Debt is squeezing poor countries. Its effect on health spending is subtler

Developing countries spent a record US$1.4 trillion servicing their foreign debt in 2023, with interest costs alone up nearly a third to US$406 billion — a 20-year high that the World Bank says is "squeezing the budgets of many countries in critical areas such as health, education, and the environment" [s1]. Whether that squeeze actually shows up as less domestic health spending, though, turns out to be a harder question than the headline implies: a peer-reviewed study of 105 low- and middle-income countries finds the channel from debt to health financing is real but small, and the common claim that IMF programmes force health cuts is not visible in the data at all [s2].

The debt numbers are not subtle

There is nothing modest about the fiscal pressure itself. The World Bank's International Debt Report 2024 found the strain fiercest for the poorest countries, those eligible to borrow from its concessional arm: they paid a record US$96.2 billion to service debt in 2023, and although principal repayments fell nearly 8% to US$61.6 billion, their interest costs hit an all-time high of US$34.6 billion — four times the level of a decade earlier [s1]. Interest now consumes close to 6% of those countries' export earnings on average, a share not seen since 1999, and as much as 38% for some [s1]. Total external debt owed by all low- and middle-income countries stood at a record US$8.8 trillion at the end of 2023 [s1].

The money is also flowing the wrong way. Since 2022, foreign private creditors have taken nearly US$13 billion more in debt-service payments out of public borrowers in the poorest economies than they put in as new financing, leaving multilateral institutions — the World Bank chief among them — as the lenders of last resort [s1]. "In highly indebted poor countries, multilateral development banks are now acting as a lender of last resort, a role they were not designed to serve," the Bank's chief economist, Indermit Gill, said [s1].

What the study actually measured

The intuition is straightforward: if a government is spending more to service debt, it has less for clinics and salaries, and patients make up the difference out of pocket. The Health Policy and Planning study, by Frederik Federspiel and Josephine Borghi of the London School of Hygiene & Tropical Medicine, tested that intuition directly. Using a panel of 105 low- and middle-income countries from 2005 to 2019 and a dynamic panel (generalized method of moments) estimator, it looked at how development aid, public external debt servicing and IMF loan conditionalities relate to two things: what governments spend on health from their own budgets, and what households pay out of pocket [s2].

The debt effect came through in the expected direction, but small. A 1% increase in the prior year's public external debt servicing per GDP was associated with a 0.007-percentage-point rise in the out-of-pocket share of current health expenditure and a matching 0.007-percentage-point fall in the government-health-spending share [s2]. In other words, heavier debt service does tilt the financing mix away from government funds and toward patients' own pockets — the outcome that most concerns advocates of universal health coverage — but the measured tilt is slight.

The findings that cut against the script

Two results complicate the standard narrative. First, development aid for health did not simply add to what governments spend. A 1% increase in the previous year's on-budget health aid per GDP was associated with a 0.024% reduction in out-of-pocket spending and a 0.026% decrease in government health spending, both measured per GDP [s2] — aid modestly displacing domestic effort rather than being wholly siphoned off, which the authors read as evidence of limited "fungibility" in the health sector. Second, and more pointedly, the study "found no relationship between IMF programme participation or conditionalities and" either government health spending or out-of-pocket payments [s2]. The much-repeated charge that IMF austerity mechanically shrinks health budgets simply did not appear in this dataset.

How to read it

The limits matter. These are associations across countries over time, not proof of cause, and the window closes in 2019 — before the pandemic-era borrowing and the interest-rate surge that produced the record 2023 figures the World Bank reports [s1][s2]. It is entirely possible the debt-to-health channel has widened since. But taken together the two sources point to a more precise story than the usual one: debt service is a large and growing drain on the budgets of the poorest countries [s1], yet its pass-through to the health-financing mix, at least through 2019, was modest and gradual, and the IMF-austerity mechanism often blamed for health cuts was not detectable [s2]. For anyone deciding where to push — debt relief, protected health budgets, or fund conditionality — knowing which lever actually moves the money is the difference between a slogan and a policy.

Sources

• [s1] Developing countries paid record $1.4 trillion on foreign debt in 2023 (International Debt Report 2024), World Bank, 3 December 2024. https://www.worldbank.org/en/news/press-release/2024/12/03/developing-countries-paid-record-1-4-trillion-on-foreign-debt-in-2023
• [s2] Aid, debt, International Monetary Fund conditionalities and domestic health financing in low- and middle-income countries, Health Policy and Planning (Federspiel & Borghi), 21 May 2026. https://doi.org/10.1093/heapol/czag070