The changing environment means that the cost of servicing debts that once appeared sustainable has become a major obstacle to development in many countries. And for the countries caught between full concessional protection and genuine market strength, the cost of financing development has become acute. For the poorest countries, The World Bank’s International Development Association (IDA)'s stable low rates have insulated them from market volatility but not from the wider financing pressures of this era, including insufficient volumes, aid cuts, and gaps in emergency support.

For middle-income countries with market access, such as Angola and Egypt, and those with a mix of concessional and market access, such as Kenya, Senegal and Ghana, the cost of finance has increased dramatically.

The implications of this squeeze for human development are clear.

As aid budget cuts from North American and European donors bite and increased fuel and food prices hit households, the money available to governments to support populations through health and social protection programs is being eroded by high interest costs.

The Iran conflict adds a further dimension. Countries carrying heavy debt burdens now face two difficult scenarios: either commodity-driven inflation pushes global interest rates higher again, directly raising debt service costs, or a slowdown in global growth undermines export revenues and tax receipts. Both pathways lead to a similar outcome—less fiscal space at the moment it is most needed.

The consequences of all this are already devastating. In 2025, the number of children who died before their fifth birthday likely increased for the first time this century, based on projections. Between 638 million and 720 million people (7.8-8.8% of the global population) faced hunger in 2024. This is likely to worsen as food, fuel, and energy prices rise, with knock-on effects on remittances and household purchasing power across the developing world.

Rising costs across creditors

Not all developing countries borrow in the same way. Where a country sits in the global financing system—what it can access, from whom, and on what terms—is largely determined by its income level and creditworthiness as assessed by the World Bank.

And while headline rates have increased, the length of maturities and stability of some forms of lending matter more to some countries than the headline rate.

For our analysis, we separate countries into three groups to better understand the constraints each group of countries faces (See Annex for a detailed list of countries in each group).

Caught in the middle: The countries facing the sharpest squeeze

To understand the differentiated impact, this paper groups countries by their position in the global financing system, using World Bank IDA classifications. The three groups introduced above face distinct exposures—and distinct risks.

Group 1: Rate insulated, but supply constrained

For the world’s poorest countries, which had little access to market borrowing before COVID, the global rate surge has largely passed them by. IDA rates have held steady at around 1%, with maturities of 31 to 40 years and long grace periods. These countries face real financing challenges—insufficient volumes, inadequate climate finance, patchy access to emergency support—but rising market rates are not among them.

Their risk is not price, it is supply: whether IDA is sufficiently funded, and whether countries can access enough of it, quickly enough, when they need it.

Group 2: Hit hard, but with options

Middle-income emerging markets have felt the rate surge most directly in absolute terms. Rates surged from 1.4% to 5.2% between 2020 and 2024, which will materially raise debt service costs.

These countries, however, tend to have stronger fiscal positions, more diversified economies, and broader access to domestic capital markets. For most, the rate increase is a significant fiscal headache. It is not a crisis of market access.

And the long-term lending of IBRD is itself a form of concessionality (typically 20+ years) that some countries seek to take advantage of even if headline rates are more expensive than market borrowing.

Group 3: Squeezed from both sides

Group 3 countries—those on IDA blend or hardened terms that also have exposure to international bond markets—is where the pressure is most acute. This is the group this paper focuses on most closely.

These countries face a double exposure. On the concessional side, they do not benefit from standard IDA rates. Hardened and blend IDA terms carry rates of approximately 1-2%, with shorter maturities and reduced grace periods. On the market side, they are directly exposed to global rate movements—through existing bonds they need to refinance, through new borrowing at sharply higher rates, or through the effective closure of market access altogether.

This group includes countries that spent the last decade positioning themselves as frontier market success stories by issuing Eurobonds, building relationships with international investors, and demonstrating creditworthiness. Kenya, Senegal, Ghana, and Benin all fit this description. For them, the rate surge has changed the calculus fundamentally.

A further divide: issuers and the priced out

Within Group 3, the picture fractures further. Some countries—those with stronger credit ratings and established investor relationships such as Kenya, Benin, Nigeria and Honduras—were able to return to bond markets in 2024, albeit at significantly higher rates. Others have not issued at all. For these countries, the bond market has become so expensive that they have stopped borrowing from it altogether.

This matters beyond the individual country. The countries facing the highest borrowing costs are precisely those that have dropped out of the data. That means that the observed average understates the true cost of market finance for the most exposed borrowers.

Multilateral Development Bank lending still benefits countries even when more expensive than before

A comparison of interest rates tells part of the story but it misses something important: the structure of a loan matters as much as the rate.

A 7-year bullet bond and a 25-year World Bank loan at the same interest rate are not the same thing. The bond requires repayment of the full principal in year seven—meaning the borrower must either repay in full or refinance, at whatever rate the market offers at that point.

The World Bank loan spreads principal repayment over decades, with a grace period before repayments begin.

To compare these instruments fairly, we accounted for both the rate and the repayment profile and calculated dollars repaid per $100 borrowed over the full term of the loan, a calculation we call the Present-Value Repayment Ratio (PVR).

A PVR of 100 means the loan is priced broadly in line with market borrowing. A PVR below 100 means the borrower is receiving financing on better-than-market terms. The distance below 100 is the implicit saving, in present-value terms, per $100 borrowed. So if a loan has a PVR of 60, that means the borrower saves $40 per $100 relative to borrowing at market-equivalent rates.

For the creditors comparison, we take the commitment-weighted average IBRD and IDA lending terms and compute what those terms would cost each country in the bond universe, using that country's own market-equivalent rate as the discount rate.

For blend countries:

Although IBRD's rate (6.57%) is only about 3 percentage points below the bond rate (9.75%), the longer maturity and grace period push principal repayments far into the future, where they are heavily discounted at Kenya's high market yield.

Every $100 borrowed from IBRD costs Kenya just $76 in present-value terms, a savings of $24 per $100 relative to bond financing. IDA's combination of a much lower rate and 30-year maturity would provide a savings of $65 per $100 borrowed in bond markets.

On Kenya's actual $1.5 billion bond issuance in 2024 the excess present-value cost relative to IBRD terms was approximately $365 million. Had the same amount been borrowed on IDA terms the present-value savings would have been roughly $975 million.

For countries priced out of markets, the value of MDB lending is even larger

The figures above calculate savings for countries against actual bond issuances but this tells only part of the story.

When market yields spiked from 2022, many countries did not borrow more expensively. They stopped borrowing altogether. Issuances from blend countries fell from $17.9 billion in 2021 to $3 billion in 2022 and $1 billion in 2023.

For example, although Cameroon issued $812 million in bonds in 2021 at 5.9%, by 2022 its market-equivalent yield had nearly doubled to 11.3%. Pakistan issued $5 billion in 2021-22 at 7-8%, and by 2023 its implied borrowing cost had reached 19.6% before coming back down to 11.1% in 2024.

That means issuer-based comparisons are conservative. They capture the premium paid by countries still able to issue but they miss the countries facing the highest effective borrowing costs because those countries disappeared from the market altogether.

For countries that did not issue bonds, we estimate what markets would likely have charged them and how much those countries would have saved by accessing finance through IDA and IBRD. We do so using secondary-market yields on outstanding bonds where available, and rating-implied sovereign spreads where they are not.

These savings are hypothetical, because many countries did not have a choice of going to market at some of these rates. But they illustrate the importance of IBRD and IDA terms in a world where the market alternative was prohibitively expensive or unavailable altogether.

What needs to happen

In a high-rate environment, concessional finance alone cannot meet countries’ needs. Strengthening the full multilateral financing system—alongside faster restructuring—is now essential. The evidence in this paper points to three clear priorities.

1. Expand MDB lending and ensure that the non-concessional window delivers real value.

IBRD remains a critical source of long-term, relatively low-cost financing for middle-income and blend countries. Even in a higher-rate environment, it continues to offer more stable and predictable terms than market borrowing, particularly when longer maturities and smoother repayment profiles are taken into account.

The question is no longer just how to make the system more efficient—it is whether the non-concessional multilateral system can continue to provide a meaningful alternative to markets in a high-rate world.

The Capital Adequacy Framework spearheaded by the G20 can unlock $300-400 billion in additional lending headroom across the MDB system through better leveraging of MDB balance sheets on capital markets. Recent announcements from S&P mean that an additional $600-$800 billion more could be mobilized without new shareholder resources. The MDBs should take steps to implement reforms such as amending its Equity-to-Loan Ratio as was done in 2024.

2. Align MDB lending with country needs

Alongside volume, MDBs should expand the speed and flexibility of their lending to ensure that countries do not take on more expensive market lending simply because MDB lending is too inflexible for their needs.

A recent survey of 650 government and MDB officials across 125 countries showed that over 80% want predictable, flexible finance, but only two-thirds think MDBs deliver it well. While 79% want faster project delivery, only 47% say MDBs perform well on this. Nearly half say project timelines are too long. Only 48% think MDBs coordinate well with each other on the ground.

To improve flexibility and country alignment, MDBs should explore longer tenors, lifting caps on policy loans, and offering loans that use national procurement and implementation systems.

The objective is not to replace market financing but to ensure that multilateral lending consistently offers a clear advantage in cost, stability, and reliability—especially for countries facing high or uncertain access to capital markets.

3. Move faster on restructuring and create space to support countries with debt overhangs.

For countries that are priced out of bond markets and facing large near-term repayments, including bullet bond maturities, the question is no longer about the cost of new borrowing. It is about whether they can service the debt they already have.

Ghana has spent nearly two years in restructuring limbo while creditors coordinated—during which market access was closed, investment slowed, and the cost of inaction accumulated. Zambia’s experience was even longer, with years of delay driven by complex creditor coordination.

The Common Framework needs teeth: clearer timelines, stronger incentives for creditor participation, and mechanisms that prevent holdouts from blocking relief for countries in genuine distress. Faster restructuring is not just fairer—it is cheaper. The longer a country remains in limbo, the more expensive the eventual resolution becomes for everyone involved.

And for countries with liquidity challenges, MDBs and the IMF should provide support that enables and incentivises preemptive action to deal with their debt overhangs before they become solvency issues.

4. Protect IDA. It is the only consistently concessional source of finance.

IDA is not just another financing option. For many of the world's poorest countries, it is the difference between borrowing at 1-2% and borrowing at 6-10%—or not borrowing at all. The present value analysis in Section 4 makes this clear: every dollar of IDA financing saves multiples in debt service compared with any market alternative.

That advantage has grown, not shrunk, as global rates have risen. The case for a larger, faster-disbursing IDA is stronger today than it was three years ago.

IDA donors should resist the temptation to harden terms further. Every step toward market rates for countries still dependent on concessional access narrows the one buffer that has held. While scaling concessional finance remains critical, fiscal constraints in donor countries mean that IDA alone cannot absorb the full financing gap.

The cost of inaction

The cost of borrowing is a major issue but there is another cost that does not appear in the data: the cost of not acting.

Every year that IDA is underfunded, every month that restructuring is delayed, every loan that is slowed down by bureaucratic processes adds up to resources that do not reach schools or clinics or power grids. They are fiscal spaces that governments cannot use because they are servicing debt instead.

The global rate environment has made development finance more expensive. The Iran shock has made the outlook more uncertain.

The countries least responsible for either of these developments are bearing the greatest cost of both.

This outcome is not inevitable. It is a choice. And a different choice can—in fact, must—be made.

Acknowledgments

The paper is an early output from the Development Finance Observatory—a first-of-its-kind data platform launching later in 2026. Produced by ONE Data and The Rockefeller Foundation, the Observatory is being built to give policymakers, investors, and citizens a single, coherent picture of global development finance flows.

Annex

The following table outlines the countries included in each of the 3 groups considered in the analysis.