LAUSR

Abstract The past four decades have witnessed an enormous increase in modern contraception in most low‐ and middle‐income countries. We examine the extent to which this change can be attributed to changes in fertility preferences versus fuller implementation of fertility preferences, a distinction at the heart of intense debates about the returns to investments in family planning services. We analyze national survey data from five major survey programs: World Fertility Surveys, Demographic Health Surveys, Reproductive Health Surveys, Pan‐Arab Project for Child Development or Family Health, and Multiple Indicator Cluster Surveys. We perform regression decomposition of change between successive surveys in 59 countries (330 decompositions in total). Change in preferences accounts for little of the change: less than 10 percent in a basic decomposition and about 15 percent under a more elaborate specification. This is a powerful empirical refutation of the view that contraceptive change has been driven principally by reductions in demand for children. We show that this outcome is not surprising given that the distribution of women according to fertility preferences is surprisingly stable over time.