The impact of abortion reforms on the human capital of subsequent generations, including health, has been documented in several developed countries. However, (i) evidence relative to the impact on health in early life is not unanimous, (ii) there is no evidence for a developing country, where health human capital is low, and (iii) existing econometric studies have been unable to disentangle the effect of abortion reform on individual behavior from that on the composition of mothers.

In this paper, I exploit variation in the timing and location of newly introduced legal abortion centers in Nepal to estimate the effect of reducing the psychological and financial cost of abortion on fertility, investments in prenatal care, neonatal mortality, and sex- selection. Consistent with the prediction that the opening of a legal abortion center nearby reduces the cost of abortion, I find that the probability of a live birth conditional on conception decreases by 8.3 percentage points (9 percent of the mean), for a given mother. Similarly, the unconditional monthly probability of a live birth decreases by 0.4 percentage points (21 percent of the mean). However, there is no evidence that improved access to abortion increases observed investments in antenatal care and little evidence that it increases unobserved investments in neonatal health.

Contrary to findings in Lin et al. (2008) for Taiwan, abortion re- form in Nepal does not appear to have led to sex-selective terminations. On the contrary, there is some evidence that it may have led to a decrease in sex-selective abortion, which could be due to the substitution of first-trimester legal abortions to illegal abortions at a later gestational stage.