Year of Publication



Martin School of Public Policy and Administration

Degree Name

Master of Public Administration

Committee Chair

Dr. JS Butler

Executive Summary

South Korea was once a typical high birth country. However, Korea is facing serious low birth rates recently as the total fertility rate is 0.98, which means that the Korean couple has less than one child on average. The rapid demographic change due to low fertility causes many negative effects such as economic vitality and welfare burden. Thus, the Korean governments have introduced various policies to overcome the low birth rate, and one of these is the childbirth grant policy: governments provide a certain amount of cash when a family within its jurisdiction gives birth. However, the world's lowest fertility rates of Korea seldom show signs of rising. So, the problem statement is whether the birth grants policy is effective in increasing childbirth in Korea. The independent variable is the childbirth grant policy, and the dependent variable is the fertility rate and the number of births. Control variables affecting their relationship are classified into economic factors such as income level, childcare facilities, and medical facilities, and social and demographic factors such as divorce rate, marriage rate, the age of first marriage, and elderly population. I used panel data regression. The panels are 25 autonomous districts in Seoul Metropolitan City from 2015 to 2018. Using these panels, I estimated the effects of childbirth incentives on birth rate have differed over the period. As a result of the estimation, the coefficients of childbirth grants are negative on both the total fertility rate and the number of childbirths statistically significant. The results are contrary to general prediction that the fertility rate will increase if there is a large amount of childbirth subsidies. However, in reality, it is the opposite causal relationship: the birth rate is reverse-causing the grant. The childbirth grants cannot affect people’s birth decision; rather, the district's low fertility rate makes local governments spend more childbirth grants. It can be interpreted that policy makers are aware of the region's low birth rate and then have increased the amount of childbirth subsidies to increase the rate. Among the control variables, social and demographic factors are found to have a major impact on the dependent variables: the elderly population ratio and average first marriage age affect dependent variables in the direction of negative, while the marriage rate has a positive impact. However, the economic factors are statistically insignificant, and do not affect childbirth. Fundamentally, if people desire to have children, they are doing that, by getting married, possibly at younger ages. Only a change in people’s goals, or much more money than is currently allocated, would have the intended effect. However, this paper has limitations; first, local areas have large effects (high rho values in the estimation), so there is a limit to generalizing the results on a nationwide because there are considerable fixed effects for each autonomous district; second, there was a limit to collect relevant data when there were countless factors affecting childbirth decisions, which can be a factor that undermines the internal validity of the variables; and the last is the limits of the data, which may be problems with accurate estimates because it was analyzed based on only four years of data.