International development finance has entered a period of profound change. Funding from the traditional donor base fell from $223 billion in 2023 to an estimated $175 billion in 2025, and it could decline significantly further by 2029, effectively taking funding levels to where they were at the start of the Sustainable Development Goal era in 2016.
The sheer scale of this contraction is only part of the story. The system emerging in its wake is increasingly fragmented, bilateral, and more closely tied to donor priorities like trade, security, and migration. While new and expanding providers will play important roles, none are positioned to replace the volume, predictability, and concessionality of the old order.
The consequences of this shift will be highly uneven. While some recipient countries and sectors can pivot from highly concessional capital to loans, global capital markets, or private investment, low-income and fragile states remain heavily in need of grants. Furthermore, private capital has yet to materialize at the scale required to offset these shortfalls, leaving pooled mechanisms struggling to support critical needs that individual donors have little incentive to finance alone.
As leaders gather for the UN General Assembly in September and for the World Bank Group and International Monetary Fund annual meetings in October, the central question is expanding beyond how much aid has been cut. The true challenge is how to manage the new development finance system taking shape. Doing so will require redesigning financing structures where credible alternatives exist, while fiercely protecting grants and pooled funding where they do not.
The Traditional Aid Order Has Fractured
For decades, global development finance was anchored by a core group of bilateral donors within the OECD’s Development Assistance Committee (DAC). Having historically accounted for 85% of official aid, this base is now shrinking. Traditional official development assistance (ODA), excluding EU institutions and non-DAC countries, peaked at $223 billion in 2023 before falling to $175 billion in 2025 amid shifting political priorities and increasing fiscal pressures. At the center of this retreat is the United States, whose share of global aid is projected to drop from roughly 30% of the global pool to approximately 22% in 2029. With other major donors, including Canada, France, Germany, the Netherlands, and the United Kingdom facing their own fiscal pressures, traditional ODA is expected to fall to between $152 billion and $160 billion by 2029, erasing roughly a decade of funding growth.
Other donors will need to play an important role, but so far, they lack the means to close the gap entirely. While bilateral donors in Europe are pulling back on funding, EU institutions are increasing their contributions and could become the single largest development donor by 2029, at an estimated $29 billion. Meanwhile, reporting Gulf States, led by Saudi Arabia and the United Arab Emirates, have entered the top 15 donors list, yet they collectively accounted for just $8.5 billion, or 3.5%, of global ODA in 2024, and major nonreporting powers have provided only a small additional offset.
The funding squeeze is accelerating a broader shift from a donor-led model toward an investment-led approach in low- and middle-income countries. Expanding the historically strict definition of aid to include other official flows (OOFs), namely capital that supports development goals but lacks the highly concessional terms of traditional ODA and multilateral financing, offers little relief. Across bilateral providers and EU institutions, total development assistance is projected to fall from $350 billion in 2023 to as low as $275 billion by 2029.
The funding landscape is also becoming more diffuse, in line with broader geopolitical trends. While the top ten bilateral DAC country donors account for 52% of total development finance today, that share is projected to fall to 44% by 2029. With activity spread across a larger group of institutions and governments, it’s increasingly difficult to forge consensus. No single actor has the scale or mandate to set a common agenda. The result is a fragmented landscape where development finance is increasingly tied to narrow, donor-specific geopolitical interests rather than broad, systemic priorities.
A More Fragmented Funding System Is Rewriting the Rules
As traditional donors pull back, the remaining capital is being redirected toward strategic priorities. These funds are increasingly delivered through bilateral channels, offering less favorable terms or requiring specific commercial or political benefits in exchange, leading to quid pro quo deals, where the demands vary by donor. Because private investment has not filled the gap, institutions and recipient governments are left to navigate a more transactional funding system.
A Shift from Soft Power to Economic Statecraft. For half a century, global aid was largely driven by poverty reduction and human development. Today, geopolitics has reasserted its dominance, with bilateral donors moving capital away from traditional development priorities toward a narrower set of national security, trade, and migration objectives.
This changing mandate is redrawing the geography of aid. In Northern Europe, defense priorities are displacing human development programs. Sweden, for example, is phasing out bilateral aid to several African countries and redirecting more than $1 billion to Ukraine, with some Nordic countries likely to follow suit. Traditional donors are also replacing grant commitments with strategic investments. Japan’s $1.5 billion TICAD9 pledge operates explicitly as “trade, not aid,” while the EU’s Global Gateway and Italy’s $5.9 billion Mattei Plan link infrastructure financing to migration control, energy security, and access to critical raw materials. As a result, despite having the highest SDG gaps and being a focal point of global poverty, sub-Saharan Africa must increasingly compete with the Indo-Pacific and Eastern Europe for development funding.
A Growing Demand for Donor Control. Governments are increasingly favoring direct relationships that give them greater control over where funding goes and how it is used. In 2024, bilateral funding accounted for a record 63% of total development flows. Meanwhile, capital channeled through multilateral and pooled mechanisms fell 31% from its peak of $38 billion in 2021 to $26 billion in 2024. When fiscal pressures forced widespread budget cuts in 2025, donors largely protected their bilateral commitments, leaving pooled funds to absorb a disproportionate share of the reductions. Development organizations must now manage more individual sovereign relationships, each with its own priorities, processes, and conditions.
A Widening Two-Tier Financial System. Many donors expect institutional investment to offset declining public funding, but those flows have not materialized at scale. For instance, blended finance volumes reached $18 billion in 2024, but 90% of transactions bypassed the poorest nations and flowed instead to middle-income countries. The consequences vary sharply by country income level. Low-income countries remain dependent on grants, which account for 80% of their inflows. Total grant volumes to low-income countries remained flat (45 billion in 2011 to 43 billion in 2024) despite growing needs. Lower-middle-income countries watched their grant share fall from 46% to 33%, increasingly replaced by debt. Upper-middle-income countries now receive on average 34 times the amount of OOF as low-income countries. The result is a two-tier system in which the countries facing the deepest development needs depend on the scarcest and most constrained pool of capital. (See Exhibit 1.)
The Structural Replacement of Grants with Debt. Recipient governments are relying more heavily on loans, non-concessional official flows, and bilateral financing with stricter conditions. Although these instruments preserve near-term funding, they dramatically increase debt service burdens and place governments under severe and additional fiscal pressure. With many nations already in or at high risk of debt distress, substituting grants with debt severely restricts policy flexibility and leaves countries with less room to finance basic development priorities. Simultaneously, as capital becomes more fragmented, individual donor governments gain greater influence over its terms, uses, and destinations.
Make Every Remaining Source of Development Finance Work Harder
Development finance leaders should plan around a smaller, more fragmented system. That requires changes in how institutions assemble and deploy capital, and in how recipient countries assess and manage the terms on which capital arrives. AI will be a vital aid in this reset by improving transparency, lowering transaction costs, and sharpening financial analysis. To put that agenda into practice, we recommend ten specific actions organized around four stakeholder-specific priorities. (See Exhibit 2.)
Recipient countries can convert stronger fiscal management into a lower cost of capital. Countries need to raise and retain more revenue at home while improving the terms on which they borrow abroad. Measures they can take to do so include:
Expand domestic resource mobilization (DRM) through AI-enabled efficiency, transparency, and accountability. Low- and middle-income countries cannot replace lost aid through DRM alone. They can, however, prevent the depletion of scarce public resources from weak collection, poor financial management, and leakage. The immediate agenda is to broaden the effective tax base, strengthen public financial management, improve transparency, and identify where money is being lost. This takes fiscal discipline, stronger tax administration, and a concerted effort to close leakages and corruption.
AI makes that agenda more actionable. Governments can use AI-enabled tools to reconcile fragmented tax and payment records, identify anomalous transactions, detect patterns of underpayment, and direct audits toward the areas of greatest probable loss. These solutions can trace how public money moves through government accounts, making it harder for funds to disappear between collection and delivery. For many low- and middle-income countries, however, the cost of AI remains a barrier to deploying these tools at scale. Lower-cost and open-source AI models are expanding the available options, while donor support can help broaden access further. The aim is to collect more of what is already owed, retain more of what is collected, and improve visibility into how public funds are used.
Negotiate lower borrowing costs through stronger macro-fiscal management. Many low- and middle-income countries enter capital markets with weak country ratings and high-risk premiums. In part, these disadvantages reflect concerns about real economic and political risk. But some low ratings are a result of incomplete information and limited local depth by rating agencies.
Individual finance ministries can improve their standing by presenting reliable data, fully disclosing liabilities, and proving that reforms strengthen repayment capacity. Countries like Kenya and Nigeria already use this type of direct engagement strategy. Although individual action is making a difference, the impacts can be tremendous if coalitions of borrower countries come together to pool analytical resources, develop common reporting standards, and engage rating agencies as a group. Using this strategy, they can show how stronger tax collection, demonstrably lower leakage, more transparent public accounts, and better macro-fiscal management translate into a more accurate assessment of repayment capacity and a lower cost of capital.
These efforts reinforce each other. Recipient countries show how macro-fiscal reforms are improving the reliability of public finances. Creditors and rating agencies reflect that evidence in their treatment of sovereign risk. The gains can compound as lower leakage improves fiscal capacity, stronger fiscal capacity improves creditworthiness, and lower financing costs leave more room for investment in development.
Donor countries should leverage multilateral channels and use concessionality selectively. Retaining a large number of bilateral programs may preserve control and political visibility, but it also fragments capital into smaller pools. Recipient governments must then negotiate multiple agreements, satisfy different reporting requirements, and assemble separate contributions around the same development need. By making three structural changes, donors and recipients can trigger a virtuous cycle that stretches every remaining dollar further:
Boost grant funding in multilateral replenishments. Institutions such as the International Development Association (IDA) and regional development banks’ concession funds, the Green Climate Fund, and the major global health funds combine contributions from many donors with reflows from earlier lending, market borrowing, technical expertise, and established fiduciary systems. This can increase the scale of available funding, allocate concessionality more efficiently across countries and priorities, and reduce the burden of assembling multiple bilateral contributions around the same development need. (See Exhibit 3.)
Pooled aid remains essential for basic services in countries that cannot fund these services themselves as well as for global and regional public goods whose benefits cross borders. Climate mitigation, pandemic preparedness, and regional stability all create value beyond the country in which the investment is made. Multilateral institutions are better placed to finance those public goods collectively.
Greater reliance on multilateral channels, however, must come with better coordination among the institutions themselves. Recipient governments can still find themselves managing dozens of funders, each with its own planning cycles, requirements, and reporting. That fragmentation absorbs scarce administrative capacity and slows deployment. Efforts such as the Lusaka Agenda, which seeks greater strategic and operational coherence among global health initiatives, are moving in the right direction, and they should be strengthened and replicated across the broader development finance ecosystem.
Protect grants as a primary vehicle and deploy catalytic capital judiciously. The contraction in aid funding is creating pressure to use every remaining grant dollar to mobilize private investment. However, private investment would direct scarce grant funding toward activities capable of generating a return and away from vulnerable populations, fragile states, and services that cannot support repayment. Grants must remain available for those needs.
Where private investment is viable, concessional capital should provide only the support required to bring it in. The test is whether the funding changes the investment decision and how little concessionality is needed to do so. Anything more risks transferring public subsidy to private investors without producing additional development impact.
Make concessional capital easier to find and cheaper to access. Grant, philanthropic, impact, and multilateral concessional funds are scattered across innumerable institutions, each with its own mandate and requirements. Some money remains unused because potential recipients cannot see where it sits, do not know how it can be used, or face application and compliance costs that are too high.
A global registry could set out the available pools, their permitted uses, their terms, and their access requirements. Fund holders should simplify their processes and balance compliance demands to the amount and risk of the funding. Standard sector templates could reduce the need to reconstruct similar transactions project by project.
AI can maintain the registry, match projects with suitable sources of finance, keep sector templates current, and automate parts of the application and compliance process. Here, the technology addresses a specific constraint: the cost of finding and securing concessional capital, which prevents existing money from being put to work.
Development finance institutions should continue to build markets that attract commercial finance. As more grant funding moves through multilateral channels, bilateral development finance institutions have a distinct role. They can bridge the space between grants and fully commercial finance in sectors that have long depended on aid. The aim is to extend the range of development needs that can attract appropriate finance. That leaves grants concentrated on health, education, climate, and other systems that cannot support a return. To achieve this goal, development finance institutions can use several strategies:
- Enable market creation in the social sectors. Health, education, and climate-related investment remain heavily dependent on ODA grants across countries at different income levels. Bilateral development finance institutions should identify the parts of those sectors that can support commercial, semi-commercial, or lightly concessional capital and concentrate their effort to deploy capital where it can have the highest development impact. This means funding innovations in business models and supporting markets previously reliant on public subsidy.
- Standardize repeatable capital stacks and structures. Financing should increasingly be considered sector by sector as well as project by project. A sector-level structure can establish which activities require grants, where catalytic funding needs to absorb early risk, what can support development-finance debt or equity, and when domestic or international commercial investors can participate. A consistent challenge has been the desire for “first ever” transactions rather than standardized and repeatable plays. A replicable deal is far more valuable than one that is innovative but difficult to replay.
- Team up with public development banks (PDBs) to prove the domestic finance model. Bilateral institutions can partner with local investors and financial institutions, including PDBs, providing longer tenors, absorbing early risk, or offering limited concessional support while a market develops. Once the economics and performance record become clearer, domestic capital can take a larger role and the level of concession can fall.
Multilateral development banks (MDBs) should push their balance sheets closer to their true limits. MDBs have already taken various measures to expand lending capacity, including through capital-adequacy reforms. They now need to take steps to test how much further their balance sheets can actually go:
- Calculate the foregone development impact of preserving top credit ratings. A triple-A rating allows an institution to borrow cheaply and pass those terms to countries that would otherwise face far higher rates. Indeed, this is an essential component of the MDB value proposition and is a powerful tool of the development mandate, but it is not the mandate itself. Development impact is the mandate. Boards and shareholders should assess the tradeoff across the whole system. How much would a rating change increase the bank’s own cost of capital? How much additional lending would it release? What additional development impact could be achieved? What would recipient countries pay if they had to raise the same money directly? Some low- and middle-income sovereign borrowers face rates of 10% to 13% in global capital markets. An MDB borrowing at AA+ instead of AAA could still provide finance on far better terms. Whether the additional lending justifies a modest increase in the bank’s funding cost, particularly given the strong repayment record of sovereign borrowers to multilateral institutions, is a proposition worth analyzing carefully.
- Test lending boundaries through scenario modeling and build snap-back strategies. The MDBs also cannot know their true prudential limit while remaining safely inside it. They should model more ambitious lending scenarios, move closer to the boundary, and prepare in advance for the possibility that a rating threshold is crossed. An institution can’t really know the limit until it is crossed. However, a downgrade may be reversible with adequate snap-back planning through changes to lending, capital, or portfolio management. The objective is not to pursue a weaker rating but to establish exactly how much prudential headroom exists. By proactively preparing a reversal strategy, institutions can stop treating the mere risk of a downgrade as a hard limit, shifting the governing question from protecting the AAA rating to maximizing development impact.
In the new development aid system, success will depend on assembling the right mix of providers, instruments, and terms around each development need. That requires sharper choices about where commercial finance can work, where coordination adds value, and where grants remain irreplaceable. Lower levels of traditional forms of aid do not have to mean less development. Rather, this reduced aid demands a fundamentally different financial system.
The authors thank Elena Molina and Bart De Langhe for their contributions to this article.
