Family planning programs have become critical elements of most developing country population control policies. Despite the generally widespread acceptance of the key role of family planning programs in reducing fertility, there is little consensus regarding the programs importance relative to the roles of general economic development and improving women;s status. To address this dialogue, we propose to investigate the following questions: To what extent can be recent fertility decline in Indonesia be explained by the following factors: improved development, and improvement in women's education and economic status? Which elements of the Indonesian family planning program have the greatest impact on fertility: the distribution information, education and communication activities? Do the effects of family planning program activities vary for different groups? For example, does fertility decline more quickly when public family planning program infrastructure is expanded in communities that are in earlier stages of the fertility transition that when infrastructure is expanded in areas at later stages of the transition? How have the expansion of the public family planning program and private family planning services, economic development, and improvements in women's status influenced the major proximate causes of fertility: contraceptive use, marriage and postpartum amenorrhea? Are there time lags in the impact of family planning program components on fertility and its proximate causes? To what extent do expansion of public program components, better access to private family planning services, contraceptive methods to more effective methods? In other words, to what extent do family planning programs affect contraceptive mix versus prevalence? Our methods unify he demographic proximate determinants and economic reduced-form models of fertility. Applications of the proximate determinants model for program evaluation have often yielded different conclusions about family planning programs than have applications of reduced-form "demand for children" models. We demonstrate the compatibility of the two approaches and develop methods to directly compare their results. We will also address the statistical problem raised by the fact that policy maker allocate program inputs for optimal impact, rather than randomly. To control for this endogeneity of program placement, we will use a community fixed effects estimator. The results are interpreted as the effects of changes in the program inputs on changes in fertility within communities.