How Family Leave Policies Influence the Employment Patterns of High-Educated VS. Low-Educated Korean Women?

Consequences of family leave policy for women’s employment remains controversy both in theoretical and empirical discussions partly because relationship between leave arrangements and women’s work decisions is complicated. It is especially unclear whether a “paid” leave policy do foster women’s labor force participation after childbirth. Since prior studies about impacts of leave policy mainly focused on Western countries, this study of impacts of leave policy in Korea—where a significant portion of women exit the labor market at the primary childbearing ages due to persistent adherence to traditional gender norms and social barriers for family to combining work and childcare—will contribute to expand our understanding about policy effects on women’s employment. To intensify maternity protection for female workers and guarantee their job security after childbirth, Korean government reformed leave policy in several aspects: extended the maximum length of paid maternity leave from two to three months; started to provide flat-rate benefits for parental leave in 2001; and doubled the amount of parental leave benefits in 2004. This study explores whether and how women’s employment changed subsequent to these reforms. This study uses the data drawn from the 1999-2007 Korea Labor and Income Panel Study, which allow us to take advantage of quasi-nature experiments because we can compare employment patterns of three “policy era” categories of women, those who gave a birth before the 2001 reform, during the transition, and after the 2004 reform. In identifying impacts of reforms, this study employs Difference-in-Difference approaches in logit analyses. Using this approach this study compares changes in employment outcomes after policy reforms among two groups of women: childless women (those not affected by reforms) and women having young children (those affected by reforms), who are classified into seven categories based on motherhood status. Finally, as another strategy of verifying the robustness of the results as well as controlling for other aspects of policy reforms, this study adds other comparing groups of women who may be differently affected by the reforms, providing stratified DD estimates based on levels of women’s education. This study controls for women’s human capital, economic needs, demographic characteristics, availability of informal childcare, husband’s socioeconomic status, and economic conditions. Findings from the multivariate logit analysis suggest that pregnant women benefited more in the transition period and women with a youngest child under one year old in the post-reform period. In addition, the impacts of 2001 policy on likelihood of employment are statistically significant and higher for pregnant women and women who have the youngest child under one year old and who have one child ages of one or two among those with low levels of education in the transition period; and for women have the youngest child under one year old in the post-reform period, when women have low levels of education. Findings suggest that women with low education benefited more in the transition period whereas the differences in the interaction effects of policy and motherhood status for low education and high education are reduced in the post-reform period.