I am great fan of Greg Kaplan’s paper `Moving Back home’ in which he motivates moving back with parents as an insurance mechanism against labor market risks. I think, apart from explaining consumption and savings responses of low income households, it highlights one of the most important economic determinants of household size for not only the US but also for other countries. When no other insurance mechanisms are available, people tend to huddle up economizing on costs of living and minimizing labor market risk through skill set diversification (much like portfolio diversification!). This explains, for example, why traditional societies like India had joint family systems for a very long period of time to the extent of becoming a defining feature of Indian society. The important thing to realize is that this system may not be here for long. The general increase in incomes, economic opportunities, and capacity to shield oneself from economic risks will reduce the need for a joint family. This is already evident in the increasing nuclearization of families across urban India. Moreover, the size of the households also might change over the business cycle!
A recent paper with José-Víctor Ríos-Rull and Sebastin Dryda, extends the idea of moving back home with parents to build a model of household size determination that is amenable to equilibrium business cycle analysis with aggregate technology shocks. The authors then use this model to explain the discrepancy between the micro evidence on the Frisch elasticities and the ones implied by the macro models. Interesting stuff!