Research Information System
International Journal of Manpower, 2025 (SSCI)
Purpose: Significant changes have been observed in Türkiye labor market over the last two decades, driven by the substantial increase in the number of universities and students and by factors such as economic crises and unemployment. This phenomenon has also affected the number of overeducated (undereducated) wage earners in the labor market. In this study, we first consider compulsory education policy to calculate the educational mismatch. Our aim is to analyze the wage returns of overeducated (undereducated) workers employed in the public and private sectors. The pooled data from the TurkStat Household Labour Force Survey for 2021 and 2022 are used. Design/methodology/approach: Unconditional quantile regression (UQR) was introduced by Firpo et al. (2009). The UQR method has become very popular and is used in several areas, such as labor economics, health economics and public policy, to examine income inequality, discrimination by gender and discrimination by race and others (Agyire et al., 2018; Gaeta et al., 2023). Firpo et al. (2009) used the recentered influence function (RIF) to estimate the outcome equation (wage equation in our case) to obtain UQR estimates. The RIF regresses the influence function (IF) which is developed by Hampel (1974) (Rios-Avila, 2020). Findings: We find that overeducated workers in the Turkish labor market received a wage penalty relative to required educated workers. The returns to years of required education are positive, which means that the returns to each year of required education have higher wages than each year of their overeducated counterparts employed in the same occupation. In the private sector, the effect of the increase in one year of education on wages in the lower percentile is lower than in the public sector for both overeducation, undereducation and required education. Originality/value: An important contribution of our study is the inclusion of the compulsory education policy in the calculation of the educational mismatch in Türkiye. The study uses data from the Household Labour Force Survey (HLFS) covering the period 2021–2022. Previous studies on the wage effects of educational mismatch in Türkiye have typically addressed this issue in the context of conditional wage distribution. However, wages can vary significantly at different points in the distribution. An important advantage of the UQR is its ability to provide comparable coefficient estimates across the wage distribution.