We study the effect of residential segregation on fertility for the socially excluded and marginalized Roma ethnic minority. Using original survey data we collected in Serbia, we investigate whether fertility differs between ethnically homogeneous and mixed neighbor- hoods. Our results show that Roma in less segregated areas tend to have significantly fewer children (around 0.9). Most of the difference arises from Roma in less segregated areas waiting substantially more after having a boy than their counterparts in more segregated areas. We account for the endogeneity of the level of segregation using (il)legal possibility to build in the area at the time of its creation as an instrument. We find that the true gap due to segregation is actually larger than that estimated by OLS (around 1.4). We finally provide evidence that exposure to the Serbian majority culture is the main mechanism at play, as opposed to differences in opportunity cost of time, migration patterns, family arrangements and returns to education.