We provide a novel interpretation of the estimated treatment effects from evaluations of parental leave reforms. Accounting for the counterfactual mode of care is crucial in the analysis of child outcomes and potential mediators. We evaluate a large and generous parental leave extension in Austria exploiting a sharp birthday cutoff-based discontinuity in the eligibility for extended parental leave and geographical variation in formal childcare. We find that estimated treatment effects on long-term child outcomes differ substantially according to the availability of formal childcare and the mother's counterfactual work behavior. We show that extending parental leave has significant positive effects on children's health and human capital outcomes only if the reform induces a replacement of informal childcare with maternal care. We conclude that care provided by mothers (or formal institutions) is superior to informal care-arrangements.