Open the mobile subscriptions map and the first thing you notice is that most of the world is above 100 per 100 people. Fiji reports 574.2. That is not a data error, and what causes it is more interesting than the headline growth.

The growth first

Across the 147 countries with figures for both years, mobile subscriptions rose from 18.0 per 100 people in 2000 to 123.5 in 2024. Not one country went backwards. It is the only indicator in this atlas where every single country improved, which is itself remarkable — development indicators almost always have a conflict or a collapse somewhere in the tail.

The starting points are easy to forget:

• India — 0.3 → 79.4 per 100
• Ethiopia — 0.0 → 65.1
• Russia — 2.2 → 186.1
• United States — 38.9 → 113.2

Ethiopia's 2000 figure rounds to zero. India had roughly three mobile phones per thousand people at the start of this period. Both now have networks reaching most of their population. No other indicator in this atlas moves that far in twenty-four years.

So why 574?

The indicator counts active SIM subscriptions, not people and not handsets. Four things inflate it, and they stack:

Multiple SIMs per person. Where networks charge heavily for off-network calls, carrying two or three SIMs is rational, and people do. This is the main driver across South Asia, Africa and the Pacific.

Separate work and personal lines. Common everywhere, and counted twice.

Machine subscriptions. Payment terminals, vehicle trackers, utility meters and tablets each hold a SIM. None is a person.

Prepaid accounts that linger. A prepaid SIM stays "active" for months after its last use, depending on the operator's definition. Countries with mostly prepaid markets count a tail of SIMs nobody is holding.

Fiji's 574 is dominated by the last two: a small population as the denominator, prepaid-heavy operators, and a tourist market cycling through SIMs. Hong Kong at 364.8 is the same story with more commuters and more machines.

What it means for reading the map

This indicator is excellent for one question and useless for another.

It is excellent for "did mobile infrastructure arrive here?" The jump from 0.3 to 79.4 in India is unambiguous and tells you something true.

It is useless for "what share of people have a phone?" For that you want mobile ownership from household surveys, which is a different and much lower number — typically 60 to 80% where subscriptions read above 100.

A choropleth of this data therefore shades Hong Kong and Fiji darker than Germany, which has roughly universal real ownership. The map is not wrong. It is answering the question it was built to answer, and that question is about SIMs.

This is the same trap as the renewable energy share, where the top of the table is the poorest countries. Whenever a world map ranks small states at the extremes, check the denominator before you believe the ranking.

Explore the full map of mobile subscriptions by country, or see how to read country statistics without being fooled. Source: World Bank Open Data, 2000–2024.