South Korea’s population is set to shrink: what would it take to stop the decline?
How much would fertility rates, life expectancy, or migration rates need to change to stop the population from shrinking?
South Korea’s population is expected to decline substantially in the coming decades. According to the United Nations’s projections, shown in the chart below, the country’s population will more than halve.
In 2026, there are around 52 million people in South Korea. By 2100, the UN projects there will be just 22 million.
South Korea’s population would not only be less than half its peak size, but also much older. Below, you can see what the population pyramid would look like in 2100. The top — for people in their 70s, 80s, and 90s — would be very “heavy” compared to the bottom half.
There would be almost as many people of retirement age as those in the working-age population.
These projections rely on assumptions about how the drivers of population change — births, deaths, and migration — will change in the coming decades.
Below, we’ve plotted the UN’s assumptions for the three key metrics in this population model: fertility rate, life expectancy, and net migration.
The UN assumes that the fertility rate — the average number of children per woman — will rebound from its current low point of 0.7 births per woman to 1 by the middle of the century and to 1.3 by 2100.
For life expectancy, the UN projects an increase to 91 years by the end of the century.
And for net migration, the UN projects a decline from 1.7% annually today to almost zero by 2050 and thereafter.
No one knows with certainty how many children people will have in 2080, or what the migration rate will be in 2060. So it’s worth asking what the population will look like if things turn out differently from what the UN assumes.
My colleagues Daniel Bachler and Sophia Mersmann have built a population simulation tool that helps to answer this — not just in South Korea, but in any country.
This tool is deliberately simple: small enough to run in a browser and nevertheless capable of modeling whole populations based on the key demographic inputs. It is far less detailed than the models expert demographers use, and our aim is not to replace them. We find this tool useful for developing an understanding of the difference that changes in demographic variables can make to the size and structure of future populations.
You can explore the model for any country in the world at the very end of this page.
I’ve used the model to explore a simple question: what would need to happen for South Korea’s population to remain roughly constant?
How much would fertility rates, life expectancy, or migration rates need to change?
I will investigate the levers one by one, so in each scenario, I’ll keep the UN’s assumptions for the other two metrics as they are.
By how much would the fertility rate need to rise to keep the population from shrinking?
One way to boost the population is for people to have more children.
South Korea has seen a particularly dramatic decline in fertility rates over the last 60 years. In 1960, the average woman had six children. Since then, it has declined to just 0.75 children.
By how much would the fertility rate need to increase for South Korea to keep its population constant over this century?
Our population model shows one possible path: if the rate were to increase to 2.1 children per woman by 2050 and remain at that level, the country’s population would stay close to current levels.
In other words, South Korea would need to see a substantial “baby boom” in the next few decades. You can adjust the path with the controls in the visualization. As you’ll see, a smaller increase in fertility rates would slow the decline, but it wouldn’t stop it.
By how much would life expectancy need to increase to keep the population from shrinking?
Another driver of population growth is increasing lifespans.
By how much would life expectancy need to increase in a future in which the UN’s expectation for the future of fertility rates and migration rates becomes true? The visualization below shows the answer: by about 50 years.
Our population model shows that if average life expectancy increases to 130 years by 2050, South Korea could keep its population constant for several more decades.1 But ultimately, low fertility rates will lead to a decline, even if life expectancy increases substantially.
Reaching 130 years of life expectancy within the next few decades would require a dramatic acceleration in the rate of improvement.
In recent decades, the countries with the highest life expectancy have taken four years to increase average life expectancy by one year. At those rates, it would take 188 years to increase life expectancy from 83 to 130 years. South Korea would need to do it in 25.
Relying on increased life expectancy to stabilize population size also raises questions about whether it’s the size of the population that matters, or its age structure.
Under that scenario, by the end of the century, South Korea’s median age would be 82 years. If the retirement age remained at 65, there would be more than two pensioners for every working person.
Unless medical innovations made it feasible for people to work well into their 80s, 90s, or 100s, the working-age population would be greatly outnumbered by retirees.
By how much would immigration need to increase to keep the population from shrinking?
South Korea’s final option is to look elsewhere. Rather than relying on population increase from births or longer lives, it could make it easier for people to move to the country.
Currently, around 1.3 people move in per 1,000 residents each year. That’s net migration, which is the balance of people entering and leaving.
How much would net migration rates need to increase if South Korea wanted to maintain its current population size?
The projection shows that it would require a net migration rate of about 9‰ — 9 immigrants per 1,000 people annually. This is about seven times higher than the current rate.
How realistic are these migration rates?
Most countries have far lower rates. Net migration in the United States and the United Kingdom has fluctuated between 3 and 6 per 1,000 people over the last 30 years. Many other high-income countries have similar rates.
Net migration rates above 8‰ are much less common, but not unheard of. As you can see in the chart below, Canada has maintained rates between 6 and 10 per 1,000 for most of the last 20 years. In Germany, rates have reached those levels over the last decade.
What would be rare is maintaining such high immigration rates for more than 70 years. And given that most other countries are also shrinking, it would be increasingly difficult to attract such high numbers of immigrants.
What did I learn about South Korea’s prospects of avoiding a shrinking population from this tool?
The gains in life expectancy or migration would be extremely large; the scale of these changes seems implausible. The most realistic of the three scenarios is an increase in fertility rates; they would need to increase to 2.3 children per woman within a few decades and stay there. Those rates are not unprecedented by any means — in fact, they’re well below the country’s historical average — but such a strong reversal would be unique following the huge decline in fertility rates across the world over the last half-century.
I think it’s very unlikely that South Korea can maintain its current population levels over the course of this century.
Explore your own projections
In this article, I’ve explored some very specific scenarios for South Korea.
But Sophia and Daniel’s tool — which you can see below — lets you see what would happen to population projections and the age pyramid under very different assumptions. You can adjust future fertility rates, life expectancy, and migration rates for South Korea, or any other country you are interested in.
Acknowledgments
Many thanks to Mallika Snyder for her valuable feedback and suggestions on this tool. We’d also like to thank Max Roser and Edouard Mathieu for comments and editorial feedback, and Marwa Boukarim for input on design and visualization.
Endnotes
It’s worth noting what this would mean in practice. Using these assumptions, almost no one in this population model would die in most years in South Korea. That seems implausible, but it actually points to the fact that the gains in health and life expectancy would need to be incredibly huge if this were to be the factor that stabilized South Korea’s population.
Cite this work
Our articles and data visualizations rely on work from many different people and organizations. When citing this article, please also cite the underlying data sources. This article can be cited as:
Hannah Ritchie, Sophia Mersmann, and Daniel Bachler (2026) - “South Korea’s population is set to shrink: what would it take to stop the decline?” Published online at OurWorldinData.org. Retrieved from: 'https://archive.ourworldindata.org/20260518-034451/south-koreas-population-is-set-to-shrink-what-would-it-take-to-stop-the-decline.html' [Online Resource] (archived on May 18, 2026).BibTeX citation
@article{owid-south-koreas-population-is-set-to-shrink-what-would-it-take-to-stop-the-decline,
author = {Hannah Ritchie and Sophia Mersmann and Daniel Bachler},
title = {South Korea’s population is set to shrink: what would it take to stop the decline?},
journal = {Our World in Data},
year = {2026},
note = {https://archive.ourworldindata.org/20260518-034451/south-koreas-population-is-set-to-shrink-what-would-it-take-to-stop-the-decline.html}
}Reuse this work freely
All visualizations, data, and articles produced by Our World in Data are completely open access under the Creative Commons BY license. You have the permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited.
The data produced by third parties and made available by Our World in Data is subject to the license terms from the original third-party authors. We will always indicate the original source of the data in our documentation, so you should always check the license of any such third-party data before use and redistribution.
All of our charts can be embedded in any site.
