Theories about neighbours’ influence on children's education that are based on social capital, cohesion, and disorganisation stress the importance of neighbourhood stability. This is because stability is regarded as necessary for building strong ties and friendships, which in turn affect educational outcomes. However, amongst the vast number of studies on the effect of neighbours on a child's education, none has tested whether neighbourhood stability matters. We fill this gap by estimating the causal effect of residential turnover on student test score gains. Estimation is based on administrative data on four cohorts of secondary school students in England, allowing us to control for pupil-level, neighbourhood-level, and school-by-cohort level unobservables and for changes in neighbourhood composition driven by students’ residential mobility. We show that a high turnover of same-school-grade students reduces value added for teenagers who stay in their neighbourhood, although turnover of other age groups does not matter. These results coupled with auxiliary findings based on survey data suggest that neighbours’ turnover damages education through the disruption of local ties and friendships, highlighting a so-far undiscovered spillover of mobility.