Do the returns to migration extend beyond migrants themselves and accrue to the children of migrants? Drawing upon data from a unique 19-year longitudinal survey from Tanzania, this paper empirically investigates this question by exploiting the variation in the outcomes of the children of migrants and the children of the migrants' siblings who stayed behind conditional upon a range of individual characteristics of their parents. I show that parental migration has important implications for child development. This relation depends on the destination and the timing of the move. More specifically, children whose parents migrated from rural areas to cities are heavier, taller and more educated for their age. The effects on height and schooling are strongest for children who were exposed to the city environment during their early childhood. In contrast, children whose parents moved to a different rural village do not appear to experience any health advantage and those moving alongside their parents even start schooling at a later age. In addition to conferring a broader view of the returns to physical mobility, this analysis contributes to the debate on the origin of spatial inequalities in developing countries.