African governments pledged in 2001 to spend 15% of their budgets on health — their own 2023 accounts show 1 of 54 keeps it

At a special summit in Abuja on 24–27 April 2001, the heads of state of what was then the Organisation of African Unity adopted the Abuja Declaration, whose paragraph 26 set "a target of allocating at least 15% of our annual budget to the improvement of the health sector". The pledge has been reaffirmed at the African Union level since, and the World Health Organization tracks it with a single indicator: domestic general government health expenditure as a share of general government expenditure — of every dollar a government itself spends, how many cents go to health. That indicator is published by the WHO Global Health Expenditure Database and mirrored by the World Bank as series SH.XPD.GHED.GE.ZS, database vintage last updated 2026-07-13. For the 54 African Union member states the database carries a 2023 figure — the 55th, Western Sahara, is not in it — exactly one clears the 15% line: South Africa, at 16.89% ^mmodeled. (The two nearest, Namibia at 14.99% and Botswana at 14.61%, fall short; Namibia misses the target by one one-hundredth of a point ^mmodeled.) Below those three the record drops away. Eight members sit in the 10%-to-15% band — after Namibia and Botswana: Cabo Verde 13.79%, Tunisia 12.10%, Lesotho 11.83%, Eswatini 10.64%, Morocco 10.61%, Ghana 10.05% ^mmodeled. Forty-five of the 54 spend under 10% of government outlays on health, and 17 spend under 5%. The median member state devotes 6.05% and the unweighted mean is 7.07% — both under half the pledged 15%. The continent's largest economies and populations sit well inside that lower band: Nigeria at 4.31%, Egypt 6.93%, Ethiopia 5.68%, the Democratic Republic of the Congo 4.00%, Tanzania 5.14%, Kenya 8.65% ^mmodeled. The lowest reported shares are South Sudan 2.14%, Eritrea 2.37%, Benin 2.55%, Togo 2.56% and Malawi 3.29% ^mmodeled.

The following is analysis, not fact. The Abuja pledge is unusual among international commitments in that the yardstick and the counter-record share an author. A defense-spending target is measured against numbers a state chooses to publish; the 15% health target is measured against the state's own executed budget, compiled into a comparable framework by the WHO from national accounts. The gap this piece reports is therefore not a gap between a government and an outside auditor — it is a gap between a government's signed target and its own books ^mmodeled. Read across all 54 members at once, the pledge behaves less like a floor and more like a ceiling almost no one reaches. That a target reaffirmed for over two decades is met by a single signatory, with the next two failing by fractions of a point, is the shape of an aspiration that never became a constraint ^mmodeled. The near-miss cases are geographically clustered: the four members closest to the line — South Africa, Namibia, Botswana, plus Cabo Verde — are middle-income states, three of them in southern Africa, where domestic revenue is large enough that a health ministry competes for it on even terms ^mmodeled. Where that is not true, the share collapses: the ten lowest reported figures are all under 4.4% ^mmodeled. The single most consequential number is Nigeria's 4.31% ^mmodeled. It is the continent's most populous state and one of its two largest economies, and it hosted the 2001 summit that produced the target; its own accounts now place it in the bottom third of the very pledge it convened ^mmodeled. The pattern the whole panel reveals is that health's share of the budget tracks not the promise but the size and reliability of the domestic tax base — which is precisely the variable the pledge was written to override, and did not ^mmodeled.

The strongest counter is that this indicator understates health priority in exactly the countries that score lowest, because it counts only domestic government money. The WHO series measures domestic general government health expenditure; it excludes external and donor financing, which in several low-income African states funds a large share of the health system and often flows on-budget. A finance ministry in a donor-dependent state can argue, fairly, that total resources reaching health are far above 4% or 5% once external funds are added, and that the Abuja denominator penalizes it for the structure of its financing rather than its commitment. Two further caveats are real. The pledge spoke of the annual "budget" — an allocation — while the indicator measures executed expenditure, and the two diverge when funds are appropriated but not spent. And "general government expenditure" is broader than a central-government budget line, spanning all levels of government and social-security schemes, so the denominator is not identical to the one a 2001 drafter may have pictured. None of this inverts the finding: even the most generous reading of donor top-ups does not lift 45 states from under 10% to over 15%, and the metric WHO itself uses to monitor Abuja is this one. But whether a 4% figure reflects a broken promise or a broken tax base is the contested part, and the record supports the second reading at least as strongly as the first.

From a health ministry in a low-income, aid-dependent member state, the 15% line looks less like a broken vow than like a category error. Officials there point out that donor programs — for HIV, tuberculosis, malaria, immunization — pay for clinics, drugs and salaries that never touch the domestic budget the WHO indicator counts, so the country can run a larger health system than its own 4% or 5% implies. The same figure that reads, from the outside, as neglect of a signed pledge reads, from inside the ministry, as evidence that the pledge measured the wrong pocket. Both can be true: the domestic commitment is small, and the health spending it undercounts is not the whole of what reaches patients.

A health correspondent covers one country's budget at a time: whether the ministry's line went up or down this year, whether the promised clinic opened. Each of those stories is true and none of them can see the shape that only appears when all 54 African Union members are laid against the single number they all agreed to in 2001. Do that, and the Abuja Declaration's 15% target turns out to be met by exactly one signatory, missed by one one-hundredth of a point by the next, and undershot by more than half at the median — with the state that hosted the summit now sitting in the bottom third of its own pledge. The finding is not that any one government spends too little on health; it is that a commitment reaffirmed for twenty-two years functions, when read across every signatory's own books at once, as an aspiration almost none of them treat as binding — and that the share tracks the size of the tax base, not the promise.