The End of Aid? South-South Finance and the Return of Domestic Responsibility

Official development assistance is a precisely defined category. To count as ODA under OECD Development Assistance Committee rules, a financial flow must be provided by official agencies, administered with the promotion of economic development and welfare of developing countries as its main objective, and be concessional in character. In practice, this definition is broad enough to encompass bilateral grants, concessional loans, multilateral contributions, technical cooperation, debt relief operations and, since rule changes in the 1980s and again after the 2015 European refugee crisis, spending on refugees within donor country borders.

The distinction between these categories matters enormously for understanding what development finance actually does. A grant transferred directly to a recipient government’s budget finances public expenditure. A technical assistance payment disbursed to a consultant from the donor country primarily finances that country’s own expertise sector. An in-donor refugee cost never crosses a border at all. All three appear identically in the headline ODA total. When analysts or policymakers cite $223.7 billion as the measure of global generosity toward development, they are citing a figure that aggregates these fundamentally different transactions.

Total DAC ODA in 2023 was $223.7 billion according to OECD ODA 2023 Data. Of this, $161 billion was bilateral and $62.7 billion was channelled through multilateral institutions. Approximately 84 per cent took the form of grants and 16 per cent were loans. These are the gross figures. The net picture, adjusting for in-donor costs and for the return flows on loan repayments from previous ODA lending, is more modest.

The nominal growth in ODA from $146.5 billion in 2015 to $223.7 billion in 2023 represents a 53 per cent increase in eight years according to OECD data. Adjusted for inflation in donor-country currencies, the real increase is substantially smaller. When the $31 billion in in-donor refugee costs is stripped out, the actual cross-border concessional resource transfer increases by a more modest margin still. The headline growth in ODA has partially reflected the accounting treatment of migration-related domestic expenditure rather than increased genuine transfer to developing economies.

The five largest DAC donors in 2023 gave the following shares of gross national income: the United States 0.24 per cent, Germany 0.79 per cent, the United Kingdom 0.58 per cent, France 0.50 per cent and Japan 0.44 per cent, according to OECD DAC data. The United Nations target of 0.7 per cent has been a formal commitment since 1970. Fifty-three years after that commitment, only Germany among the five largest economies meets it. The United States, the world’s largest economy, gives less than 0.25 cents of every hundred dollars of national income to development assistance. This is not a new observation. It is, however, a persistent structural fact whose implications compound over time.

While concessional aid flows have grown modestly in nominal terms, the debt obligations of developing economies have grown rapidly. Total external debt of low- and middle-income countries reached $9.0 trillion in 2023 according to the World Bank’s International Debt Report 2024. External debt service payments by these same countries were $443.5 billion in 2022 and $490 billion in 2023. Sub-Saharan Africa received $39.5 billion in ODA. The continent paid out multiples of that figure in debt service. The direction of the net financial transfer, from south to north, is more consequential to fiscal space than any annual variation in aid volumes.

The IMF’s 2024 List of Low-Income Country Debt Sustainability Analyses identified 28 low-income countries at high risk of debt distress and 11 already in debt distress. Together, 39 countries, representing more than half of the world’s most vulnerable economies, face financing conditions that are either at or beyond sustainable limits. This is not a marginal problem. It is a systemic condition that has become the defining constraint on development finance in the lowest-income parts of the world.

The G20 Common Framework for Debt Treatments was established in 2020 to provide a multilateral mechanism for debt restructuring for the most distressed countries. Progress has been slower than the scale of the problem warrants. Zambia reached an agreement in June 2023 and March 2024 under which $6.3 billion in debt was rescheduled over twenty years with a three-year grace period, according to IMF and Lazard data. Ghana and other countries have pursued parallel processes. The timelines involved, typically multiple years from application to agreement, mean that countries in distress spend years in financial uncertainty while waiting for restructuring frameworks that are still being negotiated between creditors with conflicting interests.

The narrative of Chinese finance as an unlimited alternative to Western concessional aid has encountered the limits of reality. Chinese overseas lending, channelled primarily through the Belt and Road Initiative and the policy banks of the China Development Bank and Export-Import Bank of China, reached a peak annual volume of approximately $117 billion in 2016 according to AidData and Boston University Global Development Policy Center data. By 2023, annual BRI lending had fallen to below $30 billion, a decline of more than 74 per cent from peak. Cumulatively, China committed approximately $1.34 trillion in overseas lending between 2000 and 2021 according to AidData’s Belt and Road Reboot report, establishing it as the largest bilateral creditor to developing economies over that period.

The retreat from peak BRI lending reflects several concurrent pressures. A growing number of BRI recipient countries entered debt distress, creating non-performing loan problems for Chinese policy banks that had structured their lending against commodity collateral and project cash flows. Zambia, Sri Lanka, Pakistan and others that borrowed heavily in the BRI era subsequently required debt restructuring that exposed the fragility of Chinese lending models outside its immediate neighbourhood. China’s own domestic economic constraints, including property sector stress and fiscal pressures on local governments, reduced the appetite and capacity for large overseas concessional or near-concessional disbursements. The BRI has not ended. It has contracted significantly.

The interest rate comparison is instructive. Chinese BRI infrastructure loans carried average rates of approximately 5.0 per cent according to AidData and World Bank estimates for 2023. World Bank IBRD lending, which is itself not concessional, was priced at roughly 5.5 to 6.0 per cent on a SOFR-plus-spread basis. World Bank IDA concessional lending, available only to the poorest countries, carried rates of 0.75 to 2.0 per cent. Chinese BRI finance was therefore neither as concessional as IDA nor clearly more expensive than commercial alternatives for middle-income borrowers. Its primary differentiation was speed, minimal conditionality and willingness to finance infrastructure that Western multilaterals would not, particularly in extractive sectors and countries with governance deficits.

The New Development Bank, established by Brazil, Russia, India, China and South Africa as part of the 2014 BRICS Summit agreement, had approved $34.8 billion in financing since its inception by 2024 according to its Annual Report. The Asian Infrastructure Investment Bank, launched in 2016 with 57 founding members and now encompassing over 100, had approved $53.7 billion in financing by the same date according to its Annual Report. Both institutions represent genuine additions to the development finance architecture. Both also operate at a scale that is substantial in institutional terms but modest relative to the financing requirements they were partly established to address.

Neither institution challenges the fundamental constraint facing development finance: the gap between available concessional capital and the infrastructure, climate and social investment requirements of low- and middle-income economies. The AIIB operates largely on near-market terms and has concentrated its lending in infrastructure and clean energy. The NDB has faced governance complications associated with Russia’s membership since 2022 and the consequent difficulties in accessing international capital markets. Both banks are structurally dependent on their own ability to borrow in international markets, which connects their lending capacity to the same global interest rate environment that has tightened financing conditions for their intended borrowers.

Advanced economies committed in 2009 at Copenhagen to mobilise $100 billion annually in climate finance for developing countries by 2020. That target was reported as met in 2022, when OECD data recorded $115.9 billion in climate finance provided and mobilised according to its Climate Finance report published in 2024. The reporting has attracted sustained criticism on two grounds. First, a substantial share of the $115.9 billion consists of loans rather than grants, meaning that it adds to rather than reduces the debt obligations of recipient countries. Second, the mobilisation methodology includes private finance that is “mobilised” by public finance through guarantees and co-investments, creating attribution questions about what the public commitment actually caused.

The replacement target is larger. At COP29 in Baku in November 2024, parties agreed a new collective quantified goal of $300 billion per year by 2035 according to the UNFCCC COP29 Baku Declaration. The gap between the 2022 reported figure of $115.9 billion and the 2035 target of $300 billion is substantial. Whether the trajectory of mobilisation can bridge it without fundamental changes in the grant-loan composition of climate finance, and without additionality requirements that prevent donors from counting existing ODA as climate finance, remains an open question. The Loss and Damage Fund, established at COP28 in Dubai in 2023 to support countries suffering irreversible climate harm, had received approximately $792 million in pledges by 2024 according to UNFCCC data. Against estimates of loss and damage running to hundreds of billions annually, this initial capitalisation is indicative rather than operational.

The most consequential development finance statistic that rarely appears in ODA debates is the remittance figure. Low- and middle-income countries received $669 billion in remittances in 2023 according to the World Bank’s Migration and Development Brief 40. This is approximately three times total ODA. Remittances are private transfers from migrant workers to their families in origin countries, carrying no conditionality, no in-donor administrative costs, no governance requirements and no repayment obligations. They go directly to households rather than to government budgets, which means their development impact is distributed differently from aid, but they represent by far the largest external financial resource transfer to developing economies and have proven more stable than either ODA or private capital flows through economic cycles.

Foreign direct investment in Sub-Saharan Africa reached $38.6 billion in 2023 according to UNCTAD’s World Investment Report 2024. This is broadly comparable to the $39.5 billion in ODA that the region received. FDI brings capital, technology and management practice, but it also generates repatriation of profits that reduces its net contribution to domestic capital formation. The comparison between ODA and FDI at roughly equivalent levels for SSA illustrates that the development finance landscape is genuinely multipolar, even before accounting for South-South finance and the NDB and AIIB.

Any assessment of development finance that concentrates on inflows without examining outflows is structurally incomplete. Africa loses approximately $88.6 billion annually in illicit financial flows according to UNCTAD’s Economic Development in Africa Report 2024. This estimate, which reflects trade misinvoicing, profit shifting, tax evasion and other mechanisms by which capital exits developing economies outside formal channels, exceeds the total ODA that Sub-Saharan Africa receives by more than double. The capital that developing economies need for public investment is, in substantial part, leaving those economies through channels that existing governance frameworks have not succeeded in closing.

The illicit flows figure has important implications for the domestic revenue mobilisation agenda. Sub-Saharan Africa’s average tax-to-GDP ratio is 15.6 per cent according to Revenue Statistics in Africa 2023, published jointly by the IMF, OECD and African Tax Administration Forum, using 2021 data as the latest comparative figure. The OECD average is 34.1 per cent according to OECD Revenue Statistics 2023. The gap between 15.6 per cent and 34.1 per cent is not primarily explained by low formal economic activity. It reflects tax exemption regimes, limited tax administration capacity, transfer pricing manipulation by multinationals and the illicit outflows that the UNCTAD estimate captures. Closing a fraction of the illicit flow gap would generate more fiscal resource than any realistic increase in ODA.

Donor fiscal constraints are real but unevenly distributed. The United States carries federal debt equivalent to 99 per cent of GDP according to the Congressional Budget Office’s Budget and Economic Outlook 2024. This creates genuine pressure on discretionary spending, including development assistance. The United States gave 0.24 per cent of GNI in ODA in 2023, a figure that reflects both fiscal constraint and political choices that long predate recent debt accumulation. The United Kingdom enacted a reduction in its ODA commitment from 0.7 to 0.5 per cent of GNI in 2021, saving approximately £4 to £5 billion annually according to FCDO data, before partially restoring the figure. These decisions reflect assessments of domestic political priority as much as fiscal necessity.

The structural competition for budget resources in donor countries has intensified. Defence spending in NATO member states has increased substantially following Russia’s full-scale invasion of Ukraine, competing directly with development assistance in annual budget processes. The energy transition is imposing infrastructure costs on advanced economies simultaneously with the foreign policy commitment to support clean energy transitions in developing ones. Demographic ageing in OECD countries increases mandatory social expenditure. Against these pressures, the political constituency for development assistance, which delivers diffuse global benefits rather than concentrated domestic ones, faces structural disadvantage in budget allocation. The constraints are real. So is the gap between the 0.7 per cent target and the 0.24 to 0.58 per cent range in which most major donors operate.

India’s trajectory illustrates the developmental trajectory that development finance is designed to support. India graduated from World Bank IDA concessional lending eligibility in 2014, following China’s graduation in 2010. Both transitions marked the movement of major developing economies from net recipients of concessional finance toward reliance on market-rate multilateral lending and domestic capital markets. South Korea, which graduated from IDA in the late 1970s and subsequently joined the OECD DAC as a donor in 2010, represents the complete version of this transition: from aid recipient to aid provider within the span of a generation.

India has extended approximately $32.6 billion in Lines of Credit to other developing countries as part of its own South-South cooperation programme, according to the Ministry of External Affairs Annual Report 2024. These credit lines, managed through Exim Bank of India, finance infrastructure, energy and manufacturing projects in partner countries, primarily in South Asia, Africa and Southeast Asia. India now simultaneously receives development financing from multilateral institutions in specific sectors and provides concessional or near-concessional finance to other developing economies. This position, as a middle-power provider in a multipolar development finance architecture, is the destination that many emerging economies seek. Getting there requires the domestic fiscal depth and institutional capacity that India built, imperfectly and gradually, over decades.

The structural argument of this dossier leads to a conclusion that has been stated many times and implemented incompletely: sustainable development requires fiscal systems that generate sufficient domestic revenue to finance public goods without dependence on external concessional flows that are structurally limited, politically contingent and declining in relative importance. Sub-Saharan Africa’s tax-to-GDP ratio of 15.6 per cent leaves governments with limited fiscal capacity to finance health, education, infrastructure and climate adaptation from their own resources. The gap to the OECD average of 34.1 per cent is not a target to be reached immediately, but the direction of travel is not optional. Development that depends on external finance at current or declining levels cannot deliver the infrastructure and service provision that rapidly urbanising, demographically young populations require.

Closing the domestic revenue gap is complicated by several factors that aid donors and development finance institutions can address but not resolve from outside. Tax exemption regimes negotiated with foreign investors sacrifice revenue for investment attraction. Transfer pricing by multinationals shifts profits to low-tax jurisdictions. Customs and VAT compliance gaps in informal economies reduce the yield from existing tax structures. Digital tax administration has shown promising results in specific contexts, with Kenya, Rwanda and Ghana demonstrating measurable improvements in revenue collection through technology-enabled systems, though the gains require time to compound and administrative capacity to sustain. The $88.6 billion in annual illicit outflows is the most important unrealised fiscal resource: not money that must be newly generated, but money that currently leaves before it can be taxed.