Return of Education for Women across Socio-Economic Status: Using Quantile Regression and Machado-Mata Decomposition Methods for Turkey

Abstract

The purpose of this paper is to identify gender-based wage differentials through wage distribution. Although studies on gender-based wage differentials with the quantile regression method have been carried out before, this study contributes to the literature by way of gender-based wage differentials as determined by the Machado-Mata decomposition method. In this paper, TurkStat's 2017 Household Budget Survey data is examined. The total sample size is 10,073 respondents aged 15 years old or older. Our findings show that a significant portion of the wage differential is based on characteristic which should exist in the labor market and little is due to gender-based; but, it should not be ignored that due to gender-based wage differentials part. Nevertheless, it is found that gender-based wage differential is higher in low-income individuals than high-income individuals. This is important for women in low-income groups who are already disadvantaged. This paper reveals that the return on education and experience on wages is higher for women with low socio-economic status.