Who Will Pay for Lost Aid? What Sub-Saharan Africa’s 2026 Budgets Reveal

In February, we assessed whether governments in sub-Saharan Africa had responded to the 2025 US aid cuts by increasing revenue or shifting public spending. Our review of budget statements and policy announcements in 18 aid-exposed countries found that only Sierra Leone and Tanzania proposed new revenue measures to replace lost aid. None announced spending reallocations to protect health programs, though we noted that later budgets might show a stronger response.

The latest budget cycle lets us reassess that prospect. This blog post extends our analysis from US aid cuts to reductions by all donors and expands the sample from 18 to 40 countries. We review official government sources, including budget speeches, finance laws, and ministry publications.

We identify more revenue measures than in February, but few explicitly address declining aid. Estimated gains are generally modest, and many measures appear to reflect routine annual tax changes rather than deliberate attempts to replace lost aid.

Have governments recognized aid decline?

Our sample includes 40 sub-Saharan African countries; South Africa, Mauritius, Zimbabwe, South Sudan, and Eritrea are excluded because of comparability or data availability concerns. We used Claude Opus 5.5 (Anthropic) to help compile information from budget speeches, finance laws, and finance ministry publications and assess governments’ responses to declining external grants this fiscal year. Eight countries budgeted for lower grants in 2026 than in 2025, with sizeable declines in Tanzania (−39 percent), Lesotho (−29 percent), Eswatini (−19 percent), Cameroon (−18 percent), and Côte d’Ivoire (−12 percent). Project grants are budgeted to fall by 27 percent in Senegal and by more than half in the Democratic Republic of the Congo. Togo reports a longer-term decline in project grants.

Some countries, including Malawi, Liberia, Madagascar, and Seychelles, expect higher aid receipts. Malawi’s projected increase assumes a sharp recovery.

One caveat applies throughout: budgeted grants capture only aid flowing through government budgets. Donors and implementing partners also fund substantial assistance directly, particularly in health. Budget figures may therefore understate the full aid shock facing service delivery.

Few taxes specifically target lost aid

Few of the 40 countries introduced sizeable tax packages, although 25 announced mostly small revenue measures alongside their 2026 budgets. Documents for five countries contained no such measures. For the other ten, official revenue documents were unavailable or unreadable, so their responses remain unknown—not necessarily absent.

Eight countries link declining aid to their fiscal plans. Five present new revenue measures as at least partly a response. Côte d’Ivoire, facing a 12 percent drop in grants, is removing VAT exemptions and taxing foreign e-commerce platforms. Malawi cites aid withdrawals and adds new excises, import surcharges, and a rental withholding tax to a November 2025 package that raised VAT to 17.5 percent. Sierra Leone extends its minimum alternate tax and raises withholding taxes. Burkina Faso aims to offset the end of World Bank program grants, partly through a tax on arms imports. The Central African Republic relies on administrative measures. Our February review identified two such cases among 18 countries: Sierra Leone and Tanzania. This time, Tanzania does not explicitly connect its measures to aid.

The other three link lower aid to their fiscal plans but identify no matching tax measures. Togo ties falling project grants to its revenue efforts but itemizes no measures. Lesotho, facing a 29 percent fall in grants, will rely more on loan-financed projects. Eswatini, where grants fall 19 percent, cites the review of US HIV funding but raises only alcohol and tobacco excises, by 3.4 percent.

Elsewhere, new revenue measures appear to be small adjustments unlikely to generate much revenue: modest increases in tobacco, alcohol, and sugary-drink excises; new taxes on digital services and betting; and new or expanded electronic invoicing. Although these could in principle raise meaningful revenue, their expected impact apparently did not warrant a budget estimate. Some governments also anticipate gains from better tax administration. We count these actions as measures when budget documents describe them as revenue-raising. Others are cutting taxes. Cabo Verde and Equatorial Guinea announced tax relief without offsetting revenue measures in the documents reviewed, while Ghana’s VAT relief package is expected to reduce revenue by about 0.4 percent of GDP.

How much revenue will the measures raise?

Only seven governments estimate the revenue impact of their new measures in the documents reviewed. Five project gains of 0.2–0.7 percent of GDP (using nominal GDP from the latest IMF World Economic Outlook). Cameroon expects a gain of 0.7 percent of GDP, partly from stronger VAT enforcement; Tanzania and Kenya, 0.6 percent each; the Central African Republic, 0.5 percent from improved revenue administration, including customs (estimated at the bill stage); and the Gambia, 0.2 percent from four new levies. By contrast, Ghana expects its VAT relief to cost about 0.4 percent of GDP.

Mali’s revised budget increases projected revenue by 1.6 percent of GDP, mainly through higher gold prices and new earmarked mining funds rather than new taxes. Its new tax on betting winnings is expected to raise about 0.25 percent of GDP. Liberia relies on a one-off mining signature bonus, not a costed package of tax measures.

Only Cameroon and Tanzania report both expected revenue from new measures and declining budgeted grants. In both, projected revenue gains exceed budgeted grant losses.

These comparisons warrant caution. Much lost aid was spent outside government budgets, so declining budgeted grants reflect only part of the funding gap. Even when projected revenue gains exceed budgeted grant losses, it remains unclear whether the additional funds will reach programs and services previously supported by donors.

How did spending change?

Fifteen countries announced spending measures, chiefly new programs, pay and benefit changes, subsidies, or decentralization initiatives. Few, if any, explicitly shifted resources to services previously funded by donors.

As in February, we find no budgetary evidence of systematic spending shifts to protect health. This does not mean no reallocations occurred, but governments have yet to demonstrate a consistent budgetary response to shield health services from declining aid.

Did the oil shock limit governments’ responses?

The rise in oil prices after war broke out in the Middle East in February 2026 may have complicated governments’ responses. Higher fuel costs were already straining household budgets, potentially making new taxes harder to introduce. As of late October, only 11 of the 40 sub-Saharan African countries in our sample still had explicit measures to cushion domestic energy costs against higher international oil prices.

Yet the budgets offer little evidence that the oil shock explains the muted response. Most 2026 budgets predated the shock. Six countries with fiscal years starting in July 2026 prepared budgets afterward. Of the four with readable documents, three—Burundi, Kenya, and Tanzania—introduced revenue measures, including the two largest quantified packages in absolute terms.

The shock’s effect is clearer on spending. Sierra Leone’s July 2026 mid-year budget added fuel and electricity subsidies, funded by a 40 percent cut in domestically financed capital spending. This shows how external shocks can increase fiscal pressures and force difficult spending trade-offs.

What do the latest budgets show?

The latest budget cycle shows more revenue measures than our initial review, but most appear to be minor tax changes rather than deliberate responses to declining aid. Few governments explicitly link their measures to aid losses. Quantified revenue gains are generally below 1 percent of GDP, and some depend on administrative improvements that have often taken longer than planned to implement and generate revenue. Some governments have also introduced substantial tax relief, while others expect more aid. Our review highlights the need for budget documents to explain the revenue impact of measures, including administrative changes, more clearly.

The main concern is that budgeted grants reflect only part of the decline in external assistance. Much lost aid was spent outside government budgets, leaving health and other social services with a larger funding gap than budget figures suggest. We find little evidence of systematic spending reallocations to protect these services.