Key Statistics
What this page is. One standardized set of headline numbers for the continent and for its ten largest economies, so that a chapter, a figure and a country card cannot quietly disagree. Every cell here is a field on one of the fifty-four country cards or a key in the verified population anchors, and the page is generated from them rather than typed, so it moves when the data moves and it cannot go stale on its own. Where a chapter needs a different concept, a different year or a different geography, it says so where it uses it; nothing on this page is a substitute for reading the chapter that argues about the number.
The ten economies below are the ten largest by GDP in current dollars, and between them they are 69 percent of the continent’s output. They are not the ten largest by population, by working-age population or by any of the growth terms, and three of the book’s eight growth poles sit outside them. The continental row at the head of each table is a total where the column adds up and the median card where it does not, and it says which it is in the cell, because the median of fifty-four countries is not the continent and the two are regularly confused.
1. Scale
| Country | Population 2024 (m) | Population 2050 (m) | Median age (years) | GDP (\(bn) | GDP per head (\)) | Urban (%) | Confidence | |
|---|---|---|---|---|---|---|---|
| Africa, 54 cards | 1,515.1 | 2,466.6 | median 18.9 | 3,137.9 | median 1,678 | median 45.7 | 11 high, 20 medium, 20 low, 3 not yet computable |
| South Africa | 64.0 | 79.2 | 28.5 | 427.1 | 6,769 | 63.8 | high |
| Egypt | 116.5 | 161.6 | 24.2 | 364.6 | 3,379 | 42.9 | high |
| Nigeria | 232.7 | 359.2 | 17.9 | 290.5 | 1,223 | 63.8 | low |
| Algeria | 46.8 | 59.6 | 28.4 | 285.7 | 6,047 | 75.8 | medium |
| Morocco | 38.1 | 43.4 | 29.5 | 182.6 | 4,842 | 63.1 | high |
| Angola | 37.9 | 74.3 | 16.5 | 141.7 | 3,614 | 71.5 | medium |
| Kenya | 56.4 | 83.6 | 19.8 | 136.5 | 2,558 | 32.2 | high |
| Ghana | 34.4 | 50.6 | 21.1 | 114.7 | 3,271 | 58.9 | medium |
| Ethiopia | 132.1 | 225.0 | 18.9 | 109.1 | 987 | 24.1 | low |
| Côte d’Ivoire | 31.9 | 55.7 | 18.2 | 98.9 | 3,006 | 54.5 | medium |
Population is in millions and is the 2024 and 2050 medium variant; the continental figures are the verified anchors, which count every African territory and are therefore a little above the sum of the fifty-four cards. GDP is in current dollars at the level the Fund publishes for 2025. The urban share is each country’s own definition of urban, which is not the built-up-area rule the cards use for a largest city, so the two are never compared. Sources, all of them through the country cards: UN World Population Prospects, the 2024 revision, medium variant, ages 15 to 64 where an age band is involved; the IMF World Economic Outlook, April 2026; the World Bank World Development Indicators, the live series, at the pull date each fetch records beside its own output; the cards themselves, for a number computed from the fields beside it, which carries no release of its own. The vintage of every release, and the day each was last read into the cards, are on the About the data page.
2. The identity terms
| Country | Term one, working-age population | Term two, employment ratio | Term three, output per worker | Implied | Observed | Gap | Confidence |
|---|---|---|---|---|---|---|---|
| Africa, 54 cards | median 2.89 | median -0.15 | median 0.74 | median 2.98 | median 2.91 | median -0.01 | 11 high, 20 medium, 20 low, 3 not yet computable |
| South Africa | 1.61 | -0.45 | -0.48 | 0.68 | 0.67 | 0.01 | high |
| Egypt | 1.96 | 0.07 | 2.26 | 4.29 | 4.33 | -0.05 | high |
| Nigeria | 2.90 | 0.29 | -1.94 | 1.25 | 1.20 | 0.05 | low |
| Algeria | 1.28 | -0.25 | 1.04 | 2.07 | 2.08 | -0.01 | medium |
| Morocco | 1.10 | -0.45 | 1.88 | 2.53 | 2.54 | -0.01 | high |
| Angola | 3.53 | 0.09 | -3.12 | 0.49 | 0.64 | -0.14 | medium |
| Kenya | 2.94 | -1.17 | 2.83 | 4.61 | 4.62 | -0.01 | high |
| Ghana | 2.52 | -1.56 | 3.48 | 4.44 | 4.45 | -0.01 | medium |
| Ethiopia | 3.46 | -1.58 | 5.91 | 7.79 | 7.92 | -0.13 | low |
| Côte d’Ivoire | 2.88 | 0.11 | 2.90 | 5.89 | 6.16 | -0.27 | medium |
All six growth columns are annual rates in percent over 2014 to 2024, which is what makes the identity add. Term one is the working-age population aged 15 to 64; term two is modelled employment aged 15 and over divided by that population, a mismatch of age bands this book states rather than smooths; term three is the residual output-per-worker series. Implied is the three terms summed, Observed is measured GDP growth and Gap is the difference the decomposition has not accounted for. The continental row is the median card column by column, so its own columns do not add. Sources, all of them through the country cards: UN World Population Prospects, the 2024 revision, medium variant, ages 15 to 64 where an age band is involved; ILOSTAT modelled estimates, November 2025; the cards themselves, for a number computed from the fields beside it, which carries no release of its own; the IMF World Economic Outlook, April 2026. The vintage of every release, and the day each was last read into the cards, are on the About the data page.
3. The balance sheet
| Country ($bn) | External debt stock | Multilateral | Official bilateral | Private | Debt service, 2026 to 2030 | Chinese commitments, 2000 to 2024 | Confidence |
|---|---|---|---|---|---|---|---|
| Africa, 54 cards | 1,175.5 | 264.5 | 148.7 | 294.7 | 389.5 | 176.3 | 11 high, 20 medium, 20 low, 3 not yet computable |
| South Africa | 175.9 | 9.2 | 4.6 | 84.1 | 52.9 | 5.5 | high |
| Egypt | 156.0 | 30.8 | 29.6 | 46.6 | 65.8 | 7.7 | high |
| Nigeria | 108.8 | 23.8 | 6.1 | 20.3 | 29.3 | 8.9 | low |
| Algeria | 6.9 | 0.7 | 0.1 | 0.0 | 0.4 | 0.04 | medium |
| Morocco | 68.0 | 21.5 | 7.2 | 17.0 | 22.2 | 1.0 | high |
| Angola | 58.7 | 6.2 | 4.1 | 35.8 | 41.4 | 49.2 | medium |
| Kenya | 42.9 | 20.0 | 8.1 | 7.5 | 18.8 | 10.0 | high |
| Ghana | 37.4 | 8.8 | 3.4 | 13.4 | 14.6 | 5.8 | medium |
| Ethiopia | 36.5 | 15.7 | 12.5 | 5.7 | 13.1 | 14.0 | low |
| Côte d’Ivoire | 40.6 | 8.2 | 5.8 | 17.2 | 19.6 | 5.7 | medium |
All columns are in billions of current dollars. The three creditor columns partition publicly guaranteed long-term debt exactly and sum to less than the external debt stock beside them, which also carries short-term paper, Fund credit and private borrowing without a government guarantee; bonds sit inside the private column rather than beside it. Chinese loan commitments are commitments and not a debt stock, which is a different quantity from the creditor columns and is never added to them, for the reason Chapter 21 gives. A zero here is a zero, a line the statistics report as nil rather than one they do not carry, and a line they do not carry reads in words instead. Sources, all of them through the country cards: World Bank International Debt Statistics, the IDS 2025 release of December 2025; the cards themselves, for a number computed from the fields beside it, which carries no release of its own; Boston University’s Chinese Loans to Africa database, the 2000 to 2024 release. The vintage of every release, and the day each was last read into the cards, are on the About the data page.
4. Data quality
| Country | Last census | Last GDP rebase | Last labour-force survey | SPI score | IMF standard | Confidence |
|---|---|---|---|---|---|---|
| Africa, 54 cards | median 2021 | 12 of 54 recorded | median 2022 | median 61.7 | 44 e-GDDS, 8 SDDS, 2 none | 11 high, 20 medium, 20 low, 3 not yet computable |
| South Africa | 2022 | 2021 | 2025 | 81.3 | SDDS | high |
| Egypt | 2017 | not recorded | 2023 | 78.1 | SDDS | high |
| Nigeria | 2006 | 2025 | 2023 | 69.1 | e-GDDS | low |
| Algeria | 2022 | not recorded | 2024 | 55.2 | e-GDDS | medium |
| Morocco | 2024 | not recorded | 2024 | 73.1 | SDDS | high |
| Angola | 2024 | not recorded | 2024 | 58.2 | e-GDDS | medium |
| Kenya | 2019 | 2021 | 2024 | 73.4 | e-GDDS | high |
| Ghana | 2021 | 2018 | 2025 | 65.7 | e-GDDS | medium |
| Ethiopia | 2007 | not recorded | 2021 | 61.7 | e-GDDS | low |
| Côte d’Ivoire | 2021 | 2020 | 2019 | 70.2 | e-GDDS | medium |
Years are the year of the exercise, not of its publication. The confidence flag is high where a country has a census within fifteen years, a statistical-performance score of 70 or better and a labour-force survey within five years, medium where one of the three fails, and low where two or three fail or the census predates 2000; where one of the inputs is itself missing the flag is left uncomputed, because not knowing when a country last ran a survey is ignorance about a statistical system and not evidence against it. Sources, all of them through the country cards: the book’s own data-quality table, entered by hand, one citation per country per field, carried in each card’s _sources map; the cards themselves, for a number computed from the fields beside it, which carries no release of its own; World Bank Statistical Performance Indicators, the live series, at the pull date. The vintage of every release, and the day each was last read into the cards, are on the About the data page.
What a blank means
There are none. A number that is missing is written out in words, and the three phrasings mean different things: not yet fetched is a series no source has supplied for that country or that does not exist for it; not recorded is a hand-entered field that has not been established, such as a rebase for a country that has not run one; not yet computable is a number derived from one of those. None of the three is a zero, and none of them is an average of the neighbours. The pattern of the emptiness is one of this book’s subjects and is argued in Chapter 1 and on the About the data page.
How this page is kept
It is generated. python3 _scripts/build_key_statistics.py rewrites it from _data/cards/*.json and the verified population anchors; --check fails if what is on disk is not what the cards would now produce, and it runs on every push. Every cell carries a tag naming the field behind it, which _scripts/check_registry.py resolves and the reader never sees. A cell that reads as empty carries a tag asserting the emptiness, so the day a source fills that field this page fails rather than contradicting the card beside it.