Education, Preferences, and Household Welfare

Using census date from Nepal, we examine how the marginal effects of male and female education on various household welfar indicators vary with education levels. Parental education is associated with better household outcomes, but marginal effects vary with education level. Higher child survival, for instance, is associated higher primary education for mothers and higher secondary education for fathers. We calculate conditional makrginal effects that correct for assortative matching of spouses and compare them to unconditional estimates. The two differ because mother and father education are partial sustitues. We also show that the marginal effects of education have fallen over time while education levels were rising. Using the relative scarcity of women in the marriage market as proxy for the wight of female preferences in household choices, we find thta educated mothers prefer better educated children, but also prefer their children to work, possibly becuase the yare more likely to work themselves.