The paper tries to explain why women in the Philippines, as yet a low middle income country, obtain higher levels of education than boys. Four empirically based reasons are posited – the substantial expansion of the education system, the growth of job opportunities, the culture that encourages girls to develop better study habits and the high returns to their education. Empirical evidence is provided to support these contentions especially on the returns to women’s schooling. The study concentrates in estimating by various methods returns to schooling using individual observations from the labor force survey. The more conventional OLS regressions are first applied to allow comparison with many studies and the semi-parametric estimates. But the semi-parametric additive method had to be used in order to check for specification robustness of the model due to the observed violation of the OLS assumption of normal distribution of error terms. The quantile regression was also applied to reflect the income distribution implications of the returns pattern. An additional insight into the returns estimation is given by the inclusion of the effect of being married and marrying well, i.e., whether the spouses are equally or upward matched in education, or not. We find that returns to education are higher the higher the level of education is and that returns to women’s education are higher than returns to men’s education. Moreover, being married and married well increase earnings. Additionally, there is a fairly high good matching between education classes, i.e., there is substantial intermarriage among college graduates and other college educated and among lower educated individuals. This implies poor social mobility considering that access to education especially at the higher levels is very much constrained by family resources. Intermarriage between college graduates preserves their high social position since access to education is restricted by income. The paper concludes with a list of social issues that emerge from the findings.
