This is an excerpt of Chapter 2 of Tackling Vacancy and Abandonment: Strategies and Impacts After the Great Recession, jointly produced by the Center for Community Progress, the Federal Reserve Bank of Atlanta, and the Federal Reserve Bank of Cleveland. It has been lightly edited and condensed for the web. In this chapter, Austin Harrison and Dan Immergluck share research that reveals the unevenness of the recovery from the Great Recession across the Rust Belt and Sun Belt. Click here to download the full chapter and read more insights on tackling vacancy and abandonment from the nation’s leading experts.
For as long as America has had houses, it has had empty houses. Seasonal vacancies, vacancies created by new builds or homes for sale, or rental vacancies: Whatever the cause, the reason, or the duration, empty houses are inevitable. Following the subprime mortgage crisis beginning in 2007, which resulted in unprecedented increases in the number of vacant homes, the empty house conversation became central to the American housing policy conversation. More than 10 years later, we apply a regional lens to changes in vacancy during the broader housing recovery since 2012. We have focused on medium-sized and large metropolitan areas in two regions of the country–the Sun Belt and the Rust Belt–that were generally hit particularly hard by the foreclosure crisis and experienced high levels of long-term housing vacancy at the beginning of the 2010s.
The U.S. housing market recovery that began around 2012 brought with it increased housing demand and generally lower levels of housing vacancy. This recovery, however, was highly uneven, with population and home values growing much more in some regions than in others. Our research focused on the extent to which the number of hypervacant neighborhoods (those with vacancy rates of 8 percent or higher) in these metro areas had declined by 2019. We have also examined the racial and poverty characteristics of such neighborhoods. It is in these neighborhoods where the cumulative negative impacts of vacancy are expected to be the most severe and where the problem of vacancy is often the hardest to solve.
Overall, we found that in the Sun Belt, in contrast to the Rust Belt, the share of tracts that were hypervacant declined over the 2012 to 2019 period, from about 10.2 percent to 6.6 percent. There was also a sizable increase in the share of tracts that fell into the low-vacancy (under 1 percent) category, from 36.4 percent to 51.6 percent.
Meanwhile, in the Rust Belt metro areas, hypervacant tracts remained roughly constant, falling only from 15.6 percent to 15.4 percent. Notably, the share of hypervacant tracts was still more than 50 percent higher in the Rust Belt in 2019 than in the Sun Belt in 2012, before the broader national recovery. And the share of hypervacant tracts in the Rust Belt in 2019 was 2.3 times the share in the Sun Belt in 2019. The Rust Belt did see a net decrease in vacancy, but it was primarily from tracts in the moderate and high levels shifting to the moderate or low levels while the share of tracts at the more extreme levels remained roughly constant.
Despite the greater persistence of hypervacant neighborhoods in the Rust Belt, such neighborhoods also exist in the Sun Belt to a significant degree. This is primarily because the Sun Belt also includes a substantial number of low-cost, low-growth metro areas, the type that tend to have the highest numbers of very high- and extreme-vacancy census tracts. In both regions, these types of metro areas saw their shares of tracts with very high or extreme vacancy levels remain about constant over the 2012 to 2019 period. This potentially supports the idea that larger regional factors are not as significant for hypervacancy as metro-level market factors are, such as cost or growth.
Low-growth metro areas do comprise a substantially smaller share of the Sun Belt metro areas than of the Rust Belt metro areas. For example, there are 41 high-growth metro areas in the Sun Belt, but only 6 in the Rust Belt. Since Sun Belt metro areas tend to be higher growth, they tended to see larger declines in vacancy, including declines in the number of very high- and extreme-vacancy tracts.
We also found that neighborhoods with higher poverty rates and/or larger Black populations were more likely to suffer from hypervacancy, especially in Rust Belt metro areas. This fact suggests that historical and current forces of segregation and discrimination may explain the existence and persistence of hypervacancy. Disinvestment remains an especially potent force, both in Rust Belt metro areas and in lower-growth metro areas elsewhere, and remains heavily racialized. Black neighborhoods continue to be generally undervalued compared with other neighborhoods by appraisers, lenders, and other actors in the real estate market. Without stronger policy interventions, including the increased enforcement and expansion of the Fair Housing Act and the Community Reinvestment Act, the forces of discrimination and segregation are likely to reinforce and perpetuate the racialized nature of hypervacancy.
Our study demonstrates that metropolitan housing market trends are strongly related to the resilience of neighborhoods when it comes to long-term vacancy rates. Whether in the Rust Belt or the Sun Belt, metropolitan growth and cost structures during the 2012 to 2019 period appear to have had a strong influence on whether, and to what degree, the very high and extreme levels of neighborhood vacancy persisted. Moreover, our findings challenge any oversimplified notion that weak market regions are predominantly located in the Rust Belt and show that, in weaker-growth Sun Belt metro areas, high levels of persistent hypervacancy remained a problem throughout the broader national recovery.