In this paper, we investigate the effects of public family planning programs on fertility in Indonesia, controlling for economic growth and improvements in the status of women. A major tumbling block to measuring the impacts of family programs on fertility is that family planning programs are not randomly placed. Moreover, if, as in the case of Indonesia, the government allocates program inputs to areas of greatest need and updates the allocations based on performance, standard methods of controlling for this problem, such as fixed effects analysis, yield inconsistent estimates. We solve this problem by explicitly modeling the government’s allocation process together with the impact analysis. Using these methods, we find that total BKKBN expenditures on contraceptive subsidies lowered fertility by about 3-6 percent. By contrast, a 50 percent improvement in the distribution network (including public health clinics and community-based distribution systems) lowered fertility by about 12 percent. However, these policies are expensive; BKKBN's contraceptive subsidies represent roughly 50% of their annual expenditures, while the costs of the distribution network are paid largely out of other ministerial budgets.