Wage returns to different levels of schooling in Turkey

Abstract

In this paper, I estimate the returns to the different levels of educational attainment, such as completing at least middle school, at least high school, and college for men and women in Turkey by using a fuzzy regression discontinuity design. I exploit the 1997 reform in which extended compulsory schooling from 5 to 8 years in Turkey as a source of exogenous variation in schooling across birth cohorts. My main finding is that the returns to schooling are increasing for higher levels of education. I estimate that the returns to completing different levels of schooling are higher for women than for men. The estimated average causal effects for completing at least high school and college are statistically significant if the sample excludes the birth cohorts 1986-1987 in which there is a fuzziness in the treatment. The possible reasons for higher returns for women than for men are the policy pushes a higher fraction of men to complete middle school and high school whereas pushes a higher fraction of women to complete college, and the rate of women who have a college degree is higher than the rate of college graduates among men in the labor market.