Abstract
This paper examines the rapid fall of South Korea's fertility to lowest-low level, focusing on rising levels of education as a contributing cause. The fall occurred mainly because of major declines in parity progression from woman's own birth to first birth and from first to second birth, combined with major postponement of first births accounted for mainly by major increases in age at first marriage. A decomposition analysis indicates that changes in population composition by woman's education account for only about seven percent of the decline in the total fertility rate between 1995 and 2005. This figure probably understates substantially the contribution of rising levels of education to the fertility decline. We speculate that rapidly rising levels of education additionally contributed to fertility decline through increased competition for good jobs and greater investment in children's education through private cram schools.
Acknowledgements
Partial support for this research was provided by a grant obtained by the Nihon University Population Research Institute from the ‘Academic Frontier’ Project for Private Universities, matching fund subsidy from MEXT (Ministry of Education, Culture, Sports, Science and Technology), 2006–2010. We thank Gayle Yamashita—computer specialist at the East-West Center—for computer programming assistance.
Notes
1. The term ‘ultra-low fertility’ has also been used. See Jones et al. (Citation2008).
2. This more modest shrinkage of 15 percent, compared with the 35–40 percent mentioned in the preceding paragraph, is due to a temporary population bulge in the reproductive ages that develops as the proportion of children declines, leading to a temporary continuation of population growth that demographers refer to as ‘population momentum’.
3. The value of 2.1 is approximate since the replacement level depends not only on the level of fertility, but also the level of mortality.
4. The age-specific proportions of singles (never-married) from which SMAM is calculated depend on age-specific first marriage probabilities pertaining to not only the current year, but also previous years.
5. The formula for calculating TFRpppr is: TFRpppr=p 0 +p 0 p 1+p 0 p 1 p 2 +p 0 p 1 p 2 p 3 + p 0 p 1 p 2 p 3 p 4+ /(1 − p 4+ ).
6. Mean age at first childbirth can be computed directly from a life table for the B–1 transition, or by adding mean age at first marriage and mean first birth interval as calculated from separate life tables for the B–M transition and the M–1 transition. The resulting estimates are not exactly the same. The numbers in were computed using the former approach, based on the B–1 transition.