Why freezing today's birth rates is not a forecast
"If today's birth rate never changes, the country will be gone by 2300." Charts like this travel fast, and the arithmetic behind them is usually right. But holding a rate fixed for centuries is a thought experiment, not a forecast. The simulator shows why in a striking way: freeze every country's fertility at once and the world does something no one expects.
What "freezing" means
A frozen-rate scenario keeps each country's fertility (children per woman) at its current value forever. It is a standard tool in demography, sometimes called a "constant fertility" variant, and it is genuinely useful: it measures how far today's behaviour is from replacement, and how much population change is already built into the age structure. The trouble starts when the result for one country is read as a prediction.
The simulator offers two versions. In closed borders, fertility and life expectancy stay at 2025 values and nobody migrates. In freeze today's rates, fertility is frozen but life expectancy still improves and people keep moving at today's rates.
One country at a time, it looks like a countdown
With closed borders, South Korea goes from 51.7 million in 2025 to 13.3 million in 2100 and 778,000 in 2200. Hong Kong goes from 7.51 million to 1.92 million in 2100 and 106,000 in 2200. These are the numbers behind most "disappearing" headlines. They follow directly from fertility rates around 0.7 to 0.8: each generation is about a third the size of the one before, and after a few generations very little is left.
Seen alone, this looks like a countdown. The problem is what the same rule implies everywhere else.
All countries at once, the world explodes
Freezing is not only about low fertility. It also freezes every country with high fertility. Nigeria, for example, starts at 4.6 children per woman. Held there, it grows from 232 million to 472 million by 2100 even with no migration.
| World population | 2050 | 2100 | 2200 | 2525 |
|---|---|---|---|---|
| UN-style: fertility drifts toward 1.8 | 9.11 billion | 9.49 billion | 8.33 billion | 2.34 billion |
| Today's rates, closed borders | 9.15 billion | 11.1 billion | 17.7 billion | 39.1 billion |
| Freeze today's fertility, migration continues | 9.3 billion | 12.1 billion | 21.3 billion | 85.8 billion |
With fertility frozen everywhere, the world passes 12.1 billion by 2100 and reaches about 85.8 billion by 2525. The UN's World Population Prospects 2024, which assumes fertility keeps changing, expects the world to peak at about 10.3 billion in the mid-2080s and then slowly decline. The simulator's UN-style scenario lands in a similar place, peaking near 9.49 billion around 2098.
No one seriously expects 85.8 billion people. The point is that the same assumption that makes Korea vanish also makes the planet impossibly crowded. If you reject one result, you have to be careful about accepting the other.
The twist: frozen migration refills empty countries
The freeze scenario with migration shows something even stranger. As high-fertility countries hit the simulator's capacity limit, people move to countries with room. South Korea and Hong Kong, which are shrinking, become destinations. In this scenario Hong Kong falls below half of today's size in 2100 but is back to 10.5 million by 2525, and South Korea reaches 72.9 million. The population that results is very different in origin from today's, mostly made up of descendants of migrants from wherever fertility was frozen highest.
This is not a prediction about any country or group either. It is what happens when you hold every rate fixed for five centuries and let compounding run. Small differences in a growth rate, held long enough, always end with whichever group has the highest rate dominating. The model is showing a property of exponential growth, not a likely future.
Rates have never stayed frozen
The strongest argument against reading frozen rates as forecasts is history. South Korea had about six children per woman around 1960 and fewer than one six decades later. Across much of the world, fertility has fallen faster than almost anyone predicted in the 1970s. Rates also rise: South Korea's went from 0.72 in 2023 to 0.75 in 2024, according to Statistics Korea. Over a century, the chance that a rate stays exactly where it is today is close to zero.
Life expectancy has changed just as much, and migration responds to economies, wars and policy. A scenario that holds all of these still for 300 years is describing a world that has never existed.
So what is a frozen-rate scenario good for?
- Measuring the gap. It shows how far a country's current behaviour is from a stable population, in a single, vivid number.
- Showing momentum. For the next 25 years, frozen and converging scenarios give almost the same answer. That part of the future is largely already decided by today's age structure.
- Testing what it takes. Freezing everything, then changing one input at a time, shows how much a fertility recovery or a migration policy would matter.
The honest way to use a frozen-rate chart is to say "if today's rates continued" rather than "will". The simulator lets you check how sensitive any headline is: switch between scenarios in the Rates tab and watch how quickly the 2100 number moves.
Compare the scenarios for South Korea
Sources
- United Nations, World Population Prospects 2024, Summary of Results (world population peak of about 10.3 billion in the mid-2080s): population.un.org/wpp.
- Statistics Korea, preliminary 2024 birth statistics (0.72 in 2023, 0.75 in 2024), reported by AFP: fortune.com.
- Simulator method, scenarios and the capacity rule: Methods and data sources.