Thirty-two African countries now spend more servicing debt than on healthcare. Twenty-five spend more on debt than on education. Around 57 percent of Africans live in a country that pays creditors more than it spends on health and education combined. The World Bank put it plainly in April: in four out of five African countries, interest payments exceed public spending on health or on education.

Those figures are widely quoted. What follows from them is examined far less often.

When an African government signs an IMF programme it accepts a set of numerical targets. Some of those targets are binding. Some are decorative. The line between the two categories tells you, with unusual precision, whose interests the architecture was built to protect.

Hard targets and soft ones

Quantitative performance criteria are the hard category. The primary balance. Net international reserves. Ceilings on central bank financing of the deficit. Ceilings on new non-concessional borrowing. Miss one and the review stalls, the disbursement is delayed, and the Board must grant a waiver before the programme moves on.

Indicative targets are the soft category. They are monitored. They are discussed in the staff report. Missing one costs a government a sentence of encouragement about the need to improve budget execution.

Social spending floors sit almost universally in the soft category. Human Rights Watch established this after reviewing every IMF loan programme for low and middle income countries across the first two years of the pandemic. The floors are indicative. Money keeps flowing when they are missed.

Oxfam reviewed the same generation of programmes and found roughly a third of the social spending floors were not met. Only half the countries examined reached their floor at all.

This is a design feature rather than an accident of implementation. Promises made to creditors carry consequences. Promises made to citizens do not.

The mechanism, country by country

Zimbabwe illustrates it cleanly. In July the IMF completed the first review of its Staff-Monitored Programme. Every quantitative target was met: primary balance, reserves, central bank credit to the public sector, new non-concessional borrowing, monetary base growth. The structural benchmark on the taxpayer register was met.

One target was missed. The indicative target on protected social and priority spending.

The detail that sharpens it is that the money existed. The Fund’s own statement records revenue collection as robust, fiscal performance as stronger than expected, and budget execution as conservative. The surplus is being saved as a buffer against possible food insecurity in 2027. The cash came in, the spending was held back, the social floor was missed, and the difference was banked.

Zambia shows a different face of the same problem. Its Extended Credit Facility reviews report missed indicative targets on non-mining tax revenue, domestic arrears clearance and reserve accumulation, alongside a missed reserves criterion. Performance against the social spending floor is barely visible in the published press releases. Creditor-facing targets get reported line by line. Citizen-facing ones often get a clause.

Ghana has met its performance criteria and indicative targets through successive reviews, which raises the question Eurodad first put in 2018. In many programmes the floors are set too low to fund basic healthcare even when governments hit them.

The pattern holds across all three. Where the floor is missed, nothing happens. Where the floor is met, it may still be set below the level that would keep a clinic open.

Where the adjustment actually lands

Care that the state declines to fund does not stop being needed. It gets done anyway, unpaid, inside households, overwhelmingly by women.

Africa has started to measure this properly. Kenyan estimates put unpaid domestic and care work at roughly 23 percent of GDP. Mali’s 2023 national time-use survey, run with UN Women, put it at 17.6 percent, with women averaging 24.7 hours a week of unpaid care against 6.6 hours for men. Across African countries with survey data, women spend between 210 and 222 minutes a day on unpaid care. Men spend between 35 and 108. The ILO records that 85 percent of employed African women also carry care responsibilities, the highest share of any region in the world.

Now set that against the fiscal outlook. Oxfam projects that 43 of the African Union’s 55 member states face public expenditure cuts of around US$183 billion over five years.

That adjustment has to be absorbed somewhere. A clinic that closes at three instead of six. A school feeding scheme that lapses for a term. A cash transfer that arrives two months late. Each one converts public expenditure into private hours. The hours are real. They are simply uncounted, and because they are uncounted they cost the fiscus nothing.

That is the quiet efficiency of the present architecture. It has located a source of fiscal adjustment that never appears in a debt sustainability analysis, never enters a creditor negotiation, and never registers as a cost in any programme document.

Naming the design

African ministries of finance make real choices and some of them deserve harder scrutiny than they get. A treasury that banks a revenue surplus while missing its social floor has made a decision, and it should answer for that decision at home.

Ministries of finance did not invent the indicative target. Creditors and the institutions that serve them decided which promises would carry consequences and which would not. Reading African debt as a story of domestic mismanagement lets that decision go permanently unexamined.

What should change

Four asks, all achievable inside the current programme cycle.

Make social spending floors quantitative performance criteria, carrying the same weight as the primary balance. If a floor is genuinely a priority, give it teeth.

Set the floors against the cost of delivering services rather than against last year’s underspend, and publish the methodology.

Require quarterly publication of execution against the floor, disaggregated by programme. Most African treasuries already publish quarterly public debt bulletins. The template exists and the capacity exists.

Fund national time-use surveys and bring them into the national accounts. Mali and Kenya have shown that African statistics offices can do this work. Every debt sustainability analysis and every arrears clearance strategy should state, in hours, what the adjustment will cost the households expected to carry it.

Zimbabwe’s Debt Consultative Group, co-chaired by the United Kingdom and France, is expected to hold its first meeting this month. Comparable conversations are under way in Accra, Lusaka, Addis Ababa and Dakar. The same question belongs in every one of those rooms. If arrears clearance and restructuring are meant to serve the people of these countries, why is the only commitment made to those people the one that carries no consequence when it is broken?

---

Sources: AFRODAD Media Initiative, 2026; World Bank, Africa’s Pulse, April 2026; UNESCO, July 2026; Human Rights Watch, Bandage on a Bullet Wound, 2023; Oxfam analysis of IMF loan programmes and projected expenditure cuts across African Union member states; Eurodad, 2018; IMF Press Releases 26/242 and 26/261 on Zimbabwe; IMF Country Report 25/225 and Executive Board statements on Zambia; IMF review statements on Ghana; UN Women and Government of Mali national time-use survey, 2023; ILO, Care Work and Care Jobs for the Future of Decent Work.