Are Housing Vacancy Rates a Good Proxy for Physical Blight?

Adam Berland
Ball State University

Housing vacancy and physical blight pose challenges in U.S. cities, particularly in shrinking cities of the Rust Belt. Although some cities have begun collecting fine-scale data on physical blight, most cities do not have detailed blight data. Researchers and policymakers may need to rely on coarser-scale data such as neighborhood vacancy rates as a proxy for physical blight. To explore whether housing vacancy rates are a reasonable proxy for physical blight, data from a comprehensive parcel condition survey in Toledo, Ohio, were used to characterize residential vacancy and physical blight at the property parcel scale. These data were then aggregated to the census tract scale, and rates of vacancy and physical blight from the local parcel survey were compared with broadly available tract-scale estimates of vacancy from the U.S. Postal Service (USPS) and the U.S. Census Bureau’s American Community Survey (ACS). Results demonstrate that although USPS and ACS vacancy estimates were closely correlated with vacancy rates from the local survey, both sources overestimated vacancy in high-vacancy areas of Toledo relative to the local survey. For all three data sources, tract-scale vacancy rates were strongly and positively correlated with physical blight (r ≥ 0.73), and there were no significant differences in the strength of correlation among the data sources. These results suggest that, despite modest overestimation of vacancy in high-vacancy areas, tract-scale housing vacancy data from USPS or ACS can be used as a proxy for physical blight in the absence of more detailed local data. By comparing patterns in three relevant data sources, this study helps work toward just and informed decisionmaking related to housing vacancy and physical blight.