In recent years, higher socioeconomic groups started to have more children than lower socioeconomic groups in Sweden and in many other developed countries. Previous research focusing on individuals as independent units had limited power in explaining this polarization in fertility. We will thus use a social network perspective to understand the role of segregated social networks in the fertility behavior of different socioeconomic groups. Using longitudinal population-level social networks consisting of multiple types of relationships (i.e. neighbors, family, and workmates), we will answer (1) how fertility behavior spreads in the social networks for different socioeconomic groups, (2) how use of assisted reproductive techniques spreads in the social networks for the different socioeconomic group, and (3) how family support in someone’s social network influences fertility among different socioeconomic groups. Data will be obtained from several Swedish registers. Quasi-experimental methods (i.e., instrumental variable and small unit analysis) will be applied to distinguish whether people influence each other’s fertility behavior or simply select their relationship based on pre-existing similarities (e.g., fertility intentions). The project will be the first to investigate the role of social networks in forming socioeconomic fertility differentials based on population-level network data, contributing to the understanding of evolving socioeconomic inequalities.