, heads of households without a bachelor’s degree or higher, renters, and families with children living in poverty. 

These indicators were measured in the following way:

• Percentage of households under 80% of Area Median Income (AMI)

• Percentage of the population that is not non-Hispanic white

• Percentage of households headed by someone without a bachelor’s degree or higher

• Percentage of households that are renting

• Percentage of families with children in poverty

Per our methodology, a neighborhood is considered vulnerable to residential displacement if it scores highly across the five demographic indicators. Each neighborhood was compared to all other neighborhoods by calculating a z-score for each variable. A z-score measures a data point’s relative position in a distribution. A tract was considered vulnerable to displacement if at least three of the five variables had a z-score above 0.5. The average of all five z-scores was also used to illustrate relative vulnerability in the Vulnerability Map.

The second part of our analysis assesses changes in demographics over time, indicating that vulnerable residents are moving out of the neighborhood while less vulnerable residents are moving in. We used four variables to measure demographic change: income, race and ethnicity, homeownership rates, and educational attainment. These variables were measured in percent change or percentage point change over a 10-year timeframe from 2011 to 2021.

Just as in the vulnerability analysis, z-scores for changes between 2011 and 2021 were computed for each tract. If at least two factors had a z-score above 0.5 and the average of all four scores was greater than 0.5, a tract was considered to have undergone demographic change. 

The third part of our analysis measures and classifies each neighborhood by the strength and growth of their housing markets. Housing market strength and change were evaluated using home sale prices, investment levels, and turnover rates. Home sale prices were calculated as the median single-family home sale price in the neighborhood. Levels of investment were calculated as the share of residential parcels with a building permit recorded (whether new construction, rehabilitation, or renovation) per household. Turnover rates were calculated as the count of residential home sales per owner-occupied household. These three measures of market strength were combined into a single market strength measure with factor analysis. The market strength factor was first estimated using 2021/2022 data, then calculated on the 2011/2012 data to measure change in the market strength factor over time using a consistent measure. 

All neighborhoods were first assessed using the three variables above. If a neighborhood could not be classified using home sale prices because there were too few sales,14 it was then classified using median rent levels from the American Community Survey (ACS).

We then used these variables to identify and classify each census tract as one of the following five types of potential market categories: 

• A Sustained tract has had a strong housing market in both periods. These tracts were in the top 40% of the market strength factor in 2011/2012 and remained in the top 40% of the market strength factor in 2021/2022; or they were in the top 40% of median gross rent in the 2008/2012 ACS and the top 40% of median gross rent in 2018/2022.

• An Appreciated tract moved from low or moderate housing market strength to high housing market strength. These tracts were in the bottom 60% of the market strength factor in 2011/2012 but increased to the top 40% of the market strength factor in 2021/2022; or they were in the bottom 60% of median gross rent in the 2008/2012 ACS and the top 40% of median gross rent in 2018/2022. 

• An Accelerating tract is increasing rapidly in market strength but has not yet appreciated. These tracts were in the bottom 60% of the market strength factor in 2011/2012 but rose at a rate in the top 20% of tracts from 2011/2012 to 2021/2022; or they had an increase in median gross rent in the top 40% 16 of all tracts from 2008/2012 to 2018/2022. 

• An Adjacent17 tract has low or moderate housing strength but is next to one or more tracts with housing market pressure. Market change often moves from one area to neighboring areas, and therefore ‘Adjacent’ tracts are likely the next places that could experience housing market change. These tracts were in the bottom 60% of the market strength factor in 2011/2012, did not rise at a rate in the top 20% of tracts from 2011/2012 to 2021/2022, but were adjacent to a ‘Sustained,’ ‘Appreciated,’ or ‘Accelerating’ tract. Tracts were considered ‘Adjacent’ if they shared at least 50% of the tract border with ‘Sustained,’ ‘Accelerating,’ and ‘Appreciated’ tracts.18

• A Stable tract did not meet any of the above criteria. These tracts are not currently at risk of gentrification or displacement but should be watched for market changes in the future.

These three variables – vulnerability, demographic change, and housing market strength and change – were then used to assign each neighborhood to one of the following six gentrification stages, per the Austin adaptation of the Bates methodology.

• Susceptible neighborhoods are near changing areas but have not yet experienced demographic change. These are neighborhoods that could have displacement if market pressure moves from a changing area nearby into these neighborhoods. Tracts that have vulnerable populations, are in the ‘Adjacent’ market category, but have not yet experienced demographic change are considered ‘Susceptible.’

• Early: Type 1 neighborhoods are in the early phase of neighborhood change. The market has begun putting displacement pressure on residents of these neighborhoods, but the neighborhoods have not shown demographic change yet. Tracts that have vulnerable populations, are in the ‘Accelerating’ market category, but have not yet experienced demographic change are considered ‘Early: Type 1.’

• Early: Type 2 neighborhoods have not yet had market changes but have begun experiencing demographic change. Tracts that have vulnerable populations, are in the ‘Adjacent’ market category, and have experienced demographic change are considered ‘Early: Type 2.’

• Dynamic neighborhoods have begun market changes and are experiencing demographic change. Tracts that have vulnerable populations, are in the ‘Accelerating’ market category, and have experienced demographic change are considered ‘Dynamic.’

• Late neighborhoods are well into the process of neighborhood change. Tracts that have vulnerable populations, are in the ‘Sustained’ market category, and have experienced demographic change are considered ‘Late.’

• Historic Loss neighborhoods have already changed over a longer timeframe. Tracts that experienced loss of Black or Hispanic population and increase in non-Hispanic white population from 2000 to 2020 or 2010 to 202020 are considered ‘Historic Loss.’

One key indicator of increased displacement pressure is the presence and location of targeted public investments and placed-based economic incentives such as Tax Increment Financing (TIF) Districts, seen overlaid here.

TIF Districts are intended to revitalize underperforming real estate markets and encourage increased private investment within a defined area, but often create or exacerbate displacement pressures for existing residents in vulnerable communities.

Most of the neighborhoods classified as Dynamic in the southern sector of the city fall within or adjacent to established TIF districts.