A study of fertility and labor force participation of married women in Korea

Abstract

The objective of this study is to determine empirically the relation between the labor force participation and fertility of married women in Korea. This relation is one key to the long run population policy and short term manpower policy of the country. In specifying the simultaneous equation model used, the study follows the lead of previous researches in the United States but also considers the special characteristics of Korea. Among the variables included the opportunity earnings of women and the cost of children are emphasized as important policy variables. Two types of data are used: (1) individual, cross-sectional data and (2) aggregate time-series data. In the cross-sectional analysis, a set of structural equations are specified in which the lifetime supply of work, fertility, quality of children, and the potential market earnings of a woman are simultaneously determined, given a number of exogenous variables (e.g., the characteristics of the individual, the family, and the surrounding labor market). Using the 1974 KNFS data, the model is estimated by two broad age groups, separately for rural and urban areas. The wages of urban females vary with their attained educational level. And their labor force participation is responsive to wages. On the other hand, the wages of rural females depend on other factors such as the presence of children and husband's educational level. However, urban or rural, more highly educated younger females work less in the labor market. In addition, younger females substitute quality for quantity of children as their actual or expected market earnings increase. Based on these results, the following policies would increase the labor force participation of the female population: higher wages for urban females, more varied job opportunities for rural females, and both for younger females independent of location. Higher wages are also suggested to achieve a further reduction in the fertility of Korean women. In the time series analysis, the simultaneous equation system used in the cross-sectional estimation is transformed into a set of reduced-form equations in which the trend in the cost of children, the wage rate, household income, and business conditions jointly determine labor force participation and fertility over time. The cost of children, derived by comparing the monthly expenditure of nuclear households with different number of children, takes the place of the quality-of-children variable used in the cross-sectional analysis. Estimation from a set of aggregate time series data confirms the overall negative effect of the cost of children on fertility and the overall positive effect of opportunity earnings on labor force participation. However, it fails to confirm the overall negative effect of the female wage on fertility. The real wage in the manufacturing industry, used as a proxy for the female wage, shows a positive effect on fertility as well as on labor force participation. This anomaly is attributed to a high correlation of the variable with growth and the rapidly accelerating household income from growth.