The Missing Infrastructure of Development Finance
Aid, institutions, and the overlooked architecture of sovereign finance
Executive Summary
The debate over UK development assistance has become dominated by a single question: how much should be spent. This essay argues that an equally important question has received far less attention: how aid should be spent if the objective is to help countries achieve more durable access to international finance.
Development partners have invested heavily in strengthening sovereign institutions that produce credit-relevant evidence. Debt management offices, public financial management systems and macroeconomic statistical capacity have all received sustained support, supported by a substantial body of evidence showing improvements in debt sustainability, investment efficiency and, in some cases, lower borrowing costs. These programmes represent some of the most carefully evaluated forms of long-term capacity building in the development finance portfolio.
The essay argues that this investment chain remains incomplete. While considerable resources have been devoted to improving the production of sovereign information, comparatively little attention has been paid to the institutional infrastructure needed to coordinate, assure and present that evidence in a coherent, comparable and continuously updated form. This is not an argument that development partners should produce sovereign credit assessments or replace existing evaluative institutions. Rather, it is an argument that the evidentiary infrastructure underpinning sovereign credit assessment deserves to be recognised as a legitimate component of development finance in its own right.
The distinction matters because development partners increasingly judge aid by its ability to mobilise private capital, expand fiscal space and reduce long-term dependence on concessional finance. If those objectives are taken seriously, then strengthening the institutional systems through which sovereign evidence becomes usable for investors, lenders and other evaluative actors should be viewed as a logical extension of existing investments in debt management, public financial management and statistical capacity.
The essay concludes that debates about restoring UK aid spending should extend beyond questions of volume. They should also consider whether development finance has overlooked a category of institutional infrastructure capable of helping countries translate stronger domestic capacity into financing conditions that more accurately reflect their underlying fundamentals.
The Missing Infrastructure of Development Finance
The future of UK development assistance has returned to the centre of political debate, ahead of a Burnham Government. Calls for an increase in spending come on the back of substantial cuts. On 19 March 2026, the foreign secretary told the House of Commons which countries would carry the largest share of the burden. Mozambique and Pakistan would see their direct grants reduced ‘significantly,’ while Ukraine, Palestine, Lebanon and Sudan would be protected (BBC News, 2026). The statement was the latest instalment of a retreat that began in February 2025, when the prime minister announced that development assistance would fall from 0.5 per cent of gross national income to 0.3 per cent by 2027/28, with the difference redirected to defence (Loft and Brien, 2026: 7). At 0.3 per cent, UK aid spending would reach its lowest level in over a quarter of a century (Watkins, 2025). The decision prompted the resignation of the development minister, Anneliese Dodds, and a warning from Sarah Champion, chair of the international development committee, that the government was pursuing what she called a false economy (Loft and Brien, 2026: 9).
The parliamentary argument that followed was conducted almost entirely in the currency of volume. Restore the 0.7 per cent target legislated under Gordon Brown, or accept 0.5 per cent as a pandemic-era compromise, or hold at 0.3 per cent until the Office for Budget Responsibility confirms that day-to-day borrowing has ended (Independent Commission for Aid Impact, 2026: 26). The Liberal Democrats called the cut a strategic error that would leave a vacuum for Russia and China to fill. The Conservatives promised a still deeper reduction, to around 0.1 per cent of GNI, roughly seven billion pounds lower again (Loft and Brien, 2026: 9). Labour, having pledged in its manifesto to restore 0.7 per cent ‘as soon as fiscal circumstances allow,’ instead cut further within months of taking office (Lazell and Petrikova, 2025: 333). Every position in that debate treats aid as a tap: a number to be turned up or down according to fiscal room and political appetite. What almost none of it asks is a different question, one about sequence rather than volume. Within whatever sum survives the Treasury’s arithmetic, does the remaining money go toward the parts of the development finance chain that already show a measurable return, and does that chain, however well built, actually terminate in the outcome donors say they want.
Investing in Institutions
That outcome, stated plainly, is that countries which strengthen their institutions should be able to borrow more cheaply, on more reasonable terms, from a wider set of lenders, with less dependence on aid itself. This sits close to the working assumption behind three decades of technical assistance. Glennie and Sumner, reviewing the evidence on when foreign aid succeeds, describe what they call an ‘institutions model’: the proposition that aid works if the right institutions are in place (Glennie and Sumner, 2014: 24). Aid, on this account, functions as a bridge to the point at which a country’s own institutional capacity, rather than continued donor generosity, governs its access to finance. The bridge has a destination. The destination is stronger domestic capacity to secure finance on terms that more accurately reflect the country’s underlying position.
Judged against that destination, the record of institutional investment amounts to one of development finance’s better-documented forms of long-term capacity building. What that record shows, on close inspection, is a consistent pattern: donors build the institutions that produce credit-relevant information with unusual discipline, and pay comparatively little attention to the institutional systems that would coordinate, verify and present that information for use by those responsible for evaluating what it means.
Start with debt management, the most technical and least glamorous layer of the apparatus. The World Bank’s Debt Management Facility, run jointly with the IMF since 2009, has delivered 612 technical assistance missions across 80 countries, with close to 90 per cent of national-level support since 2020 going to low-income and lower-middle-income countries (World Bank, 2026: vi). Within that programme, 129 Debt Management Performance Assessment missions have taken place in 71 eligible countries, several receiving more than one (World Bank, 2026: 18), and 166 medium-term debt strategy or annual borrowing plan missions have been delivered in 60 countries (World Bank, 2026: 21). The assessment tool itself is explicit about its purpose: it exists to evaluate a country’s debt management processes and institutions against sound international practice, identifying strengths and weaknesses so that capacity can be strengthened accordingly (World Bank, 2021: 1). Fifty-four countries have published a medium-term debt strategy at least once since 2020, and the number publishing a debt statistical bulletin within six months, with full coverage, has risen to 43 (World Bank, 2026: vii). This is a sustained, multi-decade programme, with a defined methodology, a shared donor pool, and outcomes that are tracked and published, far beyond what ‘marginal capacity building’ would suggest. Yet a debt management office, however well run, does not itself determine how a lender or a market interprets the figures it produces. It generates a cleaner, more reliable account of what a country owes and how that borrowing is managed. It stops short of the separate institutional process through which a set of debt figures is weighed against precedent, compared with other borrowers, assured, coordinated and presented within a common evidentiary infrastructure that can be utilised consistently by investors, lenders, credit rating agencies, and sovereigns alike.
Public financial management shows the same pattern, and the evidence on its returns is unusually direct for a field this often treated as a compliance exercise. The IMF finds that countries with stronger public investment management institutions produce investment that is more predictable, credible, and productive, and that strengthening those institutions could close up to two-thirds of the public investment efficiency gap that separates the best performers from the rest (IMF, 2015: 4-5). Moving from the lowest to the highest quartile of efficiency could double the growth impact of a given pound or dollar of public investment (IMF, 2015: 18), and countries with stronger institutions carry lower incremental capital-to-output ratios, meaning more growth for the same capital outlay (IMF, 2015: 30). The World Bank’s independent evaluation of its own support in this area draws the connection to borrowing costs explicitly: weak public investment management increases the amount of debt incurred to deliver a given set of projects while reducing the growth that debt generates, raising the debt-to-GDP ratio for no corresponding benefit (Independent Evaluation Group, 2021: 45). Because debt sustainability is a function of the gap between the real interest rate and growth, the report continues, a country's ability to service its borrowing depends materially on whether it captures value for money from the investments that borrowing financed (Independent Evaluation Group, 2021: 46). Sound public financial management functions, in other words, as a precondition for the fiscal sustainability, macroeconomic stability, and public accountability that the World Bank treats as foundational to its own mission (Independent Evaluation Group, 2021: ix), a standing considerably higher than the compliance exercise it is sometimes taken for.
Nothing that follows should be read as diminishing any of this, or as diminishing the parallel case for humanitarian assistance, education, health systems and emergency relief, all of which perform functions that no financial institution could replace and none of which are the subject of this essay. The argument developed here concerns a narrower and more specific question: whether one category of institutional investment, adjacent to all the categories just described and resting on the same underlying logic, has received disproportionately little attention despite serving many of the same long-term objectives that the rest of the development finance portfolio already pursues.
The Missing Layer
A third layer, statistical transparency, extends the same logic from institutional process to the information a country produces about itself. Here the evidence moves from plausible mechanism to measured price effect. Cady and Pellechio, examining subscription to the IMF’s data dissemination standards across 26 emerging and developing economies, find that subscription to the Special Data Dissemination Standard reduced launch spreads on new sovereign bonds by an average of 20 per cent, while participation in the more basic General Data Dissemination System reduced spreads by around 8 per cent for countries with market access (Cady and Pellechio, 2006: 1). A later IMF study reaches a compatible conclusion by a different route: greater transparency through data dissemination reduces uncertainty about economic developments, improves the basis on which prospects are appraised, and can be expected to lower the risk premiums countries face when they enter financial markets (Gonzalez-Garcia, 2022: 3). The same paper finds that the benefit is not evenly distributed. Countries with weaker governance gain the most from adopting these standards, meaning statistical transparency substitutes, in part, for the institutional trust that stronger governance would otherwise supply (Gonzalez-Garcia, 2022: 4). A government that cannot yet demonstrate strong institutions can still, through the simple discipline of publishing reliable numbers on a predictable schedule, buy itself a measurable discount on the cost of borrowing. But subscribing to a data standard is not the same act as having that data interpreted, set against comparators, and translated into a coherent evidence base that lenders and other evaluative institutions can use consistently. The standard disciplines what a country publishes. It does not supply the institution that decides what the publication means.
Laid end to end, this is a chain of investment with an unusually clean line of sight from input to outcome, and a consistent shape to the gap that sits alongside it. Donors fund the technical assistance. Institutions like debt management offices and statistical agencies absorb it. The resulting capacity is scored, published, and in the case of data standards, directly priced by capital markets. At every stage, the institution being financed produces information, records it, or discloses it. At no stage does the same investment extend to the institutional system that would turn fragmented information into a usable, comparable and continuously updated evidence base for sovereign credit assessment.
That asymmetry deserves to be taken seriously as more than an oversight, because the institutional process being neglected has the characteristics of infrastructure in a fairly precise sense, not merely in the loose way the word gets used to describe anything a government builds. It is worth being precise about what is actually missing. Debt statistics produce information. Debt management institutions produce governance. What neither provides, on its own, is the institutional infrastructure through which those outputs can be assembled, verified, and made usable within a coherent sovereign evidence base. That evidence base can then be drawn upon by the diverse institutions responsible for evaluating sovereign creditworthiness. Consider what a functioning system for supporting the formation of a coordinated and assured sovereign evidence base through coordinated, assured, and comparable evidence would actually do. It would be relied upon simultaneously by a wide range of independent actors: bond investors pricing a new issue, commercial banks assessing counterparty risk, multilateral lenders setting concessional terms, and the borrowing government itself, all drawing on the same underlying body of assured evidence rather than each commissioning a private one. It would reduce the cost of acquiring and interpreting information for every one of those actors, in the same way that a single reliable bridge reduces the cost of crossing a river for every subsequent traveller, rather than requiring each one to build a raft. Without a shared evidentiary framework of this kind, each participant bears the cost of independently assembling and interpreting the same underlying sovereign information, duplicating analytical work that a common reference point would only need to perform once. It would standardise the evidence base against which borrowers are assessed, in place of the inconsistent, ad hoc methods that currently prevail, making a Ghanaian bond and a Zambian bond legible against a shared frame of reference rather than two unrelated stories. It would generate network effects: the more borrowers and lenders that rely on a common evidentiary process, the more valuable that process becomes to each additional participant, because comparison and precedent accumulate. It would coordinate expectations across a market that would otherwise price the same country very differently depending on which analyst, which week, or which crisis happened to be salient at the time. And it would persist as a durable institutional layer rather than a one-off transaction, continuing to generate value long after any individual bond issue, technical assistance mission, or rating cycle had concluded, in the same way that a debt management office keeps producing value long after the specific loan it was built to manage has been repaid. Judged against these characteristics, the institutional process through which sovereign evidence is made usable in the formation of evidentiary infrastructure looks less like a downstream consequence of good governance and more like the missing piece of the same infrastructure the rest of this essay has already documented.
This is not simply a question of what sovereigns stand to gain, since development partners have their own reasons to take it seriously. Donors increasingly frame their objective as mobilising private capital rather than replacing it, and every pound of concessional finance is now judged, at least in part, by whether it crowds in additional investment, reduces long-term dependence on aid, and expands the fiscal space a government commands on its own account. If those objectives are taken at face value, the institutions through which sovereign evidence is made usable by those responsible for evaluating risk and producing an evidentiary infrastructure are not a tangential concern sitting outside the development finance mandate. They are part of the architecture through which that mandate is actually achieved, since a reform that never reaches the market in a form the market can price has, from the donor's own stated perspective, achieved only half of what it set out to do.
Completing the Architecture
This is where the UK aid debate returns, not as decoration but as the sharpest illustration available of the asymmetry just described. The politics of the cuts, understandably, have concentrated on what is lost: health systems, humanitarian response in Gaza and Sudan, the £2 billion replenishment of the International Development Association, which one commentary on the cuts notes generates roughly four pounds in lending and grants for every pound the UK contributes (Watkins, 2025). The same commentary records that the International Development Secretary has framed the residual budget around ‘value for money’ (Watkins, 2025). Yet value for money, in the way it is used in this debate, is almost always retrospective. It asks whether past spending achieved its stated objective. It rarely asks the forward-looking question of whether the objective itself, cheaper and more independent sovereign finance, requires a specifically designed framework that nobody in the debt management, public financial management, or statistical transparency programmes above has been asked to build. Lazell and Petrikova, reviewing the coherence of UK aid practice more broadly, note that the merger of development functions into the Foreign, Commonwealth and Development Office left the development arm as the junior partner, with a loss of local knowledge and a diminished capacity for exactly the kind of sustained, technical, multi-year engagement that debt management and statistical capacity building require (Lazell and Petrikova, 2025: 325). A shrinking budget administered by a weakened development arm is not a promising setting in which to notice a gap that was already invisible when the money was more plentiful.
None of this amounts to an argument that the debt management facilities, the public financial management evaluations, or the data transparency initiatives have been a poor use of funds. They rank among the more carefully evaluated, most consistently returned-upon investments in the development finance portfolio, and the case for protecting them, even as the wider aid budget contracts, is stronger for having been measured this precisely. The argument is a different one. Development finance has already accepted, across three decades of sustained technical assistance, the principle that financial institutions constitute development infrastructure every bit as real as a road or a power station: that a debt management office, a statistical agency, and a public investment appraisal unit are worth building because reliable finance is itself a form of public good. Recognising the institutional process that coordinates, assures and communicates sovereign evidence for credit assessment as part of that same infrastructure amounts to the next application of an existing principle, extended to the one part of the chain that has so far been left to markets, agencies and donors to assemble informally, case by case, whenever a bond is due, rather than a departure from that principle.
The Commons debate over the next round of cuts will, in all likelihood, return to the language of restoration: 0.7 per cent, 0.5 per cent, 0.3 per cent, a target legislated, suspended, and now treated as aspirational. Sarah Champion will continue to call the reduction a false economy, and she may well be right about the humanitarian arithmetic. Development finance has, over three decades, become increasingly adept at helping countries produce the evidence of creditworthiness. The next challenge may be recognising that the institutional systems through which sovereign evidence is coordinated, assured, and made usable are themselves part of the infrastructure of development.
References
BBC News, ‘Foreign secretary sets out reduced UK aid allocations to hardest-hit countries’ (BBC News, 19 March 2026)
Cady, J. and Pellechio, A., ‘Sovereign Borrowing Cost and the IMF's Data Standards Initiatives’ (IMF Working Paper WP/06/78, 2006)
Glennie, J. and Sumner, A., ‘The $138.5 Billion Question: When Does Foreign Aid Work (and When Doesn't It)?’ (CGD Policy Paper 49, 2014)
Gonzalez-Garcia, J., ‘Improving Sovereign Financing Conditions Through Data Transparency’ (IMF Working Paper WP/22/230, 2022)
Independent Commission for Aid Impact, Management of the Official Development Assistance Spending Target: A Review (ICAI, 2026)
Independent Evaluation Group, World Bank Support for Public Financial and Debt Management in IDA-Eligible Countries: An Independent Evaluation (World Bank, 2021)
International Monetary Fund, ‘Making Public Investment More Efficient’ (IMF Policy Paper, 2015)
Lazell, M. and Petrikova, I., ‘UK aid is failing: suggestions for an impactful, coherent and globally aware development practice’ (2025) 101(1) International Affairs 321
Loft, P. and Brien, P., ‘UK aid: Reducing spending to 0.3% of GNI by 2027/28’ (House of Commons Library Research Briefing CBP-10243, 2026)
Watkins, K., ‘UK aid cuts – minimising the harm, planning for recovery’ (ODI Global, Expert Comment, 6 May 2025)
World Bank, Building Debt Management Capacity: Impact, Challenges, and Future Directions – Debt Management Facility Program: 2009–25 (World Bank, 2026)
World Bank, Debt Management Performance Assessment Methodology (2021 Edition, World Bank, 2021)