Existing research shows that the quality of life urban residents experience varies significantly within and across U.S. metropolitan areas. In other words, people's health, well-being, and chances for prosperity depend on where they live. Nowadays, there is widespread consensus among scholars and urban planners that, while individuals clearly have choices in where to seek housing, those choices are often constrained. On the one hand, research has focused on how, for example, racial and ethnic group differences in resources and preferences play a role in persistent residential segregation patterns at the neighborhood level. On the other hand, few empirical studies have examined what shapes persistent spatial dynamics in housing markets at the metropolitan level. Put differently, we know a lot more about what shapes differences in quality of life within metropolitan areas than across them. To that end, this project uses audit data from the 1989, 2000, and 2010 waves of the Housing Discrimination Study, sponsored by the U.S. Department of Housing and Urban Development. The PIs merge these data to tract- and metro-level information from the 1980 to 2010 U.S. Censuses and the 2006 to 2010 American Community Survey to examine how metro-level housing market dynamics shape contemporary residential patterns, as well as changes in housing patterns since the 1980s.
Broader Impacts
This timely study seeks to inform understanding of housing patterns in U.S. cities, many of which have experienced significant changes since 1980, and in particular since the recent housing crisis. Findings may be of interest to the general public (lenders, owners, renters, advocacy groups) and policy makers. In addition, the project facilitates student training and data dissemination goals. The project will support a graduate assistant who will be involved in all aspects of data acquisition, analysis, and writing. The research team will disseminate metadata concerning the data set constructed for this study via the Kunz Center for Social Research at the University of Cincinnati, on whose website the team also expects to post relevant policy briefs.
Existing research shows that the quality of life urban residents experience varies significantly within and across U.S. metropolitan areas. In other words, people’s health, well-being, and chances for prosperity depend on where they live. Nowadays, there is widespread consensus among scholars and urban planners that, while individuals clearly have choices in where to seek housing, those choices are often constrained. On the one hand, research has focused on how, for example, racial and ethnic group differences in resources and preferences play a role in persistent residential segregation patterns at the neighborhood level. On the other hand, few empirical studies have examined what shapes persistent spatial dynamics in housing markets at the metropolitan level. This project sought to answer two main questions: first, what, if any, is the correlation between measures of residential segregation and measures of housing discrimination at the metropolitan area level? Second, does this correlation strengthen or weaken when including other characteristics of metropolitan areas that may be correlated with both housing discrimination and segregation? To address these questions, we analyzed audit data from the 2000, and 2012 waves of the Housing Discrimination Study (HDS), sponsored by the U.S. Department of Housing and Urban Development. We merged the HDS files to metropolitan area-level information from the 2000 U.S. Decennial Census and the 2008 to 2012 American Community Survey. The resulting data covered some 24 metropolitan areas, 10 of which were sampled in both 2000 and 2012. Results from this study were mixed. On the one hand, we found that, using three measures of segregation and six measures of discrimination, most of the correlations were in the expected direction; that is, more discrimination was correlated with more segregation. In addition, controlling for the percentage of mostly-white neighborhoods strengthened the correlation between discrimination and segregation, as was expected. On the other hand, the small number of metropolitan areas led to low statistical power, meaning that we could not rule out the possibility that the correlations we observed were due simply to chance variation in the metropolitan areas that were selected for study in the HDS. The results of this study are important because they inform both academic researchers and policy makers about one underexplored cause of persistent residential segregation in American metropolitan areas. And, unlike group differences in residential preferences, housing discrimination is likely a cause of segregation that can be affected by public policy. For example, HDS data show a steep decline in measures of discrimination from 1990 to 2000, and this may have occurred due to amendments to the Fair Housing Act in 1988 which took hold during the 1990s. Finally, the work our team did on cleaning and shaping the 2000 and 2012 HDS data files will pay dividends in the future to researchers who wish to analyze further the causes and consequences of housing discrimination in American cities.
