It is often argued that child labour comes at the expense of schooling and so perpetuates poverty for children from poor families. To test this claim we study the effects on children's labour force participation and school enrollments of the pure school-price change induced by a targeted enrollment subsidy in rural Bangladesh. Our theoretical model predicts that the subsidy increases schooling, but its effect on child labour is ambiguous. Our empirical model indicates that the subsidy increased schooling by far more than it reduced child labour. Substitution effects helped protect current incomes from the higher school attendance induced by the subsidy.