Designed to Fail: The SOML PforR Case Study (Nigeria P146583)

MDB Reform Platform  ·  Policy and Accountability Analysis

Designed to Fail: Governance Failure, Institutional Impunity, and Accountability in the World Bank’s Saving One Million Lives PforR (Nigeria P146583)

MDB Reform Platform  |  mdbreform.com  |  April 2026

Keywords: Program-for-Results · PforR · Nigeria · Saving One Million Lives · P146583 · Fiduciary Systems · IEG · Accountability · World Bank Board · IDA · Sovereign Guarantee · Institutional Impunity · MDB Reform

Section 1The Programme and the Record

Nigeria contributes 14 percent of global maternal deaths and 13 percent of global under-five deaths annually. The Saving One Million Lives (SOML) initiative — launched by the Federal Government of Nigeria in October 2012 — was designed to reduce avoidable mortality through primary health care delivery at scale. The World Bank supported it through a $500 million IDA PforR credit: the Bank’s largest PforR at the time of approval. The programme disbursed funds to states based on measured improvements in six health coverage indicators across 36 states and the Federal Capital Territory.

The results, in the documents the Bank itself published, are unambiguous.

The IEG review rated PDO outcomes Moderately Unsatisfactory and efficiency Negligible. The ICR, authored by the Bank team, rated the programme Moderately Satisfactory. The divergence between the two documents is not analytical. It is institutional.

Section 2Designed to Fail: What Was Known at Approval

SOML’s most striking feature is not that it failed — it is that the Bank’s own appraisal documents predicted the failure in detail. The Integrated Fiduciary Systems Assessment (IFSA), March 2015, documented the following before the Board voted:

The Bank knew. It approved a $500 million PforR of the same ministry twelve months later, and concluded in the same IFSA that SOML was “a perfect candidate for financing under the Bank’s PforR instrument.”

The Technical Assessment noted LGA accountability was “extremely opaque and lacked any systematic reporting, let alone being subject to effective accountability against budgets.” Budget execution variance in the Bank’s own reference states ran from 54 to 128 percent. The Programme Action Plan committed to establishing an independent internal auditor and a dedicated procurement officer. Neither was ever appointed: the FY2018 audit records explicitly that “the programme had no standing Procurement Officer and Internal Auditor.”

Most consequentially, the IFSA deliberately excluded state-level audit from the programme boundary (Para 30): “the audit reports of the States will not be required due to the definition adopted on Program boundaries.” This decision meant that ₦16.2 billion (~$52.9M at ₦306.75/$) in state performance grants — the programme’s primary accountability instrument — appeared on the balance sheet as a single unaudited line: “Others — Performance Grant to States: Non-Current Assets.” Eighty-three percent of the first year’s disbursements were never independently examined. The Auditor-General for the Federation, rated D+ by the 2019 Nigeria PEFA assessment, issued clean unqualified opinions on statements with this gap for five consecutive years.

Section 3Zero Accountability: The Institutional Chain of Non-Consequence

When the bonus payments and procurement allegations could no longer be managed internally, the Bank’s December 2019 Restructuring Paper acknowledged “payment of performance bonus to high level Federal Ministry of Health officials and allegations of fraud in the use of program funds and procurement processes.” The resolution: a new Programme Implementation Manual deemed “satisfactory to the Bank,” and an upgrade of the Implementation Progress rating from Moderately Unsatisfactory back to Moderately Satisfactory. No INT referral. No independent investigation. No debarment. The revised PIM — the instrument used to resolve acknowledged fraud — was never disclosed publicly.

There is a structural reason for this. The project team has no institutional incentive to refer cases to INT. A referral opens an investigation that may implicate the team’s own supervision record. It delays ratings upgrades. It creates a public record of management failure. A PIM revision and a rating upgrade achieve the same institutional objective — programme continuation — without any of those costs. The result is a systematic reduction of the funnel to INT: fraud that passes through poorly designed fiduciary assessments, is concealed in state-level audit exclusions, and is resolved through internal management instruments never reaches the integrity system at all.

SOML is Raballand’s analysis made concrete. The TTL who designed the programme and supervised it for six years to a Moderately Satisfactory ICR rating is credited with managing the world’s largest PforR. The Programme Action Plan commitments — independent internal audit, procurement officer, quarterly budget reporting — were never implemented, were documented as “implementation challenges” in the ICR, and generated no formal consequence for anyone in the management chain.

The IEG data confirms this at portfolio level. Of 92 unique Nigeria operations evaluated since the 1970s, 40 percent were rated below Satisfactory — the minimum acceptable threshold. In the FY2015–2024 cohort alone, 10 of 39 operations ($1.99 billion committed) fell below Satisfactory, and in 8 of those 39 ($1.49 billion) the Bank’s own performance was independently assessed as inadequate. The Water GP returned to the same failed programme three times across fifteen years — NUWSRP Phase 2 (MU), Phase 3 (MU), and SURWASH already MU in active supervision. No documented consequence for any team involved in any of these outcomes.

Section 4The Structural Architecture of Failure

4.1 The Approval Machine: Fiduciary Teams as Co-opted Participants

The institutional pathway of a PforR from concept to Board approval creates a specific dynamic that produces the SOML outcome reliably. Once the Country Management Unit allocates funds and the Global Practice takes technical ownership, the programme’s approval is institutionally predetermined. The fiduciary and governance specialists who join the project team face a precise career dilemma: concluding that a $500 million programme should not proceed — because, for example, the implementing ministry has a documented fraud finding, thirty states have no fiduciary assessment, and the auditor is D+-rated — requires an escalation that is structurally unusual and professionally costly.

The term used inside the Bank for specialists who raise concerns that impede approval is “roadblock.” The IFSA’s conclusion that SOML was “a perfect candidate for financing under the Bank’s PforR instrument” is not dishonest in this institutional context. It is the output of a system that rewards completion and penalises obstruction. By the time the PAD reaches OPCS review, the government has been told the programme is proceeding, the IDA allocation is committed, and the Regional VP has endorsed. Stopping it would require escalation across multiple layers of institutional self-certification. No one escalates.

4.2 The ICR as Managed Self-Evaluation

When a programme closes poorly, the institutional energy of the GP shifts from programme improvement to ICR management. The Implementation Completion Report is authored by the team — typically the TTL — whose career it benefits to produce a Moderately Satisfactory rating. The techniques are professional, not dishonest: baseline selection, regional disaggregation to surface positive sub-group stories, framing that converts “$0 of $20M Innovation Fund disbursed” into “implementation challenges,” and narrative structures that comply with ICR requirements while supporting an MS conclusion.

The IEG ICRR provides an independent check, and in the SOML case partially fulfilled this function — finding PDO outcomes MU and efficiency Negligible. But the ICRR’s findings sit in a separate report. For portfolio databases, country strategy papers, Board briefings, and future project appraisals, SOML is an MS operation. The rating that propagates through the system is the managed one, not the honest one.

4.3 The Board as Co-Conspirator

The World Bank’s Board of Executive Directors approves approximately 300 operations per year as a resident body. The 2009 Zedillo Commission named the problem: “The Board attempts an impossible trinity of roles: political representation of member state interests, technical banking supervision, and management oversight. No board in the world can perform all three simultaneously.” The co-optation consequence, documented by Brar (2026) across 8,764 operations in the IEG database, is categorical: the Board approved all operations. It held management formally accountable for the development results of none of them.

SOML illustrates the mechanism. The Board approved in April 2015 — receiving an IFSA that documented prior fraud, absent internal audit, and the exclusion of state-level audit. Having approved, the Board could not scrutinise the December 2019 fraud acknowledgment without implicating its own prior endorsement. The institutional response was the one the co-optation argument predicts: note the restructuring paper, accept the rating upgrade, continue. The Board is not a passive victim of this dynamic. It approved an accountability architecture that guaranteed its own distance from ground-level failure — and approved it again, implicitly, each time it received an ISR rating of Moderately Satisfactory while the programme deteriorated.

4.4 The Sovereign Guarantee: The Source of Impunity

Nigeria is a blend country. The $387.6 million it drew down under SOML will be repaid in full across 38 years at IDA concessional rates, regardless of whether a single additional child survived its fifth birthday because of the programme. The Bank earns its spread on every dollar repaid. This is not incidental to the accountability failure. It is its foundation.

In a commercial lending relationship, the lender bears financial consequences when its own design failures contribute to poor outcomes. The World Bank bears none. A poorly designed $500M operation, a well-designed $500M operation — the financial return to the institution is identical. The sovereign guarantee insulates the Bank from every consequence of its own institutional failures, creating the condition in which Raballand’s finding — no correlation between project quality and TTL career trajectory — is a rational equilibrium, not a management failure.

Section 5What Must Change: Eleven Specific Reforms

5.1 Design and Fiduciary Quality

5.2 Audit and Evaluation

5.3 Board Governance

5.4 Financial Accountability

Section 6Conclusion

The Saving One Million Lives Programme for Results was designed to fail. The IFSA certified as “a perfect candidate” a programme whose implementing ministry had been implicated in procurement fraud twelve months earlier. The Programme Action Plan committed to mitigations — internal audit, procurement officer, quarterly budget reporting — that were never implemented. The audit covered $1.9 million of PMU expenditure and excluded $53 million in state performance payments from its scope. Performance bonuses to senior ministry officials were in the audited accounts six months before the Bank publicly acknowledged them. When acknowledged, they were resolved with a PIM revision. The rating remained Moderately Satisfactory. The PIMs are lost.

This outcome was not produced by bad luck or difficult conditions. It was produced by an institution whose internal reward structure compensates approval and penalises obstruction; whose ICR process allows the GP to manage the permanent record of its own work; whose Board is co-opted into every approval and therefore cannot hold management accountable for results; and whose sovereign guarantee insulates it from every financial consequence of its own institutional failures.

Raballand documented the incentive structure in 2015. Brar documented the portfolio consequences in 2026. The IEG has documented the outcome pattern annually for a decade. The SOML case documents all of it at project level, with source citations from the Bank’s own public record.

Nigeria’s children and mothers paid for this programme. They will continue paying, through sovereign debt service, for 38 years. That debt should carry a record of what it failed to achieve — and a commitment from those who govern the institution that accountability for institutional failure will not remain, as it has for twenty-five years, the one reform that the institution consistently declines to make.

MDB Reform Platform  |  mdbreform.com  |  Licensed CC-BY 4.0  |  Primary sources: World Bank Documents & Reports portal and IEG dataset. All citations verifiable against public record. Contact: fcvstrategy@gmail.com