Nigeria’s higher state spending shifts the test to services that work

Original research record

We connect Nigeria’s new state-spending breakdown with the earlier warning that stronger national finances had not fully reached households. The next useful test is local delivery: which projects work, who can use them and whether running costs are funded. A lower education share does not, by itself, mean education spending fell.

What’s changing? The World Bank’s 8 October release reports that state revenue and spending rose strongly in real terms during 2023–2025, while capital spending took a larger share. It also says spending on education, health and social protection rose, although more slowly than economic infrastructure. This is a change in relative priorities, not evidence of an education cut. [New breakdown](https://www.worldbank.org/en/news/press-release/2026/10/08/higher-revenues-better-lives-strengthening-state-spending-for-growth-jobs-services) The page gives only a date. We cannot confirm that it was first published within the last 24 hours.

The Bank’s April presentation had already identified a shift towards infrastructure and warned that household benefits still lagged stronger finances. The October release adds detail to that earlier picture; we do not call the whole trend newly discovered. [Earlier assessment](https://thedocs.worldbank.org/en/doc/1239f54ed8f7334e68cc40639b37cb87-0360012026/original/nigeria-ndu-april-2026-presentation-final.pdf)

Our inference is that revenue growth now needs to be matched with checks on completed work and ongoing services. A new building and a funded service are different stages.

Key points The reported real revenue and spending gains cover 2023–2025. The education-share comparison covers 2021–2025. We do not multiply those figures to invent an education-spending change. We would follow project completion, access, staffing and maintenance. Spending totals cannot tell us whether a school, clinic or road is functioning well. An aggregate can hide large state differences. State-level comparisons need the same reporting period, population basis and coverage.

What could change this view? Infrastructure can improve access to schools, clinics and markets, so a greater capital share can support social outcomes. The signal would weaken if well-run projects and reliable services already account for the spending shift. It would strengthen if completion delays, unpaid operating costs or poor access persist despite larger budgets. We have not audited projects or inferred wrongdoing.

Bottom line We would judge the revenue gain through specific services and completed work. Better public finances create room to improve living standards, but they are only the first link in that chain. This is a public-service research signal, not a naira, bond or equity forecast.

How likely is this call? · Three frameworks This is a noninvestment signal. Market-direction percentages do not apply. Bayesian · what changed the estimate? The October breakdown adds detail about spending choices to an existing concern about household outcomes. We would change our assessment with comparable state-level evidence of completed, accessible and reliably funded services. No likelihood model or probability of policy success was fitted. de Finetti · are the comparisons fair? World Bank reports can reuse the same official accounts, so they are not independent samples. States also differ in starting conditions and responsibilities. We would compare like-for-like services and periods rather than treat each state as an interchangeable trial. Kolmogorov · do the numbers agree? Market probabilities and investment horizons are N/A. Revenue, expenditure, capital share and education share use different denominators. A falling share can coexist with rising real spending. We keep financial inputs, physical completion and service outcomes separate.

Evidence behind this view We began with OTR Database Data’s existing energy and public-delivery themes, then checked a new World Bank disclosure and its earlier presentation. A separate official FAAC release was also read. Its inconsistent totals are excluded from our numerical case rather than silently corrected. [FAAC release](https://fmino.gov.ng/fg-states-and-lgcs-share-n2-338-trillion-from-n3-685-trillion-gross-revenue-for-august-2026/)

What past price moves show No market-price history is needed. The new release says real state revenues rose about 93% and expenditure about 92% during 2023–2025; capital spending’s share rose from 46% to 61%. Education’s share fell from 14.9% in 2021 to 12.1% in 2025. These are different comparisons. They support a shift in the mix, not a calculated fall in education funding. [World Bank](https://www.worldbank.org/en/news/press-release/2026/10/08/higher-revenues-better-lives-strengthening-state-spending-for-growth-jobs-services)

Evidence against this view The earlier state comparison excluded Delta, Rivers and Sokoto because complete quarterly reports were unavailable. We cannot carry that sample forward as if it automatically describes the October analysis. The new full report’s state-level tables were not obtained. More spending can reflect useful repairs to earlier gaps; benefits may also appear after construction is finished.

Why the direction changed No earlier asset call is reversed. The new focus is the chain from national revenue to state priorities to delivered services. The earlier OTR oil-importer and energy-cost cards do not establish this Nigeria-specific spending result. The 367-card Signals set had no Nigeria or Nigerian match, so this is new coverage in that checked set.

What people say vs the facts We read a bounded set of institutional statements, not a household-opinion survey. Even an official page needs checking: the September FAAC release gives ₦2.58 trillion in its opening but ₦2.338 trillion in its title and final total; its VAT balance also differs between paragraphs. We do not choose whichever figure fits our story. [Conflicting source](https://fmino.gov.ng/fg-states-and-lgcs-share-n2-338-trillion-from-n3-685-trillion-gross-revenue-for-august-2026/)

How past forecasts performed The World Bank’s April outlook was about 4.2% growth over 2026–2028; its October release projects an average 4.4%. That is an updated external outlook, not a realised result or an OTR HIT. We found no earlier OTR forecast on this specific state-delivery connection in the reviewed corpus. [April outlook](https://thedocs.worldbank.org/en/doc/1239f54ed8f7334e68cc40639b37cb87-0360012026/original/nigeria-ndu-april-2026-presentation-final.pdf) [October outlook](https://www.worldbank.org/en/news/press-release/2026/10/08/higher-revenues-better-lives-strengthening-state-spending-for-growth-jobs-services)