The nation’s homeownership rate is affected by both the length of time that households remain as homeowners rather than reverting to being a renter and the length of time households rent or live with their parents following the termination of a period of homeowning. Thus, differences in ownership rates between whites, African Americans, and Hispanics could be due, in part, to differences in their durations of owning, renting, and living with parents. Studies of this issue have been piecemeal, with none being comprehensive. The analysis presented here reveals that it is not true that “once an owner, always an owner”. Also, the rate of exiting from homeownership differs substantially by race and ethnicity of the owner. There has been little recognition that post-ownership minorities are particularly unlikely to return to homeownership.

This study uses a national data set (the National Longitudinal Survey of Youth-NLSY) that follows a cohort of individuals for 21 years. Their residence histories are tracked, measuring the time spent in each type of tenure. This data set provides extensive information about the socio-demographic characteristics of individuals, allowing one to study the determinants of the length of an ownership spell, or a rental spell. The data are weighted to make the sample nationally representative.

The analysis begins by distinguishing between the length of stay in a particular dwelling and the duration of stay in one or more owned homes. All existing published studies focus on the length of stay in a dwelling (either owner or rented), or on the time to mortgage default. These studies focus on the time spent in a particular dwelling unit, not the length of time in a particular “state of the world” such as owning or renting. While these studies are of interest for some questions, they do not provide the needed information to determine whether different lengths of continuous spells of owning or renting contribute to racial gaps in homeownership rates.

The empirical analysis in this study consists of two parts. The first presents descriptions of ownership and rental spells, focusing on differences by race, ethnicity, and income. The second part is a formal econometric analysis of spell length, identifying which factors contribute to lengthening spells of owning and which ones contribute to reducing spells of renting or living with parents that follow spells of ownership.