U.S. development assistance represents a signi?cant source of funding for many population programs in poor countries. The Mexico City policy, known derisively as the global gag rule, restricts activities of foreign NGOs that receive such assistance. The intent of the policy is to reduce the use of abortion in developing countries — a policy born entirely of U.S. domestic politics, which turns on and o? depending on the political party in power. I examine here whether the policy achieves its aim, and how the policy a?ects reproductive outcomes for women in Ghana. Employing a woman-by-month panel of pregnancies and woman-?xed e?ects, I estimate whether a given woman is less likely to abort a pregnancy during two policy periods versus two non-policy periods. I ?nd no evidence that any demographic group reduces the use of abortion as a result of the policy. On the contrary, rural women signi?cantly increase abortions. This a?ect seems to arise from their increased rate of conception during these times. The policy-induced budget shortfalls reportedly forced NGOs to cut rural outreach services, reducing the availability of contraceptives in rural areas. The lack of contraceptives likely caused the observed 12% increase in rural pregnancies, ultimately resulting in about 200,000 additional abortions and between a half and three-quarters of a million additional unintended births. I ?nd that these additional unwanted births have signi?cantly reduced height- and weight-for-age, relative to their siblings. Rather than reducing abortion, this policy increased pregnancy, abortion, and unintended births, resulting in more than a half-million children of signi?cantly reduced nutritional status.