This paper exploits an exogenous shift in the trade policy in India to study the impact of industrialization on son preference. Using a difference-in-differences strategy, we find that households are more likely to have a male child in regions with higher trade openness relative to regions with lower trade openness. Moreover, higher trade openness seems to have affected only the Hindu households; there is no analogous effect on the Muslim households. We further analyze the underlying mechanisms through which industrialization might have affected relative survival of daughters. We find a significant increase in real dowry payments in regions experiencing greater trade openness. Most interestingly, dowry inflation is experienced by the Hindu households, but not by the Muslim households. The results are robust to falsification tests using cohorts born much before the liberalization period and are not driven by systematic migration into areas with greater trade openness.