Fig. 6.1
Trends in the total fertility rate (TFR) for Organization for Economic Cooperation and Development (OECD) member countries, 1960–2012 (OECD 2014a)
At 6.0 children per woman, South Korea’s TFR was particularly high in 1960, well above fertility levels in all other OECD countries except Mexico and Turkey. From that point, fertility in South Korea dropped extremely quickly, dipping to below 1.3 in 50 years. By 2012, South Korea had gone from the third highest fertility level in the OECD to one of the very lowest. Fertility also dropped rapidly in Mexico and Turkey, but has not dipped below 2.0 children per woman in either country. Between 1960 and 2012, the TFR in South Korea dropped by 4.77 children per woman. Among OECD member countries, only Mexico experienced a fertility decline of greater magnitude. Some other Asian countries, such as Taiwan and Singapore, experienced a similar pattern of very high fertility in 1960 dropping to very low levels by 2012 (Fig. 6.2). The difference in TFR over the 52-year period was 4.48 for Taiwan and 4.47 for Singapore. Nevertheless, the magnitude of South Korea’s fertility decline was the greatest in the region.
The third special feature of fertility in South Korea is that the TFR has remained at a very low level—less than 1.3 children per woman—for many years. As of 2012, only nine of the 34 OECD member countries, plus Singapore and Taiwan, had ever experienced fertility less than 1.3 (Table 6.1). Fertility remained at this low level for 12 years (2001–2012) in South Korea, compared with 11 years in Italy (1993–2003) and Slovenia (1995–2005), 10 years in Singapore (2003–2012), 9 years in Taiwan (2004–2012), 4 years in Germany (1992–1995), and 3 years in Japan (2003–2005).
Table 6.1
Number of years, as of 2012, with TFR at less than 1.3 children per women, selected OECD and Asian countries
Country | Years with TFR less than 1.3 | Time period |
|---|---|---|
South Korea | 12 | 2001–2012 |
Italy | 11 | 1993–2003 |
Slovenia | 11 | 1995–2005 |
Spain | 10 | 1993–2002 |
Singapore | 10 | 2003–2012 |
Taiwan | 9 | 2004–2012 |
Greece | 8 | 1996–2003 |
Slovak Republic | 8 | 2000–2007 |
Germany | 4 | 1992–1995 |
Poland | 4 | 2003–2006 |
Japan | 3 | 2003–2005 |
Demographic and Socio-Economic Implications of Very Low Fertility
Demographic Implications
Statistics Korea (2011a, b) has projected the size of South Korea’s population under various fertility and mortality scenarios. Under the medium scenario, Statistics Korea assumes that TFR will increase from 1.23 in 2010 to 1.42 in 2045 and then remain at that level. Between 2010 and 2060, life expectancy at birth is assumed to increase from 77.2 to 86.6 years for men and from 84.1 to 90.3 years for women. Over the same period, the net annual immigration rate is assumed to decrease from 1.67 to 0.53 per thousand. Given these assumptions, South Korea’s population is projected to reach a peak of 52 million in 2030 and then start to decline, falling to 44 million in 2060.
We extended this population projection from 2060 to 2100, assuming that fertility and net immigration rates will remain constant at 2060 levels and that life expectancy at birth will increase by 2100 to 89.3 years for men and 93.2 years for women. Given these assumptions, South Korea’s population will shrink to 28 million by 2100.
A breakdown of these forecasts by age group reveals a pattern of extreme population aging (Fig. 6.3). The population of young people, age 0–14, will decrease from about 8.0 million in 2010 to 4.5 million in 2060 and 2.8 million in 2100. The working-age population, age 15–64, will peak at about 37 million in 2016 and then fall to 22 million in 2060 and 13 million in 2100. Beginning in 2020, the elderly population will increase rapidly, as the baby-boomer generation, born in the 5-year period around 1960, reaches age 60 and above. South Korea’s population age 65 and above will increase from 5.5 million in 2010 to about 12.7 million in 2030 and 18.0 million in 2049. At that point, the size of the elderly population will start to fall along with the decline of South Korea’s population overall. The population age 65 and above is projected to drop to 11.9 million by 2100. By contrast, the proportion of elderly people in the total population is projected to increase steadily throughout the twenty-first century, rising from 10 % in 2009 to 20 % in 2026, 30 % in 2037, 40 % in 2060, and approaching 50 % in 2100. This pace of population aging is one of the fastest in the world.
Fig. 6.3
Age structure of South Korea’s population in 2010 (a) and projections for 2060 (b) and 2100 (c) (Statistics Korea 2011a)
Economic Implications
As the working-age population is predicted to increase until 2017 and then start to decline, the size of the labor force in South Korea will rise from 23.5 million in 2008 to 24.8 million in 2019 and then drop to 17.0 million in 2050. Between 2000 and 2010, South Korea’s labor force increased at an annual rate of 0.97 %, making a strong contribution to economic growth. By contrast, the labor force is projected to decrease by an annual rate of −0.60 % between 2020 and 2030, by −1.31 % between 2030 and 2040, and by −1.65 % between 2040 and 2050 (Moon et al. 2004). The impact on the labor force of very low fertility in the 2000s will appear in earnest from 2020 onward. The first baby boomers number 7.1 million, including 5.3 million in the labor force, and they started retiring in 2010. Between 2010 and 2018, 3.1 million workers will retire, but only 0.6 million will enter the working-age population. The resulting labor shortage is estimated at 0.6 million in 2015 and 1.5 million in 2020 (Bang et al. 2004).
The age structure of the working-age population will also be affected by the rapid pace of fertility decline. The proportion of the working-age population at age 50–64 will increase from 24.7 % in 2010 to 37.0 % in 2030 and 39.1 % in 2050 (Fig. 6.4). Accordingly, the average age of South Korean workers will increase from 39.1 years in 2010 to 42.0 in 2030 and 43.0 in 2050.
The aging of South Korea’s labor force is expected to decrease labor productivity. Some studies have shown that labor productivity peaks between age 30 and 50 and then goes down due to physical decay, human-capital depreciation, and obsolescence (Vandenberghe and Waltenberg 2010). Moon and colleagues (2004) found that after controlling for human and physical capital, an increase of 1 % in the proportion of workers age 55 and above decreases labor productivity by an estimated 0.09–0.17 %.
Declining fertility is also expected to reduce domestic demand for a variety of goods and services. The Korean Medical Association (2008), quoting from the Health Insurance Review and Assessment Service, notes that the number of obstetrics and gynecology hospitals and clinics decreased by 8.5 % between 2003 and 2007, and the number of pediatric hospitals decreased by 7.8 %. In addition, retirement of the baby-boom generation will decrease the demand for housing and other goods and services. In South Korea, only a small proportion of workers receive a pension. Rather, the elderly tend to hold a large proportion of their assets as home equity, and they are likely to sell their homes to fund their retirement. As these houses come on the market, Bae and colleagues (2007) predict that the number of new houses to be constructed annually will decrease from 471,000 units in 2008 to 451,000 in 2020 due to rising supply and falling demand.
The predicted drop in domestic demand may lead to a decrease in jobs, imbalance across industries, and an economic recession. Hwang (2009) estimates that changing consumption due to low fertility and population aging could lead to a loss of about 90,000 jobs in 10 years. The average annual growth of South Korea’s gross domestic product (GDP) may well slow down—from 4.0 % in 1995–2011 to a predicted 1.0 % in 2030–2060 (Johansson et al. 2012). The pace of the economic slowdown will accelerate from about 2020 onward, when South Korea’s baby boomers start retiring in large numbers.
South Korea’s national pension program has recently expanded, but the number of contributors is expected to shrink in the long term, given the prolonged period of low fertility. Contributors are projected to increase from 20.4 million in 2013 to a peak of 20.6 million in 2015 and then start to decrease, down to 11.0 million in 2078. Over the same period, the number of pensioners will increase from 1.8 million in 2013 to a peak of 14.0 million in 2065 and then decrease to 11.8 million in 2083. Thus the proportion of pensioners compared with contributors will rise dramatically, from 13.0 % in 2013 to peak at 112.9 % in 2068 and then decline slightly to 111.9 % in 2083.
The pension fund is forecast to peak at South Korean won (KRW) 2,561 trillion (US$2.35 trillion at 5 Feb 2015 exchange rate) in 2043 and then start to shrink, with annual earnings less than expenditures, beginning in 2044 (Fig. 6.5). At current rates, the fund will be exhausted in 2060. To keep the fund solvent, contributions will need to increase from the current rate of 9 % of reported income to 12–16 % (National Pension Finance Estimate Committee 2013).
Fig. 6.5
Projection of pension funds for South Korea, 2013–2060, billion KRW (National Pension Finance Estimate Committee 2013)
Similarly, the number of contributors to South Korea’s national medical insurance program will go down, and the number of beneficiaries will go up, so that the ratio of expenditure to revenue will increase from 1.00 in 2012 to 1.53 in 2030. Total expenditure already exceeds total revenue, and the deficit is predicted to rise from KRW 2.8 billion (US$2.58 million) in 2012 to KRW 47.7 billion (US$43.90 million) in 2030. It is estimated that the premium rate for medical insurance will need to increase from 5 % in 2012 to 12 % in 2030 (Park and Kim 2010).
As expenditures increase for pensions, medical insurance, long-term care of the elderly, and other areas of social welfare, the ratio of South Korea’s total social expenditure to GDP will increase (Fig. 6.6). In 2013, public social expenditure accounted for 9.8 % of GDP; in 2060 this proportion is expected to rise to 29.0 % (Social Security Committee 2014).
