Neighborhood Networks and Program Participation

We explore the existence of social interactions in program participation within small neighborhood networks. Our population of interest is pregnant women and their participation in Medicaid during pregnancy. Using geographically detailed data, we show that a pregnant mother is substantially more likely to participate in Medicaid, if recently pregnant mothers on her exact census block also received Medicaid benefits. To deal with endogenous sorting into neighborhoods, we only compare mothers across small neighborhoods within a broader geographic area defined as an agglomeration of nearby census blocks. The reflection problem is avoided by restricting peer groups to only mothers who have recently given birth. We also document substantial heterogeneity in the estimated network effect across various dimensions. Furthermore, increased Medicaid participation seems to translate into healthier behavior among pregnant women with earlier and more intensive participation in prenatal care.