Marriages between blood relatives – also known as consanguineous unions – are widespread in North Africa, Central and West Asia and most parts of South Asia. Researchers have suggested that consanguinity has adverse effects on child development, but assessing its impact is not straightforward as the decision to marry a relative might be endogenous to other socio-economic factors. Using a unique dataset collected in rural Pakistan, this paper assesses the extent to which consanguinity is linked to child cognitive ability and nutritional status. As economic benefits of marrying cousins may lead to upward bias to estimates of the effects of consanguinity on child outcomes, prior work likely underestimates the negative impacts of consanguinity on child outcomes. After controlling for current household wealth and parent education, this paper exploits (current and past) grandfather land ownership and maternal grandparent mortality to identify the effect of endogenous consanguinity on child cognitive ability and height-for-age. Children born into consanguineous unions have lower cognitive scores, lower height-for-age, and a higher likelihood of being severely stunted. More importantly, adverse effects are significantly larger after accounting for the endogeneity of consanguineous unions, suggesting that negative impacts on child development are substantial, and likely to be larger than suggested in previous studies. Reducing incentives for consanguineous unions should therefore be of concern among policy makers aiming at improving child development outcomes where marrying cousins is common.