| This study analyzes to what extent different mechanisms can explain the raw correlations between cohort sex ratios and female educational attainment in post-1949 China, and, in the process, sheds light on the most important proximate factors behind rising relative female educational attainment during this period.;After careful construction of cohort sex ratios from various China Population Census samples that accounts for systematic sampling bias, we find that the raw correlation between province-birth year cohort sex ratios and female educational attainment in post-1949 China is negative and significant for those born before the onset of China's "One-Child" fertility policies, and positive for those born after.;Various hypotheses that can explain these relationships are reviewed. To analyze the relative contributions of each mechanism, carefully chosen proxy variables are included in a linear model.;The results show that among the rural hukou population born between 1950 and 1975, the negative correlation can be explained by a measure of healthcare expansion, growth in health beds per capita. We hypothesize this measure corresponded with increased female returns to education through decreasing maternal mortality rates, and decreased sex ratios through female infant survival rates. In addition, the magnitudes of the estimates suggest that the aggregate increase in relative female education is largely explained by this factor.;For those born after 1975, during which the One Child Policies and economic reforms began, we find that among the rural hukou population, increasing female birth weight is associated with increasing relative female years of education, consistent with theories of female comparative advantage in skill-biased activities given marginal investments in nutrition. Meanwhile, among the urban hukou population, female birth weight is negatively associated with educational attainment, which reaches strong significance for the dependent variable junior college attainment. We hypothesize that this could be due to negative selection of female infants born to higher status households or a premarital investment mechanism. |
