We develop a theory of fertility, distinguishing its intensive margin from its extensive margin. The deep parameters are identified using facts from the 1990 US Census: (i) fertility of mothers decreases with education; (ii) childlessness exhibits a U-shaped relationship with education; (iii) the relationship between marriage rates and education is hump-shaped for women and increasing for men. We estimate that 2.5 percent of women were childless because of poverty and 8.1 percent because of high opportunity cost of childrearing. Over time, historical trends in total factor productivity and in education led to a U-shaped response in childlessness rates while fertility of mothers decreased.

We test Le Play's (1875) hypothesis that the French Revolution contributed to France's early fertility decline by imposing equal partition of inheritance among all children, including women. We combine new data on local inheritance rules before the Revolution and individual-level demographic data from historical sources and crowdsourced genealogies. Difference-in-differences and regression-discontinuity estimates show that the inheritance reforms enacted during the Revolution reduced completed fertility by 0.5 children. A key mechanism was the desire to avoid land fragmentation across generations. These reforms closed the fertility gap between regions with different historical inheritance rules and crucially contributed to France's demographic transition.