A Quantitative Model for Mapping the Consequences of Public Housing Demolitions

Among a city’s assets, few are as critical as its housing stock. Because it is so essential, a city’s housing supply can pose difficult questions for city planners and policymakers. In the mid-1990s, such was the case in the city of Chicago, where officials debated — and ultimately approved — a plan to demolish thousands of public housing units. Enabled by tens of millions of dollars in federal funding,¹ demolition crews razed buildings in 59 neighborhoods, with most of the work happening between 1995 and 2010 (the final year that federal money was available). Did the demolitions, which amounted to one of the largest such programs in the U.S., improve welfare for Chicago and its surrounding communities?

At first glance, it may seem that the displacement caused by demolitions, in which public housing tenants often relocated to other neighborhoods, would have been the main factor influencing neighborhood welfare. But in their paper, “Urban Renewal and Inequality: Evidence from Chicago’s Public Housing Demolitions,” three researchers uncover other dimensions to consider when assessing how neighborhoods adapt to a postdemolition landscape.

Drawing on data maintained by government sources (including the U.S. Census Bureau and the Chicago Housing Authority), the authors — University of Chicago Assistant Professor of Economics Milena Almagro, University of Texas at Austin Associate Professor of Economics Eric Chyn, and Philadelphia Fed Economic Advisor and Economist Bryan A. Stuart — studied the impacts of demolitions of public housing units in Chicago across all census tracts in Cook County, Illinois. All told, their analysis covered households living across 1,240 tracts, with an average population of roughly 4,000 residents per tract. Their findings indicate that tracts with higher levels of demolition underwent notable changes in their demographic and socioeconomic characteristics. They observed the following:

• The higher the intensity of demolition in a tract, the larger the subsequent increase in its share of White residents (and correspondingly, the lower its share of minority residents, with the share of Black residents falling particularly sharply).
• An increase in demolition activity in a tract is associated with a subsequent uptick in household income within that tract.
• Tracts with higher demolition rates experienced faster rent growth than tracts with no demolitions.

Having documented these shifts in neighborhood makeup (among other transitions), the authors decomposed the forces driving such changes. They designed a model that delves deeply into the postdemolition choices that households make when deciding where to live (focusing tightly on choices that involve neighborhoods in or near recent demolition sites). Importantly, they track housing preferences across racial groups, which they sort by income, distinguishing between poor and nonpoor households.

“To compare preferences across groups,” the authors write, “we calculate [their] implied willingness to pay for different neighborhood characteristics.” The calculations show that residents prefer, among other things, neighbors of the same race and income group. Black and Hispanic residents, for instance, display an implied willingness to pay higher rents to live in areas that represent their ethnicities. Furthermore, members of all racial groups willingly pay more to live in areas with higher-income residents and fewer public housing units.

Over the course of several additional estimations, certain households are shown to generally be better off after demolitions, as measured by an approach that expresses changes in household welfare in dollar terms. Poor White households, for instance, derive an increase in welfare that is comparable to a $230 decrease in their annual rent. At the same time, the authors point out that indeed some households are worse off; Black and Hispanic households on the lower end of the income spectrum, for example, tend to experience a decrease in welfare after demolitions.

Delving into additional applications of their model, the researchers investigated how demolitions affected neighborhoods in census tracts that did not contain demolition sites. By doing so, they show how the effects of demolitions (especially in terms of housing costs) radiated beyond their immediate neighborhoods. Spillover effects from demolitions appeared to be felt citywide, although neighborhoods farther from demolition sites experienced a weaker increase in housing costs: Prices in tracts within 0.1 mile of a demolition site appreciated by 2.5 percent (on average), while neighborhoods farther afield — 20 miles away — recorded gains of only 1 percent.

In light of the broad and lasting effects of public housing demolitions, the authors assert that it’s only natural to ask if the demolition sites were eventually redeveloped. And if so, what effect did that redevelopment have on the neighborhood? After all, a substantial portion (37 percent) of census tracts withstood the demolition of more than half of their housing stock, raising questions about how they recovered in the wake of such significant culling. “The original plan developed by the [housing authority],” the authors say, “was to create ‘mixed-income’ housing … provided through new construction of both public housing and market-rate units.” But as the authors report, redevelopment was slow: Thirty-five percent of the land stood vacant as recently as 2015.

Using their neighborhood-by-neighborhood approach, the authors estimate the effects of redeveloping the land, focusing on the merits of building new housing (whether through private or public channels). They conclude that such efforts are worthwhile, with even moderate redevelopment (amounting to 20 percent of destroyed housing, for instance) proving to be beneficial, especially for lower-income minority households. Why is this so? Among other reasons, the authors point out that redevelopment increases the supply of housing in neighborhoods that were, prior to demolitions, among the least expensive in the Chicago area — neighborhoods that provide vital residential options for lower-income households.

Taken together, the findings described in “Urban Renewal and Inequality: Evidence from Chicago’s Public Housing Demolitions” show that public housing demolitions have consequences that aren’t immediately apparent. Housing markets are particularly affected, and housing costs begin an upward climb that continues for years following a demolition. The modeling also shows that postdemolition neighborhoods often undergo demographic and socioeconomic changes as housing markets adjust and households decide where to relocate. The research provides compelling evidence of the little-studied ways in which demolitions change their immediate and surrounding neighborhoods — ways that go beyond visible, physical transformations.