Association for Public Policy Analysis & Management

Homeownership rates grew nationally by a factor of .04 between 1990 and 2005, and declined back to their 1990 rates by 2014 (US Census Bureau, 2014). Some of the decline is attributed to foreclosure waves and some to the more stringent post-recession underwriting standards, both of which most severely affected lower-income families. The intriguing fact, however, is that homeownership rates also dropped, and at even higher marginal rates, for families with no significant financial constraints. Survey results show that homeownership is still considered the superior choice, even after the recession (Drew & Herbert, 2012), so what could explain the decline of homeownership rates among the financially-capable families? One hypothesis is that what we observe is not a wholesale withdrawal from homeownership, but rather a menaingful delay: The financially-secure families (“owner-ready,” O-R) wait longer than before to become homeowners, and that explains lower ownership rates. This hypothesis is informed by the research that shows migration trends back to urban areas with better job opportunities and amenities (e.g. Kolko, 2009; Ehrenhalt, 2013), which at times can impede or delay homeownership for various reasons. In this research I test for any changes in time-to-ownership after the recession, and ask how much of the delay can be explained by the preference to live in areas with higher quality of life (QOL).

The study sample is a panel of families across the US, who start as renters in 1999 and qualify for Fannie Mae’s conventional 30-year fixed-rate mortgage in their census-tract/county of residence each and every year. The primary data is from the Panel Study of Income Dynamics (PSID) between 1999 and 2013, which I combine with data on QOL elements such as jobs, amenities, foreclosure, and neighborhood characteristics from various sources. In the first part of the analysis, I use logit models to measure the effect of QOL on the likelihood of homeownership. Results show that O-R families are significantly less likely to become homeowners after the recession, especially in areas with higher-paying jobs. Average local income and education, on the other hand, seem to increase the likelihood. In the second analysis, I utilize survival models to see whether time-to-ownership is affected by QOL. Results from parametric loglogistic models show that assuming state and family fixed-effects, time-to-ownership increases by a factor of .42 after the recession. The delay is more pronounced in areas with more FIRE-industries’ jobs, whereas areas with more educated and affluent residents reduce the delay. These results imply that at least part of the drop in homeownership rates among the O-R is due to extended waiting periods after the recession. The most impactful QOL elements are high-end jobs, average local education and income. However, since the results do not show an all-ecnompassing effect for all QOL elements, the question about the appeal of homeownership among O-R families remains worthy of further research. In the upcoming phase of this research I am studying the effect of fear-of-foreclosure by comparing across different housing markets and the way they were affected by the recession.