This paper studies how gender inequalities in survival probability and educational attainment are shaped by severe adverse shocks and their interaction with social norms in the setting of China’s Great Famine. We propose a novel and unified model to show how cohort sex ratios and gender inequality in schooling investment were shaped by China’s Great Famine and its interaction with the culture of son preference. Using the data from 2000 China Population Census, we exploit regional variations in the culture of son preference and cohort variations in exposure to the famine to empirically estimate the impact of famine and its interacting effect with the culture of son preference on cohort sex ratios and the gender gap in education. We find that famine shocks lead to a reduction in t...[ Read more ]

This paper studies how gender inequalities in survival probability and educational attainment are shaped by severe adverse shocks and their interaction with social norms in the setting of China’s Great Famine. We propose a novel and unified model to show how cohort sex ratios and gender inequality in schooling investment were shaped by China’s Great Famine and its interaction with the culture of son preference. Using the data from 2000 China Population Census, we exploit regional variations in the culture of son preference and cohort variations in exposure to the famine to empirically estimate the impact of famine and its interacting effect with the culture of son preference on cohort sex ratios and the gender gap in education. We find that famine shocks lead to a reduction in the fraction of males among survivors and an increase in the gender gap in years of education, but both are less true in regions with a culture of son preference. Our results suggest that the culture of son preference buffers the impact of famine on cohort sex ratios and the gender gap in years of education. Our additional empirical evidence using CFPS 2010 on height (endowments) verifies the potential channel in our theoretical framework. We also show that our results are not driven by selective fertility in response to the famine.