In many households there is an unspoken order at the table. Guests are served first, then the elders, then the children. Nobody writes it down. Everybody knows it.
Across much of the developing world, national budgets now keep a similar order, and it is a harsher one. The lender eats first. What is left goes to the clinic, the classroom and the district health office.
Too often, the children eat last.
A table laid for the creditor
The numbers are not subtle. UN Trade and Development (UNCTAD) published A World of Debt 2026 in October. It finds that developing countries paid $995 billion in net interest on their public debt in 2025. The median developing country spent 8.1% of government revenue on net interest that year, nearly double the 2010 level.
Now read the next figure twice. According to the same report, 3.7 billion people live in countries that spend more on interest payments than on either education or health. That is 1.2 billion more people than in 2010. UNCTAD counts 51 developing countries that spend more on interest than on either education or health.
Africa sits near the front of that queue. UNCTAD reports that the share of Africa’s population living in countries spending more on interest than on health or education rose from 24% in 2010 to 65% in 2023. In Africa, governments now spend more per person on interest than on either health or education.
The clinic waits outside
My work keeps me close to health systems, and I know what those ratios look like on the ground. They look like a district hospital with a vacancy list nobody is allowed to fill. They look like stock-outs, unpaid allowances and a training plan postponed for another year.
I argued recently that Africa’s health worker shortage is a hiring problem, not only a training problem. Debt is a large part of why the hiring does not happen. A finance ministry that must pay the lender before it pays the nurse will, sooner or later, stop paying for new nurses.
The World Bank’s Africa Economic Update, published on 6 October 2026, is blunter still. In nearly four-fifths of the countries in sub-Saharan Africa, it finds, “interest payments exceed public spending on health and/or education”. Its launch statement adds that “Public debt has broadly stabilized at around 57% of GDP, but high debt-service costs continue to limit spending on health, education, and infrastructure.”
Stable debt is good news. A stable weight is still a weight.
The other guests have left
For years, aid softened the blow. A country could pay its creditors and still lean on donors for vaccines, medicines and HIV programmes. That cushion is thinning fast.
The OECD’s preliminary 2025 aid data, published in April 2026, show that official development assistance from its donor members fell by 23.1% in 2025, to USD 174.3 billion. It is the largest annual contraction on record. Bilateral aid to sub-Saharan Africa fell by 26.3%, and the OECD projects a further 5.8% drop in total aid in 2026.
So the table is being laid with less food, and the lender’s plate has not shrunk. Something has to give. In budget after budget, it is the social sectors.
The case for paying first
Let me give the other side its due, because it is a serious argument. Debts are promises. A government that misses payments loses access to markets, pays more on its next loan and can tip into a crisis that hurts the poor far more than any budget squeeze. Borrowing also built roads, power lines and hospitals that are still standing.
The Book of Proverbs put the relationship bluntly long ago: “the borrower is servant to the lender.” Nobody in a finance ministry needs reminding.
I accept all of that. My quarrel is not with paying debts. It is with the price, and with what we have come to accept as normal.
UNCTAD estimates that if developing countries could borrow at the same rates as developed countries, they could collectively save around $500 billion a year in interest. That gap is not a reward withheld for bad behaviour. Much of it is a premium charged for being poorer, smaller and seen as risky by investors who rarely visit. Part of the bill is a tax on perception.
A question asked forty years ago
None of this is new. In 1986, according to the development economist Kevin Watkins, Tanzania’s first president, Julius Nyerere, asked his country’s creditors: “Must we starve our children to pay our debts?” Another account, co-written by Watkins in 2020, renders it as “Should we starve our people to pay our debts?”, and I have not found an original transcript.
The wording varies. The question has not aged a day.
The answer then was a long campaign for debt relief. The answer now cannot simply be to wait for the next relief round, because by the time it arrives another generation of children will have eaten last.
Changing the seating plan
What would a fairer table look like? UNCTAD sets out a menu. Internationally, it calls for stronger global and regional financial safety nets, action on structurally higher borrowing costs, more lending by multilateral and regional development banks, more technical assistance, and progress on a borrowers’ platform so that indebted countries can coordinate. African governments should push for these together, not one at a time.
But I want to be just as direct about the work at home, because UNCTAD names that too: stronger macroeconomic management and public institutions, and better-structured debt portfolios. Three things matter most to me.
First, know what you owe. A government that cannot see its full debt profile, including arrears to its own suppliers, cannot negotiate well. That is a research and data problem before it is a finance problem.
Second, cost the trade-off in human terms. When a budget raises interest and cuts health, the budget statement should say how many posts, doses or clinic hours that cut represents. A clear theory of change for each social programme makes that arithmetic possible, and makes it harder to hide.
Third, spend what is left as though it were scarce, because it is. Every dollar that survives the creditor’s plate should be tied to a result someone can measure. Waste is a luxury no indebted health system can afford.
The order of service
Every table has an order, and orders can change. Lenders deserve to be paid. They do not deserve to be served before a child’s vaccine, every year, without anyone asking why.
Until the price of borrowing reflects real risk rather than old assumptions, and until our own institutions can show exactly what each interest payment costs in nurses and classrooms, the lender will keep eating first.
At Praxis, our advisory and partnerships work helps ministries, programmes and partners build the evidence and institutional capability to make trade-offs like these visible and defensible. If your institution is facing them, let us talk.
Facing a similar challenge in your institution?
We work with teams on research, data, digital systems, and learning that hold up in practice.





