Abstract
We provide a historical decomposition of fertility in the United States by family type. We find that intergenerational coresidence was systematically associated with lower fertility than nuclear families, with the di erence shrinking over time. This pattern is robust to controlling for several demographic and socioeconomic confounders. We build a simple, analytical model and show that a theory featuring both endogenous fertility and endogenous coresidence can rationalise the observed cross-family fertility di erence. Simulations from a calibrated dynamic general equilibrium version of the model show that the model has the right qualitative behaviour, and is quantitatively meaningful. Using individual data, we discuss (and dismiss) several potential alternative explanations.
We provide a historical decomposition of fertility in the United States by family type. We find that intergenerational coresidence was systematically associated with lower fertility than nuclear families, with the di erence shrinking over time. This pattern is robust to controlling for several demographic and socioeconomic confounders. We build a simple, analytical model and show that a theory featuring both endogenous fertility and endogenous coresidence can rationalise the observed cross-family fertility di erence. Simulations from a calibrated dynamic general equilibrium version of the model show that the model has the right qualitative behaviour, and is quantitatively meaningful. Using individual data, we discuss (and dismiss) several potential alternative explanations.
