We introduce a general framework to analyze the trade-off between education and family size. Our framework incorporates parental preferences for birth order and delivers theoretically consistent birth order and family size effects on children's educational attainment. We develop an empirical strategy to identify these effects. We show that the coefficient on family size in a regression of educational attainment on birth order and family size does not identify the family size effect as defined within our framework, even when the endogeneity of both birth order and family size are properly accounted for. Using Danish administrative data we test the theoretical implications of the model. The data does not reject our theory. We find significant birth order and family size effects in individuals' years of education thereby confirming the presence of a quantity-quality trade off.
