contents go

KDI - Korea Development Institute

KDI - Korea Development Institute



KDI FOCUS Raising Old-Age Threshold: Feasibility and Benefits September 06, 2022


Raising Old-Age Threshold: Feasibility and Benefits

September 06, 2022
  • 프로필
    LEE, Tae Suk
|   Related information   |
Currently, the standard for the elderly in Korea is 65 years old. As life expectancy increases and fertility rates decrease, the burden of supporting the elderly population in Korea will be the highest among OECD countries after 2054. Should I raise the age of the elderly? 

 Author: Lee Tae-seok, head of the KDI Population Structure Response Research Team

#Elderlyage #Pension #100-year-oldage #Elderlyjob
|   Script   |
The Welfare of Senior Citizens Act enacted in 1981 in Korea defines seniors as individuals aged over 65.
With the trend of increasing life expectancy and decreasing birth rate, Korea's old-age dependency ratio is expected to be the highest in the OECD from 2054 on.

Major countries are extending the pensionable age in consideration of their fiscal condition and trends in life expectancy. And the effective retirement age is also increasing.

On the other hand, concerns about fiscal sustainability are growing, as the number of years of receiving welfare services for the elderly and receiving pension is rapidly increasing, due to poorly adjusted old-age criteria.

As the health of the elderly population improves and the possibility of participation in the labor market increases, Korea also needs to review the criteria for the elderly population.

How should the age of the elderly in Korea be adjusted in the future?

Previous studies have suggested that the age of the elderly needs to be adjusted based on the remaining life expectancy, such as the age at which life expectancy remains at 15 or 20 years.

If we calculate the change in the old-age criterion based on the remaining life expectancy of 15 years, we can see that it increases by about 2.3 years on average over the past 10 years and by about 1 year on average every next 10 years. The age limit in 2100 is projected to be 80 years old.

We may encounter diseases or disabilities at some point in our lives.
The health-adjusted life expectancy (HALE) was applied to take into account such cases.

The old-age threshold based on the HALE is 68 years old as of 2016, showing an average increase of 2.7 years over the past decade.
This is two times faster than the OECD average.

Although HALE for women is higher than for men, the gap is narrowing from six to four years. Another gap between regions is also gradually closing.

Life expectancy can also differ depending on income, but the difference between income quintile is expected to be relatively small compared to other factors.

Considering these differences in gender, region, and income, If the old-age threshold increases by 1 year every 10 years from 2025, it will reach 73 years old in 2100.
If the old-age criterion is adjusted in this way, Korea’s old-age dependency ratio will be 36%p lower than the current 65-year-old threshold.

We need to determine the margin and timing of age adjustment for the elderly in consideration of the speed of health improvement of the vulnerable group, and prepare a gradual upward adjustment plan considering the gap by gender, region, and income.

Above all, it is important to have a sufficient notice period in consideration of the period necessary for forming expectations of the private sector, changing behavior, and adjusting social systems, and it is necessary to prepare support measures to mitigate damage caused by the adjustment process.

※ The provided materials below have been translated into English using computer-assisted translation.

Under the current threshold of old age at 65, Korea’s burden of caring for the elderly population is projected to rank the highest among OECD member countries from 2054 onward.

From 2025, when the burden of old-age support begins to increase rapidly, if the threshold for old age is raised by one year every 10 years in line with the speed of improvement in health conditions, the old age threshold in 2100 will be 73 years old and the adjusted elderly population ratio is expected to 60%, 36%p lower than the 65-year-old population ratio, and the burden of old-age support will be greatly reduced. In a situation where it is inevitable to alleviate the burden of supporting the elderly through the upward adjustment of the age of the elderly, the size and timing of the upward adjustment must be carefully determined in consideration of the speed of health improvement of the vulnerable elderly. At the same time, the roadmap for the adjustment of old-age threshold should take into account the adjustment period necessary for the formation of public expectations, behavioral changes, and social system adjustment, and the reform should be sufficiently announced in advance to mitigate the side effects caused by this reform.

Lastly, the demographic structure change should have complementary measures ready as part of well-rounded policy efforts.

I. Issues

Korea's life expectancy exceeded the OECD average in 2001 and is nearing the world’s highest level in Japan (83.7 years in 2021). Conversely, Korea’s total fertility rate nosedived, remaining at the lowest level in the OECD since 2003 after falling below the OECD average in 1984. In recent years, along with Hong Kong and others, Korea has ranked among the world's lowest (0.8 children as of 2021). The trends of rising life expectancy and falling total fertility rate are prompting a massive shift in demographic profile, represented by overall population decline and an increasing share of the elderly population.

In Korea, life expectancy is on the rise, and the total fertility rate is falling, indicating a declining total population and an increasing share of older people.

Korea used to have the highest percentage of young people in the OECD, but now it is becoming the demographically oldest OECD country. Its old-age dependency ratio¹  is forecasted to grow fastest among major countries for the next 30 to 40 years, exhibiting the highest proportion of seniors worldwide. Despite the lowest old-age to working-age population ratio until the 1980s in the OECD group, Korea is expected to surpass the group average in 2027 and post the highest rate starting in 2054. Meanwhile, Korea's total dependency ratio, which considers not only the elderly (older than 64) but also the youth (younger than 15), has been among the lowest among OECD countries since the early 1980s due to its declining fertility rate and is likely to remain low by 2025. As of 2022, Korea still has a higher share of the working-age population compared to major countries. The demographic burden has not fully hit Korea just yet. However, the total dependency ratio is to outstrip the OECD average from 2034 and go over 100% in 2058, the highest level ever recorded among OECD member countries. Exceeding 100% of the total dependency ratio means there are not enough working-age people to afford the dependent young and old. Accordingly, revisiting the age definition of the elderly is unavoidable, contingent on improvements in health status and the possibility of their rejoining the workforce.

Korea’s old-age dependency ratio has been lower than that of major countries but will be similar to or higher after five years. 

While Korea's aging-related expenditures have not surpassed major countries, it is projected to ascent fast due to the rapid aging of its population. As to how to make age-related spending efficient, this study investigates the feasibility of raising the threshold age of the elderly, together with essential considerations and expected effects of the reform.²

II. Welfare Programs for Older Adults: Current Age Eligibility 

Age criteria for elderly welfare programs are determined based on a comprehensive list of project objectives, characteristics of policy targets such as health status by age, and budgets and fiscal resources. Although it differs from program to program, the most commonly used starting age to identify old age is 65. As of 2022, the minimum age required for major elderly welfare programs varies from 50 to 75. However, out of 49 major programs, 24 (49%) are intended for 65+ and 14 (29%) for 60+. 

차트 샘플

The bias in age eligibility criteria of 65 years or older may have resulted from the fact that laws and projects related to the welfare of the elderly adopted the minimum age set by the Welfare of Senior Citizens Act. The minimum age had not been amended for about 40 years and the same age criteria apply to the laws introduced in the 2000s, including the Long-Term Care Insurance Act and other laws concerning housing and mobility for disadvantaged persons. 

In accordance with the Welfare of Senior Citizens Act enacted in 1981, the age eligibility for senior citizen welfare services in Korea is 65 years of age or older, unless otherwise specified in the special law.

III. Changes in age thresholds of old age

Since the 19th century, the conventional threshold of old age has been 65. Despite the lack of clear theoretical evidence to support this, the long-established beginning of elderliness at age 65 prevails across many countries. Nevertheless, the global demographic megatrend emerging after Second World War started to increasingly uncover the inadequacy of the customary understanding of old age, namely the acceleration of population aging and the improvement in senior health. In addressing the population challenge, major countries are contemplating raising the pensionable age, given rising life expectancy and fiscal conditions. Another global trend is that the average age of leaving the labor market (effective retirement age) is increasing gradually. 

Given rising life expectancy and fiscal conditions, major countries are pushing back the pensionable age, and the effective retirement age is gradually increasing. 

The average effective retirement age across OECD countries displays a U-shaped pattern, similar to the trend of the average pensionable age, with a gradual increase from the 2000s. Countries that have raised the pension age since the 2000s have also taken action to improve their systems by increasing or even abolishing the retirement age to push up the effective retirement age. These policy efforts, combined with improvements in the elderly’s physical and economic ability, seem to have paid off. Since 2000, the OECD average of the effective retirement age has been adjusted in parallel with the increasing rate of the remaining life expectancy. It is found that the expected retirement period has been adjusted by 20 years or shorter for men and 24 years or shorter for women. In Korea, the effective retirement age has remained higher than the OECD average, while the expected retirement period is about 15 years for men and 20 years for women. 

차트 샘플

Analogous to advanced countries, Korea pushed back its pensionable age for national and government employee pensions due to relevant reforms in 1998 and 2005, respectively. As shown in Figure 3, Korea’s pension age is planned to gradually hike to 65 by 2033, but this plan lacks discussions on how to make the upward adjustment. As already mentioned, the threshold age for older persons has been 65 since the enactment of the Welfare of Senior Citizens Act and has not changed since then.

As the pace of adjusting the old-age threshold has not kept up with the increase in life expectancy, the period for pension take-up and elderly welfare benefits has rapidly extended. This trend is likely to persist.

The period of receipt for pension and elderly welfare benefits has been rapidly growing due to the lopsided changing pace of the lower limit of elderliness compared to the increasing life expectancy, which is likely to continue in the coming years. The life expectancy at 60 was 14.6 years back in 1960 at the time of the introduction of the Government Employees Pension; however, in 2022, the average number of years remaining at 61, the pension eligibility age in that year, is 21.4 years, up by 6.8 years. As for the national pension scheme, the life expectancy at the pensionable age of 60 was 18 years in 1988 when first adopted but later added 6 years to 24 years in 2022. The life expectancy at 65, the threshold age for the elderly stipulated by the Welfare of Senior Citizens Act, was 14.5 years at its enactment in 1988 and later went up to 21.4 years in 2022. 

차트 샘플

The public pension system faces the most severe consequences of the fiscal sustainability burden from the aging population. Major advanced countries have opted for automatic adjustment mechanisms that change the pensionable age tied to rising life expectancy. This mechanical system was able to address the issue in part, but as fiscal sustainability is not limited to public pensions, more systematic efforts are required to reassess the age criteria of elderliness across overall welfare programs for older people.

IV.  Adjusting Old-Age Threshold: Theoretical Review of Changing Threshold Age of Seniors

In adjusting the old-age threshold, the first order of business is to gather objective data supporting the inevitableness of gradually moving it up due to longer life expectancy. ³  Relevant discussions should include political considerations while mindful of different types of conflicts of interest. However, a theoretical discourse of non-political nature is underway ahead of such politically charged dialogue to clearly show the concerned policy agenda and viable alternatives. Most prior theoretical studies on linking old age, aging processes, and health status suggest defining elderliness based on specific years of remaining life expectancy (Ryder, 1975; Siegel and Davidson, 1984; Fuchs, 1984; Sanderson and Scherbov, 2016). Earlier literature supports using 10 years in life expectancy, while later studies recommend a more realistic approach of 15 years. Given various heterogeneities not captured in prospective old age threshold and life expectancy found in major countries, 20-year life expectancy should also be considered. 

This study examines Korea’s aged population based on 15-year life expectancy and makes international comparisons via the 2022 World Population Prospects. The result shows an increasing trend of 2.3 years every 10 years until 2022, at which point old age begins at 73, and afterward, 1 year in 10 years. The onset time of old age is projected to extend from 62 in 1975 to 69 in 2005, 76 in 2051, and 80 in 2100. 

Korea’s criteria for old age based on 15-year life expectancy has steadily risen to 73 in 2022 and will increase by about an average of 1 year every 10 years. 

The 2017 Global Burden of Disease provides data on the burden of diseases and disabilities not captured in life expectancy. This study looks at the trajectory of old age calculated using 15 years of health-adjusted life expectancy of 29 countries that became OECD members before 2000: 63 years on average back in 1990 but reached 67 in 2016, adding 1.5 years every decade. Korea, too, shows an upward movement from 61 in 1990 to 64 in 1999,  66 in 2007, and even 68 in 2016. Its average increase is 2.7 years in 10 years, about two times faster than the OECD average. 

Korea’s old age, calculated with the 15-year health-adjusted life expectancy, had a 2.7-year increase every decade to reach 69 in 2016, rising about two times faster than the OECD average. 

차트 샘플

By gender, as HALE typically shows extended years in women than men, ages considered to be old for females are generally higher than for males. The average gender gap among OECD member countries is shrinking from 5 years in 1990 to approximately 4 years in 2016. Korea also has a narrowing gender gap from about 6 years to around 4 years, but still, old ages for men are estimated to be lower than for women. Meanwhile, since HALE is not without estimation error, it may have 3~4 years of estimation error between the upper and lower bounds of 95% confidence intervals (CI). The relatively low bottom limit of men’s old ages has a wider distribution than the entire population, implying that inter-country heterogeneity may increase further. The threshold age of old Korean males at the 95% lower bound of 15 years of HALE in 2016 is 64, which is similar to the current level. 

Considering the gender gap and estimation errors, the threshold age of Korean male elderly as of 2016 at the 95% lower bound based on 15-year health-adjusted life expectancy is 64, similar to the current age threshold. 

Another life expectancy gap to consider is intraregional. Statistics Korea’s national, provincial, and municipal life table data show significant gender and regional disparities in life expectancy. However, calculating older ages with 15-year life expectancy as of 2005 and 2017 based on Korea’s gender and regional gaps in life expectancy reveals that the starting age of elderliness is moving up over time while the gaps between genders and regions are gradually narrowing. Still, Korea continues to exhibit a regional gap of about 2-3 years and a much larger gender gap of about 5-6 years. That is, age calculated from the entire population’s life expectancy may not accurately reflect the current gender and regional gaps in life expectancy. 

The gender and regional gaps in life expectancy are gradually closing in Korea.

Other than health status and life expectancy, earnings may also have gaps. Khang et al. (2019) present trends and projections of gaps in life expectancy based on health insurance data, showing that the gender difference in life expectancy narrowed from about 7 years in 2004 to 6 years in 2017 with a forecast of gradual reduction. They further show that the disparity by income quintile widened from around 6 years in 2004 to about 6.5 years in 2017, projecting about 7 years in 2030. Moreover, the unavailability of life tables by income quintile rendered calculating old ages by income impossible. However, as the gap in life expectancy and the old age gap based on 15-year life expectancy show similar patterns, if the life expectancy gap by earnings widens, the old age gap is also likely to show gradual expansion. 

The earning gap in life expectancy is somewhat widening, but the gap remains relatively stable.

차트 샘플

To sum up, life expectancy and survival rates have consistently increased worldwide with time, thanks to improving health status at all ages. If a specific number of additional years to live determines old age based on a theoretical model of life expectancy, the threshold will continue to rise in the long run. In theory, HALE can serve as the most reasonable criteria to define elderliness, but that is not easy to apply due to difficulties like uncertainty in projections and restricted access to medical data. In this regard, this study recommends more practical approaches, like additional discussions about the extent of the age adjustment and the stability of the age gaps in light of old age thresholds calculated with life expectancy. 

Given the improving trend of life expectancy, the old-age threshold is expected to increase continuously over time.


V. Raising Old-Age Threshold: How to Proceed and What to Expect


Theoretically, Korea can keep raising the threshold age for elderliness by one year every 10 years. Still, given the likelihood for the private sector to anticipate and adapt to the new criteria, there should be enough preliminary discussions to build a consensus for an adjustment plan and a regular review mechanism. Meanwhile, the age threshold based on life expectancy for the total population may yield over-adjustment for particular social strata or groups of specific chronic diseases. It is because life expectancy does not cover the burden of disease and disability and the gap in genders, regions, and earnings. Therefore, a long-term perspective should be employed when considering the feasibility of increasing the age threshold. At the same time, deciding the timing and extent of the upward adjustment must use discretion in light of the burden of disease and disability and the existing gaps among different genders, regions, and income levels and their long-term trends. 


If Korea keeps raising the threshold for old age by 1 year every decade from 2025, the age baseline in 2100 will be 73, with its old-age dependency ratio at 60%, which is 36%p lower than the ratio under the current threshold of 65.

차트 샘플

The age difference among Korea's elderly is gradually closing while the age difference between the expectation of life remaining 15 and 20 years is 6 to 8 years. In view of the expected years in retirement at present, along with the trends in life expectancy gaps in genders, regions, and earnings, it is worthwhile to look at 20-year life expectancy. Suppose Korea raises the threshold age by 1 year every 10 years after 2025 when the care burden for the aged begins to emerge in full swing, given the pace of the private sector's adaptation (Figure 6). The defining age will hike to 73 in 2100, coupled with the old-age dependency ratio at 60% (Figure 7), which is 36%p lower than the ratio under the current threshold of 65.

차트 샘플

In general, the old-age dependency rate increases with the increasing share of social security spending. Still, the pace of welfare spending’s growing share may vary among countries, depending on their way of responding to the rising dependency ratios. The percentage of elderly welfare spending needs to be managed by increasing the old-age threshold in the future. 

The old-age dependency ratio generally tends to rise in tandem with the share of social security spending. However, despite the heavier and rising burden of elderly care than other countries, Japan and Italy have been decelerating the pace of increase in the share of their fiscal spending on social security. Lively discussions on an upward adjustment of baseline age for elderliness have taken place in those countries, which have a particular implication for Korea as it endeavors to systematically manage the strains from elderly welfare programs. A step-by-step upward adjustment strategy should be in place to alleviate the related financial burden. Such a scheme should take on a long view based on objective evidence while considering the burden of illnesses and disabilities and the gaps in genders, regions, and earnings. In addition, support measures should also be prepared to mitigate damages to the groups vulnerable to the age adjustment and provide prior notices to the private sector allowing ample time to enhance the prospect of their acclimatization to the change. At the same time, since a higher age baseline for elderliness means a higher ceiling for the working-age population, policy efforts to promote labor supply and demand for older workers should also be present and consider this particular working group's characteristics.


 [References]Fuchs, V. R., “Though Much is Taken: Reflections on Aging, Health, and Medical Care,” The Milbank Memorial Fund Quarterly: Health and Society, 62(2), Special Issue, Spring, 1984.
Khang, Y. H., J. Bahk, D. Lim, H. Y. Kang, H. K. Lim, Y. Y. Kim, and J. H. Park, “Trends in Inequality in Life Expectancy at Birth between 2004 and 2017 and Projections for 2030 in Korea: Multiyear Cross-sectional Differences by Income from National Health Insurance Data,” BMJ open, 9(7), 2019.
Lee, Taesuk et al. Structural Reforms in Response to Demographic Changes in Korea, Research Monograph 2020-08, Korea Development, 2020 (in Korean). 
Ministry for Health, Welfare and Family Affairs, The Guide of Aged Health and Welfare Project, 2022 (in Korean). 
OECD, Pensions at a Glance 2021: OECD and G20 Indicators, 2021.
Ryder, N. B., “Notes on Stationary Populations,” Population Index, 41, 1975, pp.3~28.
Sanderson, W. and S. Scherbov, “A Unifying Framework for the Study of Population Aging,” Vienna Yearbook of Population Research, 14, 2016, pp.7~40.
Siegel, J. S. and Davidson, M., “Demographic and Socioeconomic Aspects of Aging in the United States,” Washington, DC: US Government Printing Office, 1984. 
UN, 2022 Revision of World Population Prospects, 2022.

Korea Law Information Center (, last access: Apr. 29, 2022).
OECD Statistics (, last access: Apr. 29, 2022).

1  The old-age dependency ratio means the share of the population aged 65 and over to the working-age population aged 15 to 64, and the total dependency ratio refers to the rate of the population of those aged 14 and under plus senior citizens to the working-age population. 
2  This study introduces some key findings from Lee et al. (2020) about the necessity of raising the old-age threshold and some policy elements to consider. For details on relevant discussion and policy recommendations for other key areas, refer to Lee et al. (2020).
3  Lee et al. (2020) recommend a series of policy strategies based on the operational experiences of the UK Pensions Commission regarding how to establish necessary grounds, share policy information, help the participants better adapt, and adopt complementary measures.
4  In Part 1, Lee et al. (2020) recommend several policy measures for a more efficient use of older workers: easing the constraints on senior citizens’ production activity, securing the incentives for the elderly to participate in the labor market, maintaining senior citizens’ labor productivity, and creating labor demand for older workers.

  • Ⅰ. Issues

  • Ⅱ. Welfare Programs for Older Adults: Current Age Eligibility
  • Ⅲ. Changes in age thresholds of old age
  • Ⅳ. Adjusting Old-Age Threshold: Theoretical Review of Changing Threshold Age of Seniors
  • V. Raising Old-Age Threshold: How to Proceed and What to Expect
related materials ( 10 )
  • Key related materials
Join our Newsletter

World's Leading Think Tank, Korea Development Institute

Security code

We reject unauthorized collection of email addresses posted on our website by using email address collecting programs or other technical devices. To access the email address, please type in the characters exactly as they appear in the box below.

KDI Staff Information

Please enter the security code to prevent unauthorized information collection.

KDI Staff Information

Please check the contact information.

KDI Staff Information

Please check the contact information.