Aid Cuts Are Also Institutional Exits

Why responsible withdrawal is now a core test of development effectiveness

AI-generated illustration of an institutional handover meeting; not a photograph of an actual IREGS assignment or documented event.

A development programme can close on time, reconcile its accounts and satisfy contractual requirements while leaving the institutions around it less able to function. That distinction between completing an award and preserving its development value is becoming urgent as aid budgets contract.

OECD preliminary data indicate that net official development assistance fell by 23.3 per cent in 2025, the largest annual contraction on record. The OECD projects a further 6.9 per cent fall in 2026, to USD 152 billion, which would take aid to its lowest real level since 2014. The decline is uneven: bilateral assistance to least developed countries and sub-Saharan Africa is projected to fall again, while humanitarian assistance and support to government and civil society face substantial two-year reductions.[1]

Much of the policy debate understandably asks where scarce funding should be concentrated. That is necessary, but incomplete. A portfolio cut is not implemented at the level of a budget chart. It is implemented through suspended contracts, terminated grants, reduced country teams, cancelled technical assistance, closed monitoring systems and local organizations suddenly expected to carry obligations that were designed around external finance.

The central institutional risk is therefore not contraction alone. It is unmanaged exit. When funders treat withdrawal as an administrative closeout rather than a transfer of functions, assets, knowledge and risk, they can destroy part of the value they are trying to preserve. Responsible exit should now be treated as a core discipline of development effectiveness.

The contract ends before the system is ready

Every programme contains two kinds of architecture. The visible architecture is contractual: budgets, outputs, delivery schedules and reporting requirements. The less visible architecture is institutional: who convenes actors, pays recurring costs, maintains data, supervises staff, operates equipment, holds relationships and absorbs risk when plans fail.

Abrupt reductions usually act first on the contractual architecture. The institutional architecture is expected to adjust afterwards.

Recent U.S. experience shows the consequences at unusual scale. A September 2026 review by the U.S. Government Accountability Office found that approximately 6,780 of 7,510 USAID awards were terminated between January 2025 and March 2026. It also found that the Department of State lacked documented guidance for consistent foreign-assistance decisions and lacked complete, accurate data on awards and staffing.[2] The case is exceptional in speed and magnitude, but it exposes a general problem: selection decisions made without usable portfolio information are unlikely to produce orderly transitions.

Closeout compliance does not solve this problem. In Egypt, USAID's Office of Inspector General found that implementers generally followed asset-disposition plans after award terminations. Yet beneficiaries could not use about USD 2 million in assets as intended because the training required to operate them had been cancelled with the awards.[3] The inventory could be reconciled while the development value was still lost.

This distinction matters for value for money. A donor can recover assets, settle invoices and close an award cleanly while leaving an institution unable to perform the function for which those assets were purchased. Administrative completion and sustainable transition are different results.

Risk does not disappear; it moves downward

When funding stops, risk is often described as reduced donor exposure. In practice, much of it is transferred to organizations with less liquidity and political leverage. National NGOs may inherit staff liabilities, leases, safeguarding commitments, community expectations and records that must still be protected. Public agencies may be left with equipment but no operating budget, or with a policy mandate but no personnel to implement it. International intermediaries may protect their core systems by compressing subgrants first.

This is a political-economy problem, not merely a technical one. The actor that controls the funding timetable can externalize the costs of adjustment. Partners that depend on the relationship have limited ability to contest the timing or terms of exit, even where they hold the operational knowledge needed for a viable handover.

That dynamic cuts against localization. A system cannot credibly ask local actors to lead while using them as financial shock absorbers. Localization during contraction requires more than shifting delivery responsibility. It requires authority over transition choices, adequate recovery of closeout costs, access to programme data and a negotiated decision about which functions should continue, change or end.

The risk is greatest in fragile settings, where substitution is least plausible. OECD projections suggest that multiple major providers and multilateral channels are contracting at the same time. The OECD estimates that core funding to UN entities fell by 27 per cent in 2025 and projects further multilateral reductions in 2026.[1] In such conditions, “another donor will take over” is not a transition plan.

Evidence and accountability are vulnerable at precisely the wrong moment

Budget pressure also changes what institutions can know. Monitoring visits, evaluations, partner assurance and learning staff are easy to classify as overhead when services are being protected. Yet reducing those functions during a portfolio shock can leave funders unable to identify which closures create the greatest harm, whether mitigation is working, or what liabilities remain.

The problem is particularly serious because institutional capacity is already a common constraint. The World Bank's Independent Evaluation Group found institutional-capacity and financial-management challenges in 75 and 76 per cent respectively of the closed operations in which those factors were assessed. It also found these challenges more prevalent in fragile and conflict-affected situations.[4] Removing accompaniment, coordination or oversight before responsibilities are absorbed can turn a known implementation weakness into a transition failure.

Independent scrutiny of UK aid illustrates the wider pattern. The Independent Commission for Aid Impact has repeatedly linked budget volatility to disrupted programming, weaker forward planning and damaged partner trust. Its 2025 review of UK aid to Sudan found that in-year reductions postponed activities, closed a water, sanitation and hygiene programme early, and delayed support to women-led organizations. Its July 2026 analysis also warned that traditional bilateral programmes were being sharply reduced in long-standing partner countries, even as three-year allocations improved predictability.[5]

Predictability does not remove the pain of a cut, but it changes what institutions can do about it. Time permits governments to establish budget lines, organizations to retain critical staff, donors to coordinate, and evaluators to preserve evidence about unfinished outcomes. Uncertainty consumes the same period in contingency planning and repeated redesign.

Not every programme should be preserved

The case for responsible exit should not become an argument for institutional self-preservation. Fiscal constraints are real. Portfolios accumulate weak programmes, duplicative structures and activities whose political or strategic rationale has changed. Continuing them because closure is difficult can waste scarce resources and protect incumbent organizations rather than public value.

Nor can all transitions be gradual. Governments may lawfully change policy, fiduciary failures may require suspension, and conflict can make normal handover impossible. In some settings, domestic financing cannot replace external aid on the timetable available.

These constraints strengthen the case for explicit exit governance. They do not weaken it. A disciplined process distinguishes programmes that can end with limited residual risk from functions whose abrupt removal threatens lives, rights, institutional continuity or the integrity of prior investment. It also makes trade-offs visible. Without that discipline, urgency can become a rationale for decisions that are quick but poorly informed.

There are established models to draw from. The Global Fund's 2024 Sustainability, Transition and Co-Financing Policy requires sustainability and, where relevant, transition planning to be reflected in funding requests; it emphasizes country ownership, domestic financing, national systems and preparation before external support ends.[6] These arrangements are not directly transferable to every aid programme, and planned eligibility transitions differ from sudden political cuts. Their value lies in the architecture: transition is treated as a managed outcome, not an afterthought.

What responsible exit requires

A practical standard for responsible exit could have five elements. The following is an analytical proposal drawn from documented transition failures and existing donor practice, not a claim that a universal donor standard already exists.

First, portfolio decisions need a dependency map, not only a performance ranking. Decision-makers should identify which services, coordination roles, datasets, supply chains, staff capabilities and recurrent costs depend on each intervention, and who could realistically assume them.

Second, closure criteria should include fragility, reversibility and substitution risk. A programme with modest short-term results may still maintain a critical institutional function. Conversely, a high-performing project may be unsustainable if it has never transferred authority or recurrent costs.

Third, affected partners should negotiate a time-bound transition compact. As proposed here, such a compact would specify notice, decision rights, liabilities, asset and data transfer, staffing, safeguarding, communications and resources required for closure. Consultation without influence is insufficient.

Fourth, providers should coordinate exits at country and sector level. OECD guidance now explicitly calls on donors to plan exits responsibly and coordinate them. Its 2025 peer review of Luxembourg similarly recommended transparent communication, clear timetables and pragmatic transition plans as the country withdrew from central Sahel partnerships.[1][7] Coordination must test whether replacement finance is real, rather than recording another institution as a theoretical successor.

Finally, evaluation should extend beyond the final disbursement. One proportionate option would be a post-exit review six to eighteen months later, timed to the programme and context rather than treated as a universal standard. It should check whether intended services and coordination functions continued, transferred assets remain operational, recurrent costs have funding, safeguarding and data obligations are met, and financial or operational risks have shifted to less-resourced partners. Interviews with affected partners, basic service and asset-use indicators, and an explicit comparison with the agreed transition plan would provide more insight than a final-delivery checklist. The purpose is not to prolong every grant; it is to learn whether the exit protected or impaired development value.

An older joint donor evaluation reached a conclusion that remains uncomfortably current: early warning was rare, unilateral decisions were often presented as faits accomplis, and stakeholder participation in detailed planning was a good indicator of a more successful exit.[8] The sector has had the diagnosis for years. What has changed is the scale of the exposure.

In an era of contracting aid, institutional quality will be revealed not only by what donors continue to fund, but by how they leave. Responsible exit cannot make a smaller budget sufficient. It can prevent scarcity from being compounded by avoidable loss, transferred risk and institutional amnesia. That is now part of what development effectiveness means.

Article references

1. OECD, ODA projections for 2026 and the near-term: Implications for vulnerable countries and sectors, 19 June 2026. https://doi.org/10.1787/d7c74fa2-en

2. U.S. Government Accountability Office, Foreign Assistance: Agencies Made Broad Changes, but State Lacked Guidance for Decisions, GAO-26-108607, 30 September 2026. https://www.gao.gov/products/gao-26-108607

3. USAID Office of Inspector General, Terminated USAID Awards in Egypt, 6 May 2026. https://oig.usaid.gov/node/8120

4. World Bank Independent Evaluation Group, Results and Performance of the World Bank Group 2024, 2024. https://ieg.worldbankgroup.org/evaluations/results-and-performance-world-bank-group-2024/overview

5. Independent Commission for Aid Impact, UK aid to Sudan, October 2025; and UK aid spending set to fall by around 42%, 16 July 2026. https://icai.independent.gov.uk/wp-content/uploads/ICAI-review-UK-aid-to-Sudan.pdf ; https://icai.independent.gov.uk/uk-aid-spending-set-to-fall-by-around-42-as-major-shifts-in-development-priorities-take-effect-watchdog-finds/

6. The Global Fund, Sustainability, Transition and Co-Financing Policy, Board-approved November 2024. https://resources.theglobalfund.org/media/14383/core_sustainability-transition-cofinancing_policy_en.pdf

7. OECD, Development Co-operation Peer Reviews: Luxembourg 2025, 2025. https://doi.org/10.1787/3713cf73-en

8. Joint donor evaluation, Managing Aid Exit and Transformation: Summary of a Joint Donor Evaluation, 2008. https://sida.se/en/about-sida/publications-archive/managing-aid-exit-and-transformation-summary-of-a-joint-donor ; https://www.oecd.org/content/dam/oecd/en/toolkits/derec/evaluation-reports/derec/sweden/42208151.pdf

‍