Development Finance and Institutional Resilience in Emerging Markets

In brief

The development finance system that serves emerging markets and low-income economies is under simultaneous strain on three fronts: a financing requirement that has widened faster than available capital, a debt overhang that is crowding out public investment, and a cost of capital that keeps private money away from precisely the countries and projects that most need it. The United Nations Conference on Trade and Development (UNCTAD) estimates the annual financing gap for the Sustainable Development Goals (SDGs) in developing economies at roughly US$4 trillion, up from about US$2.5 trillion when the goals were adopted in 2015. Concessional public flows are an order of magnitude smaller. The arithmetic is unforgiving: aid volumes matter, but they cannot close the gap on their own. The system's central task has shifted from disbursing concessional finance to mobilising domestic resources and private capital at scale — and, so far, it is not doing so.

The second strain is debt. After a decade of low global interest rates that encouraged emerging-market borrowing — including a wave of first-time sovereign bond issuers in Sub-Saharan Africa — the tightening cycle that began in 2022 raised refinancing costs sharply and closed international markets for many sovereigns. Roughly half of the low-income countries eligible for IMF concessional lending are now at high risk of, or already in, debt distress, with interest payments squeezing health, education and climate-resilience spending, while a restructuring architecture centred on the G20 Common Framework has proved slow and unpredictable.

The third strain, and in our assessment the most decisive, is the cost of capital. The obstacle to investment in emerging markets is rarely a shortage of viable projects; it is that the risk-adjusted returns demanded by international investors, layered on top of high sovereign borrowing costs and currency risk, render otherwise sound projects unfinanceable. A clean-energy project in a lower-income market can face financing costs several times those of an identical project in an advanced economy, chiefly because of macroeconomic, currency and perceived-risk premia rather than the technology. This is where the multilateral and bilateral development finance institutions (DFIs) have the clearest comparative advantage — and where reform effort is concentrated.

This report maps the system's structure and scale, examines these constraints, and sets out three scenarios to 2030. Our central judgement is that incremental, cumulative reform — balance-sheet optimisation at the multilateral development banks (MDBs), better use of guarantees and local-currency instruments, and slow repair of the debt architecture — is the most probable path, and narrows the gap without closing it. What separates the scenarios is less the quantity of global capital than the political economy of directing it toward the highest-need, highest-cost destinations.

Key judgements

• The gap is a mobilisation problem, not an aid problem. With the SDG financing gap estimated near US$4 trillion a year and concessional flows in the low hundreds of billions, the decisive variable is how much domestic and private capital each public dollar can crowd in — and current leverage is far below what the "billions to trillions" agenda assumed a decade ago.

• Debt distress is widespread and the workout machinery is slow. Around half of concessional-eligible low-income countries are at high risk of or in debt distress. The G20 Common Framework has resolved only a handful of cases, each after prolonged negotiation, and creditor fragmentation — traditional bilaterals, China, private bondholders — complicates comparability of treatment.

• Cost of capital, not project pipeline, is the binding constraint. Currency risk, sovereign risk premia and perceived (as distinct from realised) risk push financing costs for emerging-market infrastructure well above advanced-economy levels; closing even part of that wedge would matter more than any single new fund.

• MDB reform can add meaningful headroom without new capital. Implementing the G20 Capital Adequacy Framework review — adjusting leverage, capturing callable capital and offloading risk — could expand collective MDB lending capacity by an estimated US$300–400 billion over roughly a decade, though adoption is uneven.

• Blended finance has under-delivered at the frontier. The annual market is an estimated US$15–20 billion, and private-capital leverage in low-income countries typically sits below one dollar mobilised per public dollar — far short of the multiples the model needs to scale.

• Remittances are the quiet giant. At an estimated US$650–685 billion a year to low- and middle-income countries, remittances exceed FDI and official aid combined for many economies, yet remain lightly instrumented and burdened by high transfer costs.

• The provider landscape has fragmented. China's rise as a bilateral creditor, the growing role of Gulf and other non-DAC providers, and new multilateral vehicles have widened the menu but reduced coordination — raising both the supply of capital and the transaction costs of restructuring and standard-setting.

1. Context and why it matters

Development finance is the connective tissue between global savings and the investment needs of economies that cannot yet fund those needs domestically or on commercial terms alone. It comprises official development assistance (ODA) from bilateral donors and multilateral agencies; the lending and investment of the MDBs and bilateral DFIs; private capital mobilised alongside public money; and the domestic resources that governments raise through taxation and local capital markets. For most of the post-war period the debate centred on the volume and quality of aid. That framing is now inadequate to the scale of the challenge.

Three shifts explain why. First, the ambition embodied in the 2015 SDGs and the Paris Agreement raised the investment bar sharply — particularly for energy systems, adaptation and resilient infrastructure — while the concessional envelope did not grow commensurately. Second, the sequence of shocks since 2020 — the pandemic, the rate-tightening cycle, and commodity and food-price volatility — simultaneously increased spending needs and reduced fiscal space, leaving governments with less room to invest just as the required investment rose. Third, the geopolitics of capital has changed: the provider landscape is more crowded and less coordinated, and the assumption that a broadly unified group of Western creditors and institutions sets the terms no longer holds.

Why this matters beyond the countries directly affected is straightforward. Under-investment in health systems, energy and climate resilience produces cross-border consequences — in migration pressure, pandemic risk, food-price transmission and emissions — that no single government can wall off. For investors, emerging and frontier markets represent both the fastest-growing share of the global consumer base and the largest concentration of un-served infrastructure demand. For the multilateral system, the credibility of the institutions built after 1944 rests on demonstrating relevance to a problem of this magnitude. The stakes are systemic.

2. Market structure and scale

Development finance is best understood not as a single market but as a set of overlapping channels with very different volumes, price points and risk appetites. Concessional flows are small but carry the highest subsidy; commercial and private flows are far larger but reach only a narrow band of investment-grade or near-investment-grade destinations. The table below sizes the principal channels of external finance reaching developing economies. Figures are gross annual flows in the mid-2020s and should be read as order-of-magnitude estimates; several channels overlap (MDB-mobilised private capital, for example, appears both under private mobilisation and, in part, within FDI), so the rows must not be summed.

The ranges above make the scale mismatch visible — the low and high bounds of each channel plotted side by side:

Two structural features stand out. The first is the dominance of private, market-driven flows — remittances and FDI — over official channels. In many low- and middle-income economies remittances alone exceed both FDI and ODA, making household transfers, rather than institutions, the largest external funder. Because remittances tend to be counter-cyclical and stable, reducing transfer costs — still averaging well above the SDG target of 3 per cent — is among the highest-return interventions available.

The second feature is the mismatch between where concessional risk-bearing capacity sits and where the largest pools of private capital are willing to go. Global institutional investors manage assets in the tens of trillions, but allocations to emerging-market infrastructure and low-income countries remain a rounding error within that total. The DFIs exist to bridge that gap by absorbing risks the market will not price efficiently, yet their combined balance sheets, even after reform, are modest relative to the requirement. This is why the leverage on public capital — how many private dollars each public dollar mobilises — is the metric that matters most, and why raising it from levels well below what the system needs is the central design problem of the next decade.

3. Drivers and structural dynamics

Several forces are reshaping the system at once. The first is the reappraisal of MDB balance sheets. The 2022 G20-commissioned review of Capital Adequacy Frameworks concluded that the MDBs had been managing their balance sheets conservatively relative to their actual risk, and recommended measures — recalibrating capital-adequacy ratios, giving analytical weight to callable capital, expanding risk transfer to reinsurers and other MDBs, and instruments such as hybrid capital — that could expand lending capacity substantially without new paid-in capital. Our estimate, consistent with the range in the official review and subsequent disclosures, is that full implementation could add on the order of US$300–400 billion in cumulative headroom over roughly a decade. This is meaningful but not decisive relative to the gap, and adoption has been incremental because it interacts with the banks' triple-A credit ratings, which shareholders are reluctant to jeopardise.

The second driver is the climate-development nexus. Adaptation and clean-energy investment needs run to hundreds of billions of dollars a year, concentrated in exactly the economies with the least fiscal space and the highest cost of capital. This has pulled the MDBs and DFIs toward instruments designed to make climate projects financeable — guarantees, first-loss tranches, currency-hedging facilities — and toward a mandate debate about whether the banks should price and reward global public goods (emissions reductions, pandemic preparedness) that a single sovereign borrower has little incentive to fund. The Bridgetown Initiative and the 2023 Summit for a New Global Financing Pact in Paris crystallised these ideas, including automatic debt-service suspension in disasters and expanded use of the IMF's Special Drawing Rights (SDRs).

The third driver is monetary and macro-financial. The post-2022 rate environment reset the arithmetic of emerging-market debt: sovereigns that issued Eurobonds in the low-rate 2010s faced refinancing at far higher yields or lost market access entirely, and currency depreciation against a strong dollar raised the local-currency burden of foreign-currency debt. Even as global rates ease from their peak, the term premium demanded of frontier issuers remains elevated, and the "original sin" of borrowing in foreign currency continues to transmit global financial conditions directly into domestic fiscal stress.

The fourth driver is the recycling of SDRs. The IMF's 2021 allocation of about US$650 billion delivered reserves disproportionately to advanced economies that did not need them; efforts to rechannel a portion — through the Poverty Reduction and Growth Trust, the Resilience and Sustainability Trust, and proposals to route SDRs via MDBs as hybrid capital — aim to move idle reserves toward countries that do. Pledges have reached the symbolic US$100 billion mark, but conversion into deployed, leverage-generating capital has been partial, constrained by the reserve-asset characteristics that make SDRs attractive to hold but awkward to lend.

4. The debt overhang and the cost-of-capital problem

The debt problem and the cost-of-capital problem are two faces of the same constraint. On debt, the headline concern is less a single systemic default event than a slow, grinding liquidity squeeze: a rising share of government revenue diverted to interest payments, particularly external interest, at the expense of investment. For a cohort of low- and lower-middle-income economies, debt service now consumes a larger share of revenue than health or education — a development reversal even where headline solvency is not in question, because it forecloses the investments that would raise future capacity to pay.

The restructuring architecture has not kept pace with creditor fragmentation. The G20 Common Framework, established in 2020, was intended to bring China and other non-Paris-Club bilateral creditors into a common process alongside traditional lenders and to require comparable treatment from private bondholders. In practice, early cases moved slowly, comparability across heterogeneous creditors proved contentious, and the reputational cost of entering the process discouraged pre-emptive restructuring — the opposite of what an efficient system would reward. Recent cases show that agreements are reachable, but resolution remains slow enough to impose real economic damage during the interregnum, when access to new finance is frozen.

The cost-of-capital problem is the more consequential of the two because it applies even to solvent, well-run economies. The wedge between financing a project in an advanced economy and an identical project in a lower-income market is driven overwhelmingly by macroeconomic and currency risk and by perceived rather than realised default risk. Evidence from multilateral guarantee portfolios and infrastructure-loan default databases suggests that realised loss rates on emerging-market infrastructure are often lower than credit spreads imply — a gap between perception and outturn that, if corrected through better data, guarantees and risk-sharing, would reduce financing costs without subsidy. That is the case for the DFIs' guarantee and risk-transfer instruments and for publishing standardised credit-performance data. Currency risk is the harder half: local-currency and hedging facilities shift exchange-rate risk away from the borrower, but hedging thin or volatile currencies can be prohibitively costly, and public hedging capacity remains small relative to demand.

5. Instruments, mobilisation and the competitive landscape

The instruments through which development finance is delivered have proliferated, but their effectiveness at mobilising private capital varies widely. Concessional loans and grants carry the deepest subsidy and reach the poorest countries but do little to crowd in private money; pure market instruments — sovereign bonds, commercial project finance — reach only investment-grade or near-investment-grade destinations. The policy interest lies in the middle: guarantees, first-loss and mezzanine tranches in blended structures, local-currency facilities, and portfolio risk-transfer.

Blended finance — using concessional public or philanthropic capital to improve the risk-return profile of a transaction for private investors — is the instrument most often cited as the bridge to scale. The reality is more sobering: the annual market is an estimated US$15–20 billion of deal flow, small relative to the need and lumpy year to year. More importantly, the leverage achieved is disappointing where it is most needed: in low-income countries, each public or philanthropic dollar typically mobilises less than a dollar of private capital, well below the multiples of four or more sometimes assumed. The reason is structural — the poorest markets carry macro, currency and regulatory risks that concessional first-loss layers only partly offset, so private investors demand a large cushion that consumes the subsidy without multiplying it. Blended finance works best in middle-income markets and creditworthy sectors — also where private capital would more readily flow anyway, raising legitimate questions about additionality.

The competitive and institutional landscape has widened. Alongside the World Bank Group and the long-established regional development banks sit newer multilateral vehicles, a growing set of bilateral DFIs across Europe and North America, and non-DAC official providers — most consequentially China, whose policy-bank lending reshaped the creditor map in the 2010s before moderating toward smaller, greener and more risk-conscious engagement, with Gulf funds and other emerging providers adding capacity. This proliferation increases the aggregate supply of capital and the menu of terms available to borrowers, but it also raises coordination costs: standard-setting on transparency, safeguards and debt sustainability is harder across a fragmented base, and creditor heterogeneity complicates restructuring — more choice for borrowers, less coherence for the system.

6. Regional and comparative lens

The pressures described above are unevenly distributed. Sub-Saharan Africa concentrates the sharpest version of the cost-of-capital and debt problems: a large cohort of countries that gained market access in the 2010s, high exposure to currency and commodity shocks, credit ratings clustered below investment grade, and adaptation needs among the highest in the world relative to fiscal capacity. The debate over an "Africa premium" — that the continent's sovereigns and projects are priced for more risk than their realised default experience warrants — is, in our reading, partly supported by loss-rate evidence, though data gaps make the mispricing hard to quantify.

South and South-East Asia present a different profile: deeper domestic capital markets, more diversified export bases, and — for the major economies — near-investment-grade access that changes the mobilisation calculus. Here the constraint is less market access than the scale of energy-transition investment required and the challenge of channelling domestic institutional savings, including large pension and insurance pools, into long-dated infrastructure. Latin America occupies an intermediate position, with reasonable market access offset by political-cycle volatility and, in places, elevated debt burdens. Small island and climate-vulnerable states face a distinct problem: too small to attract efficient private flows, highly exposed to disaster shocks, and often classified by income status in ways that restrict access to concessional finance — the case that motivated the Bridgetown proposals for vulnerability-based eligibility and disaster-contingent debt clauses.

7. Risks, uncertainties and open questions

Several risks could widen the gap rather than narrow it, and honest analysis must weight them. The first is a renewed tightening of global financial conditions — whether from persistent inflation, geopolitical shocks or a risk-off repricing — which would raise emerging-market borrowing costs, close market access for frontier issuers and deepen debt distress. The second is fiscal retrenchment among traditional donors: ODA reached record nominal levels in the early 2020s partly because of in-donor refugee costs and one-off crisis spending, so the underlying trend in country-programmable aid is more fragile than the headline suggests.

The third risk is that MDB reform stalls at the level of communiqués, constrained by shareholders' protectiveness of credit ratings and the absence of a mechanism to compel adoption. The fourth is that the debt-restructuring architecture fails to improve, leaving future distress cases to grind through prolonged, ad hoc negotiations that destroy value. The fifth is a coordination failure among a fragmented provider base, in which competition on terms erodes transparency and debt-sustainability discipline. Against these sit genuine upside possibilities: faster SDR rechannelling, a step-change in guarantee usage, or a durable narrowing of the perception-reality risk gap through better data could each lift mobilisation materially. The distribution of outcomes is wide, and it is driven more by political-economy choices than by the global availability of capital.

Three scenarios to 2030

We frame the outlook through three named scenarios. These are not forecasts but internally consistent narratives, each with a qualitative judgement about how much of the financing gap is likely to be narrowed. Probabilities are illustrative and reflect our assessment, not a measured distribution.

Scenario A — Incremental Reform (our central case; roughly a 50–55 per cent judgemental weight). MDBs implement the Capital Adequacy Framework agenda gradually, adding headroom in the estimated US$300–400 billion range over the decade; guarantee and risk-transfer usage grows steadily; SDR rechannelling progresses but remains partial; and the debt architecture improves at the margins, with restructurings resolving somewhat faster than early Common Framework precedents. Private mobilisation rises but leverage stays below the multiples needed to close the gap. The annual financing gap narrows meaningfully but remains large — measured in trillions rather than eliminated — and progress is uneven, strongest in middle-income economies and weakest in the poorest and most climate-vulnerable states.

Scenario B — Coordinated Acceleration (roughly 20–25 per cent). A sequence of summits and shareholder decisions converts reform intent into binding action: full CAF implementation, a substantial hybrid-capital injection from rechannelled SDRs, standardised and published emerging-market credit data that compress risk premia, a working guarantee platform at scale, and a debt architecture that rewards early, orderly restructuring. Private leverage rises toward the multiples the model requires. The gap narrows sharply, though even here it is contained rather than closed by 2030. This scenario depends less on new money than on political alignment among major shareholders and creditors.

Scenario C — Fragmentation and Retrenchment (roughly 20–25 per cent). Global financial conditions tighten or remain elevated; donor budgets contract; MDB reform stalls amid rating-agency caution; and creditor fragmentation defeats orderly restructuring. Debt distress spreads, market access narrows for frontier issuers, and climate and development investment falls further behind need. The gap widens — not through a single dramatic crisis so much as a compounding of missed opportunities and slow-burn liquidity stress across a broad cohort of economies.

Across all three, the central point holds: the constraint is not the global stock of capital, which is ample, but the terms and channels through which it reaches high-need, high-cost destinations — a matter of institutional and political choice, not macro luck.

Implications by audience

Governments and policymakers (borrowing economies). The highest-return domestic agenda is mobilising own resources — broadening tax bases, strengthening revenue administration, and deepening local capital markets so that long-dated domestic savings can fund long-dated domestic investment. On external finance, the priority is to sequence borrowing prudently, favour concessional and local-currency instruments, build contingency and disaster clauses into new debt, engage creditors pre-emptively at the first sign of stress, and invest in the transparency and data that lower perceived risk.

Governments and policymakers (shareholder and donor economies). The most cost-effective lever is not necessarily larger aid budgets but fuller implementation of MDB balance-sheet reform, which expands capacity without new paid-in capital; expanded and accelerated SDR rechannelling, particularly as MDB hybrid capital; and support for a debt architecture that makes orderly, timely restructuring the rational choice. The concessional base for the poorest countries, where private leverage is weakest, should be ring-fenced from broader mobilisation ambitions.

Business, investors and financial institutions. For institutional investors, emerging-market infrastructure and the energy transition represent a large, under-allocated opportunity whose principal barrier is structuring rather than availability of assets. Engagement with guarantee platforms, currency-hedging facilities and blended structures can access risk-adjusted returns that headline spreads overstate, particularly where realised loss experience diverges from perception. For corporates, project bankability increasingly turns on early involvement of DFIs and on instruments that address currency and off-take risk.

MDBs, DFIs and multilateral institutions. The institutional imperative is to demonstrate that reform translates into higher private mobilisation per public dollar — the metric on which the system's relevance rests. That means prioritising guarantees and risk-transfer over balance-sheet lending where they crowd in more private capital, publishing standardised credit-performance data to compress risk premia, scaling local-currency capacity, and coordinating standards across a fragmented provider base so that competition raises supply without eroding discipline.

Donors and funders (philanthropic and concessional). Concessional and philanthropic capital is scarcest and most valuable where private leverage is weakest — the poorest and most vulnerable economies. Directing it toward first-loss layers, pipeline development and public goods that no single sovereign will fund, rather than middle-income transactions that private capital would reach anyway, is the additionality test that should govern deployment.

Methods and confidence

This report is a synthesis and interpretation of publicly available information rather than the product of primary data collection. Our approach triangulates across four kinds of source: (1) official multilateral data — the OECD Development Assistance Committee for ODA, the IMF and World Bank for debt-sustainability assessments and country classifications, UNCTAD for FDI and financing-gap estimates, and the disclosures of the development finance institutions; (2) specialist market data on blended finance and mobilisation; (3) the analytical literature and official reviews on MDB capital adequacy, sovereign debt restructuring and the cost of capital; and (4) our own structured judgement, applied in the scenario analysis.

Readers should treat the quantitative claims with the confidence levels we have attached to them. A small number are measured and well-established: ODA volumes reported by the OECD DAC, remittance series compiled by the World Bank, and the IMF's debt-distress classifications are reported data, subject to the usual revision and definitional caveats. The large aggregate — the roughly US$4 trillion annual SDG financing gap — is itself an estimate produced by UNCTAD under stated assumptions, and different methodologies yield different figures; we use it as a widely cited order-of-magnitude benchmark, not a precise measurement. All flow figures in the market-structure table are estimates presented as ranges, with overlapping definitions that preclude summation. The reform-headroom figure (an estimated US$300–400 billion over a decade) is derived from the official Capital Adequacy Framework review and subsequent disclosures, and depends heavily on implementation choices, and the scenario probabilities are explicitly judgemental. Where we express a view, its basis and uncertainty are stated, consistent with the Institute's standard of methodological transparency.

References and sources

Official multilateral series and the primary reform documents below were read directly; specialist market data on blended finance and mobilisation are used as published, with methodologies noted.

• United Nations Conference on Trade and Development (2023). World Investment Report 2023: Investing in Sustainable Energy for All. UNCTAD, Geneva.
• OECD (2024). Development Co-operation Report 2024 and official development assistance statistics. OECD Publishing, Paris.
• International Monetary Fund and World Bank (2024). Debt Sustainability Framework for Low-Income Countries: list of LIC DSAs. IMF and World Bank, Washington, DC.
• G20 Independent Review of Multilateral Development Banks' Capital Adequacy Frameworks (2022). Boosting MDBs' Investing Capacity. G20, Washington, DC.
• Convergence (2024). State of Blended Finance 2024. Convergence, Toronto.
• World Bank and KNOMAD (2024). Migration and Development Brief 40: remittance flows to low- and middle-income countries. World Bank, Washington, DC.
• International Monetary Fund (2024). Fiscal Monitor and World Economic Outlook. IMF, Washington, DC.
• Government of Barbados (2023). The Bridgetown Initiative 2.0. Bridgetown.
• Summit for a New Global Financing Pact (2023). Chair's summary and roadmap. Paris.
• World Bank Group (2023). Evolution Roadmap: a new vision and mission. World Bank, Washington, DC.