Abstract: Reforms that expand reproductive rights change behavior only when they change effective access. We study Nepal, which legalized abortion in 2002 to reduce unsafe procedures, and ask when and where legal access affected the sex composition of births in a society with son preference. We combine quasi-random variation in firstborn sex with the staged rollout of abortion services and predetermined connectivity to city services in a triple-difference design. The rollout separates two margins. During the initial first-trimester rollout, later-birth hazards declined where services were reachable, but the male- and female-birth responses were not statistically distinguishable, and the sex composition of births did not change. Once access expanded to include second-trimester services, medication abortion, and midlevel providers, the annual probability of a female birth fell while male births did not decline. Nationally, the probability of a female birth fell by 0.49 percentage points, and the female share of later births in firstborn-girl families fell by 2.83 percentage points. The responses were concentrated within two hours of a city. Within this radius, the female birth probability and the female share declined by 1.30 and 6.52 percentage points, respectively. Back-of-the-envelope calculation maps the national composition estimate to an annual flow equivalent of about 11,900 missing girls at birth, or 1.91 percent of all births.

View Latest Version

Abstract: In the US, age-adjusted mortality rates are higher in 'red' states, i.e., states with high support for the Republican Party. We ask whether this is attributable to state-level policies as opposed to confounding variables such as culture. Using a variety of empirical approaches, we find that state government explains between 0 and 20% of the mortality gap. Scaling this by the size of the (large) gap, red state policies increase mortality risk by 0-3% relative to blue state policies.

View Latest Version

Abstract: While son preference in India is well documented, measuring gender differences in children's consumption remains challenging because consumption is typically observed only at the household level. We estimate gender-specific child consumption using a collective household model and data from rural India. We find substantial within-household inequality: boys consume systematically more household resources than girls. Translating these allocations into individual-level poverty, roughly 63.5 percent of girls are poor, compared to 56.3 percent of boys. These gaps account for age- and sex-specific needs, persist throughout childhood, and are largely invisible to standard household-level poverty measures. We further show that son-biased fertility stopping behavior contributes to the gender gap in child poverty by increasing girls' exposure to larger families with fewer resources per child, explaining about one-fifth of the overall gap. Finally, combining our model-based poverty estimates with administrative mortality data, we show that hidden gender gaps in child poverty can account for one in five Indian missing girls.

View Latest Version

(Includes guest lectures in R.)

(Includes discussion sections.)

(Includes discussion sections.)

(Includes discussion sections.)