Spring/Summer 2020 Featured Research Profile
Spouse and house? GCER Fellow Minsu Chang connects the dots between the two.
The prospects of marriage and divorce have changed dramatically over the past decades. For instance, the average marriage rate for singles between ages 25 and 64 fell by more than 30%. The average divorce rate more than doubled from the 1970s to the 1990s. Yet, during this period the rate of homeownership remained roughly constant until the recent house price boom.
In her paper entitled “A house without a ring: the role of changing marital transitions for housing decisions,” Georgetown economist and GCER fellow Minsu Chang addresses this puzzle. She provides a novel quantitative framework that shows how marital transitions are a major factor in housing decisions in the United States.
Chang’s research takes a close look at both homeownership rates and the composition of assets across individuals with different marital statuses. Her work reveals a major shift: non-elderly singles were much more likely to be homeowners, while young couples were less likely to lock their assets in housing.
Professor Chang then builds and estimates a structural life-cycle model to account for these observations. In the model, households consider the prospect of marriage and divorce when making decisions. They decide how much to consume, rent, work, and save in non-housing and housing assets. A homeowner incurs substantial transaction costs whenever he/she wants to adjust house size. This is especially important during times of marital transitions when the current house is no longer suitable after a marriage or divorce.
Chang’s mechanism complements other channels, such as borrowing constraints and labor market uncertainty, to explain the evolution of aggregate housing demand. Her analysis produces three key findings. First, comparing 1970 and 1995, 29% of the increase in homeownership by singles can be accounted for by the change in marital transitions. Second, the decrease in married individuals’ housing asset share cannot be generated without incorporating observed changes in marriage and divorce probabilities. Lastly, when extending the analysis to the mid-2000s, a period that includes the housing price boom, Chang shows that the continuing decrease in marriage contributed to a 7% increase in young singles’ homeownership rate.