A study on regional return to education in South Korea: comparison of male and female wages

ABSTRACT

This study examines the return to education in South Korea by comparing metropolitan areas with non-metropolitan areas. It utilizes the Korean Labor and Income Panel Study from 2018 and 2019 for analysis, alongside the Mincer equation. The findings indicate that female workers have a higher return to education compared to male workers. The Oaxaca decomposition method reveals that private academies contribute to increasing differential treatment between men and women, whereas public education reduces the gap. Additionally, the regression discontinuity design method shows that the university premium is significantly different by region.

KEYWORDS:

• Human capital
• regional return to education
• Oaxaca decomposition
• regression discontinuity design

SUBJECT CLASSIFICATION CODES:

• E24
• I26

Acknowledgment

We express our gratitude to Donna Ginther, David Slusky, and ChangHwan Kim for providing us with valuable feedback on previous versions of this paper. We are truly grateful for the valuable feedback from the two anonymous reviewers. We would also like to extend a special thank you to Zongwu Cai, Shahnaz Parsaeian, Youjin Hahn, Jinwook Jeong, Jiyoung Park, Dongwoo Kang, Jack Osman, Llamosas Rosas Irving, Antonella Mancino, participants at the 62nd Western Regional Science Association Conference, 87th Midwest Economics Association, 60th Missouri Valley Economic Association, and seminar attendees at the University of New Hampshire for their helpful comments and suggestions. We are grateful to Ronald Oaxaca for his invaluable feedback.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Private tutoring is different from private schools, and is similar to test preps in the U.S.

2 According to the Housing Apartment Index in 2018 and 2019, it can be seen that the index of Seoul increased from 2018 (1.11) to 2019 (1.20) and both were significantly higher than that of Gyeonggi (1.01 & 1.01) and Incheon (0.99 & 0.98). The Housing Apartment Index data is provided by the Korea Real Estate Board.

3 The potential experience has proved to be very fittable in various empirical analyses (Heckman and Polacheck, 1974; Murphy and Welch 1990).

4 According to Becker (1981), Lam (1988), and Nakosteen and Zimmer (2001), if the income from marriage in the marriage market is a source of public goods for families, it mainly has a positive relationship. Thus, the spouse’s education year as a proxy variable should be premised that all wage workers are married. In addition, it is not an appropriate proxy variable that can be used at the current time when the marriage periods of wage workers are delayed. Therefore, this study uses the father and mother’s education years as family background variables for unobservable factors.

5 Kim (2013) emphasizes that the average method may not fully address the identification problem. As an alternative, he proposes the grand-mean centering method.

6 In order to make the adjusted score following the Hoekstra (2009), we subtract the actual test score from the universities’ admission minimum score. The recentered minimum score would be zero. In addition, an individual score for those who have entered a university shows positive value.

7 However, many good quality companies still remain located in the metropolitan area.

8 When analyzing only married women and dividing them into metropolitan and non-metropolitan areas, the results show a sharp decline in the return as they enter their 50 s.