The Recognition Gap

Status and substance are not the same thing. The Power Atlas measures both: what a country has, and what the world treats it as having.

195 states · 1816 to 2026 · Correlates of War, Maddison, site power score · built 2026-09-05

States
195
ranked in 2026
Most under-recognised
+12.8
China
Most over-recognised
-27.3
United States
Tracked since
1816
15 powers with a full series

Today

Points of world power share: positive is mass not yet priced in, negative is status the base does not carry.

How this is measured

Latent material mass minus recognised standing, both as a share of world power in the same year.

Latent power is material mass. Recognised power adds what a state spends, how far it reaches, and the standing it has been granted: a permanent seat, a nuclear arsenal, leadership of a bloc. A country can have one without the other, and the interesting cases are the ones where the two have come apart.

1
latent 25.1 · recognised 12.3 · power rank 2
+12.8
gap
2
latent 10.9 · recognised 3.4 · power rank 8
+7.5
gap
3
latent 2.1 · recognised 0.5 · power rank 19
+1.6
gap
4
latent 2.1 · recognised 1.0 · power rank 11
+1.2
gap
5
latent 1.3 · recognised 0.1 · power rank 38
+1.1
gap
6
latent 1.1 · recognised 0.1 · power rank 48
+1.0
gap
7
latent 1.6 · recognised 0.6 · power rank 42
+1.0
gap
8
latent 1.1 · recognised 0.1 · power rank 49
+1.0
gap
9
latent 1.6 · recognised 0.6 · power rank 17
+1.0
gap
10
latent 1.4 · recognised 0.5 · power rank 14
+0.9
gap
11
latent 1.2 · recognised 0.3 · power rank 33
+0.9
gap
12
latent 2.3 · recognised 1.5 · power rank 15
+0.8
gap

Since 1816

Every state that has held a top eight place, with its gap in every year it was ranked.

How this is measured

Series start at 1816, where the Correlates of War capability data begins. Earlier Power Atlas years are benchmark-interpolated and carry wide error bars.

An open question, not a findingunresolved

Whether the gap leads war onset is an open question on this site, not a finding. It needs an initiation coding the war dataset does not yet carry.

The obvious thing to ask of this series is whether a wide gap comes before a war. The site holds 623 wars back to 1500 and could join them to this board tomorrow. It will not, yet, for three reasons: the war data records two sides but not who started it, there are no controls for how violent a given era was, and part of the recognised score is a curated status layer assigned with hindsight, which could be circular with war itself. Any of the three is enough to make a result look real when it is not.

When those are resolved the answer goes on the Ledger, with a date, whichever way it comes out.

Sourcesbuilt 2026-09-05
  • Correlates of War NMC v6.0
  • Maddison Project Database 2023
  • site country score and curated status layer