Numerous papers report a negative association between parental divorce and child outcomes. To provide evidence whether this correlation is driven by a causal effect, we exploit idiosyncratic variation in the extent of sexual integration in fathers' workplaces: Fathers who encounter more women in their relevant age-occupation-group on-the-job are more likely to divorce. This results holds also conditioning on the overall share of female co-workers in a firm. We find that parental divorce has persistent, and mostly negative, effects on children that differ significantly between boys and girls. Treated boys have lower levels of educational attainment, worse labor market outcomes, and are more likely to die early. Treated girls have also lower levels of educational attainment, but they are also more likely to become mother at an early age (especially during teenage years). Treated girls experience almost no negative employment effects. The latter effect could be a direct consequence from the teenage motherhood, which may initiate an early entry to the labor market.