Egypt overtakes Japan in 2028 on our path. Japan has now shrunk for fourteen years running.

Satoria's forecast

Satoria forecast for Egypt: how many people live there 133.0 million people in 2034, up from 118.4 million people in 2025 - clearly higher. 80% range 127.8 million people to 139.0 million people.

Two countries pass Japan. One does not.

Our central paths put Egypt at 123.5 million people in 2028 against Japan's 121.8 million, so Egypt goes past that year. The Philippines follows in 2030, at 121.8 million against 120.9 million. Vietnam stays about ten million behind through 2034.

A count on one day is not a midyear estimate.

The series behind our forecast counts all residents of a country whatever their legal status, in the middle of each year. Japan's own count is for 1 October. Both are right, measured at different moments. In the midyear estimates for 2025 Japan had 123.4 million people and Egypt 118.4 million.

Fourteen years of getting smaller, every single year.

Japan's own statistics bureau counted 123,802 thousand people on 1 October 2024. That was 550 thousand fewer than a year earlier, a fall of 0.44 percent, and the fourteenth yearly fall in a row. The United Nations expects Japan to lose about twenty-one million more people by 2054.

Where does Egypt land in 2034?

Egypt had 89.2 million people in 2010 and 118.4 million in 2025. Japan went the other way over the same years. Move the slider to where you think our model puts Egypt ten years on, then see the answer and the range.

Our ranges on this measure are too narrow, and we can prove it.

In testing, the 80% range we publish for population covered only 0.69 of the held-out points instead of 0.8. So treat the band as the least uncertainty there is. The Egyptian and Japanese ranges already overlap in 2028, which puts 2027 and 2029 in play.

2028 is the year to check, and 2030 after it.

If the midyear estimates put Egypt above Japan in 2028, our path held. If Japan's fall slows, or Egypt's growth slows faster than we expect, the crossing slides later and we will put that in our public record. Japan's next yearly count is the first place to look.

What Japan counted

Japan's statistics bureau counted 123,802 thousand people on 1 October 2024, 550 thousand fewer than a year before. That is a fall of 0.44 percent, and the fourteenth yearly fall in a row.

The United Nations expects this to carry on: it puts Japan second only to China for absolute population loss between 2024 and 2054, at about twenty-one million people. That is the United Nations' own projection, not ours.

One thing to hold onto before the numbers start moving: the series our forecast runs on counts all residents whatever their legal status, in the middle of each year, so it will never exactly match a count taken on 1 October.

A very large gap has almost closed

In 2010 Egypt had 89.2 million people and Japan 128.1 million. In the midyear estimates for 2025 Egypt had 118.4 million and Japan 123.4 million. What was a difference of tens of millions is now a few million.

Egypt did not do anything sudden. It gained people steadily for fifteen years while Japan lost them steadily. Two ordinary trends, running in opposite directions for long enough.

Our central path puts Egypt at 133.0 million by 2034, with a published range from 127.8 million to 139.0 million. A straight line drawn through the measured years would land well above that, so our model is already reading Egypt's growth as slowing.

What our model says next, and where it is weak

On the central paths Egypt reaches 123.5 million people in 2028 against Japan's 121.8 million, and the Philippines reaches 121.8 million in 2030 against Japan's 120.9 million. Japan is at 119.4 million by 2034. Vietnam grows to 109.0 million and stays behind.

Across 51 country series, on data held out from fitting and model choice, the model's errors were 67% smaller than assuming no change. Population is one of the easier things to forecast a decade out, because the people who will be alive in 2034 are mostly alive now.

The weak part is the uncertainty, not the central line. In testing, the 80% range we publish for this measure covered only 0.6902 of held-out points instead of 0.8. The band we draw is therefore too narrow, and since the Egyptian and Japanese ranges already overlap in 2028, a crossing in 2027 or 2029 is entirely inside what we should be allowing.