The clich¨¦ ¡°once a homeowner, always a homeowner¡± is not true. We study the causes of terminations of spells of first-time homeownership.Using a national panel data set, we find that the likelihood of a household terminating a spell of homeownership is predictable at the time of purchase. Specifically, the lower the probability score that a household becomes an owner at the time of purchase, the greater the likelihood of termination of the subsequent ownership spell. This finding suggests that post-purchase counseling programs can be targeted toward those most at risk at the time of home purchase. We also find that post-purchase events affect the likelihood of termination. Important factors include changes in household earnings and wealth, house value, unemployment rates, family size, and marital status. There are substantial racial differences in termination rates. Some of these differences are explained by differences in household characteristics at the time of home purchase, and some by differences in post-purchase events or households¡¯ reactions to them.
