We here address the causal relationship between maternal depression and child human capital in a UK cohort study. To do so, we exploit the conditionally exogenous variations across mothers' genomes in an instrumental variable approach and describe the conditions under which genes can be used as valid instruments. We show that an additional episode of maternal depression between the child's birth up to age 9 reduces both their cognitive and non-cognitive skills by 20 to 25% of a SD throughout adolescence. Our results are robust to a battery of sensitivity tests addressing, among others, concerns about pleiotropy and genetic inheritance.