Income Inequality in Cities: How Much is Explained by Land Use and Housing Typology?

• In Event: Impacts of Local Land Use Policies

Abstract

It is a well known fact that income inequality has been rising over the past four decades.  In the U.S., the starkest income disparities have shown up in cities.  In this paper, we first establish that most of the change in income inequality in cities between 1990 and 2020 is explained by variation within cities over time rather than variation across cities.  Using the Thiel index to decompose the source of inequality among U.S. metro areas, we observe that, by 2020, nearly 90 percent of the metro-level income inequality is explained by the within-metro inequality.  Furthermore, there is quite a bit of variation in inequality within each metro area.  While the trend among communities (measured as PUMAs) in any single metro area is generally towards increasing inequality, there are still many sub-metro areas that have maintained the same level of inequality or declined.  These set of facts motivate our main question: what explains this within-metro variation in income inequality over time?

The second part of the paper tries to tease out one possible set of drivers of within-city inequality: land use and the housing typology.  Citywide inequality can be affected by the flexibility with which the municipality can respond to economic changes and provide shelter for a range of household types and incomes.  We expect that a city’s land use regime and residential building typology will mediate the degree of income inequality over time.  First, less restrictive land use regulations will allow for a more responsive housing market to ensure a higher volume and more diversity in housing production.  Second, since land use regimes change slowly, existing housing infrastructure should also provide an informative indication of the city’s capacity to support an economically diverse population.  For example,  places with a diversity of housing options, such as single family houses and larger multi-family apartment buildings, of varying quality tiers, may mitigate income inequality at the city level.   We will test for the impact of land use and housing typology on income inequality in a sample of cities, with varying growth and economic profiles, across the U.S.  

We adapt measures of income inequality and housing segregation to generate metrics of housing typology diversity (for example building off of Airgood-Obrycki et al. 2025, Freemark et al. 2023 and Aladangady et al. 2017).  We use established metrics of land use restrictiveness, such as housing supply elasticity (Baum-Snow and Han 2024), density of as-of-right development and a land availability index (Lutz and Sand 2023).  Using a four-decade panel of PUMAs across a sample of large and mid-sized U.S. cities and rich data on land use and housing typologies from ACS, IPUMS NHGIS and WSF-3D satellite data, we test for correspondence between within-metro income inequality and housing diversity.  Finally, we implement a community-targeted skill premium shock (following Fogli et al. 2025), to test for the responsiveness of the local housing and land use regime to changes in the localized demand for space.

Authors

• Rachel Meltzer, Harvard University
  Non-Presenting Co-Author

• Pooya Ghorbani, The New School
  Presenting Author