Kofi & the Six Ships — World Bank & IFC · Analytical NotesDay 0–7 · 2025–2026
Kofi & the Six Ships — Series Landing Page
Introduction to the multi-part analytical series examining the World Bank, IMF, and Regional Development Banks through the lens of one fictional African graduate student encountering each institution for the first time. Covers the Six Ships narrative, the Nigeria country series, and the full catalogue of analytical notes.
Series OverviewWho Is Minding the Ship? The Board That Approves Everything Cannot Be Held Accountable for Anything
53% of projects satisfactory at completion. $84.4M Board cost annually. The Board has never formally held management accountable for development outcomes in 82 years. The Zedillo Commission identified why in 2009. Nothing changed.
13,273 IEG-rated operations since 1973. The S+ rate fell from 82% in the 1970s to 26% by 2005–09. A Board that co-approves every loan cannot independently oversee it. The Wappenhans Report (1992), the Zedillo Commission (2009), the AIIB comparison. Three reforms that would change the architecture — none requiring a quota review.
Read the PaperIFC Fragile States: 11% Satisfactory and the Additionality Problem
11% satisfactory in fragile states. The PSW: $2.5 billion to go where the private sector would not — used to replace capital the IFC would have deployed anyway. That is a refund, not additionality.
Croupiers in Washington, tables in Kinshasa. The IFC’s mandate in fragile states is the most important and the least delivered. The additionality claim does not survive the data. Non-PSW IFC commitments in PSW-eligible countries fell in aggregate during IDA18–IDA20.
Read the Paper Analytical NoteKofi & the IMF — Papers & Analytical NotesDay 2 · Five Notes
The Ship That Does Not Keep Score
Zero project ratings in 149 countries. The IEO produces thematic evaluations — valuable, but not a rating of whether the IMF’s work delivers results. You cannot fix what you do not measure.
The IMF’s evaluation architecture compared to the four MDBs. What IEO does. What it does not do. Why the absence of project ratings is not a technicality — it is the accountability gap.
Read the Paper Analytical NoteNigeria and the Africa-Wide Emergency Funding: The $3.4 Billion RFI and the Evaluation That Could Not Name It
$3.4bn disbursed in weeks. No performance conditions. The IEO Sub-Saharan Africa evaluation named 19 countries. Nigeria — the largest recipient — was not among them.
COVID emergency lending, institutional self-protection, and the accountability gap that the largest single-country RFI in Sub-Saharan Africa exposed. The Idris arrest. What the IEO evaluation omitted and why it matters.
Read the Paper Nigeria Note Africa-Wide AnalysisThe Cold War Across 19th Street: The Fiscal Case for Ending IMF–World Bank Mandate Duplication
$750M–$1.1bn per year in wasted capacity. Same Finance Ministers. Same street. Same week. 35 years of concordats. Nothing has structurally changed.
Five domains where both institutions do the same work. Five country cases where contradictory advice cost governments real money. The 35-year timeline of failed coordination. Six actions that would actually change the architecture.
Read the Paper Analytical NoteThe Homework It Does Do — And Why the Score Has Not Moved: Isomorphic Mimicry and AFRITAC
Five regional centres. $200M in technical assistance. Not rated. Schick’s seven basics: payroll, procurement, internal control, cash, basic reporting. SSA PEFA says these are not being done. Two decades of capacity building. The score has not moved.
What AFRITAC teaches and what the PEFA data shows countries still cannot do. The form-function gap in IMF technical assistance. Why the score has not moved despite two decades of capacity building. The isomorphic mimicry argument applied to PFM reform.
Read the Paper Analytical NoteEurope’s Chair: Who Sits at the Table, Who Decides, and What It Would Take to Change It
The IMF Managing Director has always been European. The World Bank President has always been American. Both are customs, not rules. A custom you can only change by deciding to be embarrassed by it.
The governance arrangement that has persisted since 1944. What it costs in legitimacy. What changed at the WTO in 2021. What the Spring Meetings could look like if the shareholders chose to act.
Read the Paper Analytical NoteKofi & the Regional Development Banks — Papers & NotesDays 4–6 · AfDB, ADB, IDB
AfDB Results: The 94% Problem — Five Named Cases and a Validation Architecture That Cannot Correct Itself
94% satisfactory, every year. Adjusted plausible range: 61–75%. IDEV validates a sample and produces no independent rating. The gap is unpublished.
The task manager who wrote the rating was wrong, and kept it. Medupi. The smile that does not change. Five cases where the documented evidence and the official number tell different stories.
Read the Paper Analytical NoteADB Results: The Best Evaluation Architecture in Development Banking — and a 12-Point Gap That Is Not Closing
IED validates 100% of PCRs. Publishes the management-IED gap. Signs its name to the disagreement. The gap: 12 points sovereign. Not closing. Nine Japanese presidents in fifty-nine years. A rule you can change at a meeting. A custom you can only change by deciding to be embarrassed by it.
The ADB evaluation architecture compared to the World Bank and the AfDB. What transparent disagreement between management and independent evaluation looks like in practice. The governance question: same principle applies to the IMF Managing Director and the World Bank President.
Read the Paper Analytical NoteIDB: Evaluation Architecture and the Rating Gap — Sixteen Years of the DEF and a 28-Point Management-OVE Divergence
Management: 81% satisfactory. OVE: 53%. Same projects. Same year. A decade. Capital decisions made on the management figure.
The most comprehensive self-evaluation history in development banking. The only institution to evaluate its own evaluation framework — and find the cultural change did not occur. Four named cases. Six failure patterns. The DEF as a model: what it got right and why it did not move the number.
Read the Paper Analytical NoteKofi in Nigeria — Country SeriesDays 1–10 · Benin City, Lagos, Abuja
Kofi in Benin City — What the World Bank Did and Did Not Do in Nigeria’s South
Day 1 of the Nigeria series. Kofi arrives in Benin City and encounters the Bank’s operational record in Edo State: the SEEFOR project, the power sector, and the gap between what the Bank financed and what was delivered on the ground.
Read the PaperRethinking the IDA Private Sector Window — From Internal Allocation to Competitive Deployment
206 PSW projects, $6.18bn. Subsidy ≥100% in 28 Local Currency Facility transactions. Non-PSW IFC commitments in eligible countries fell in aggregate. Additionality realised in 33% of FCS projects. That is a refund, not development finance.
Full project-level empirical assessment of the PSW portfolio across IDA18–IDA20. The Local Currency Facility finding. The additionality test. The political economy of why the PSW persists despite the evidence.
Read the PaperKofi & SEEFOR — The Community Development Fund That Did Not Reach Communities
Day 3 of the Nigeria series. The SEEFOR (South-South and South-East Economic Recovery and Empowerment Programme) examined against the IEG evaluation record. What fiduciary compliance looks like when accountability is absent.
Read the PaperKofi & Kobo — The FinTech That Works and the Financial Inclusion Portfolio That Does Not
Day 4 of the Nigeria series. Kofi encounters Kobo360 and the Nigerian fintech ecosystem — and the contrast with the Bank’s own financial inclusion portfolio performance in Nigeria. What works in private digital finance and what does not work in public lending.
Read the PaperKofi the Contractor — Procurement, Corruption and the Accountability Gap
Day 5 of the Nigeria series. Kofi meets a local contractor and hears what the Bank’s procurement system looks like from the other side. The gap between formal compliance and functional integrity. What Prior Review means in practice in a high-fiduciary-risk environment.
Read the PaperKofi at the Palace — The Palace of the Oba and the Bank’s Political Economy
Day 6 of the Nigeria series. Kofi visits the Palace of the Oba of Benin and reflects on what the Bank’s engagement with traditional authority structures in Nigeria has and has not achieved. The political economy of project design in a contested governance environment.
Read the PaperDays 7–10: Lagos, Abuja & the Federal Portfolio — Coming
Four further instalments covering Lagos infrastructure, the Abuja federal engagement, the SOML health programme, and the Nigeria Power Sector analysis in the context of the full Nigeria country series.
The Richmond Reckoning — Tanzania, Parliament & the Cost of AccountabilityEssay · 2026
The Richmond Reckoning
What happens when a courageous Speaker of Parliament appoints a young MP to chair a parliamentary committee to enquire into an irregular procurement in the energy sector? The enquiry report led to the resignation of the sitting Prime Minister and two other Ministers — the first time this happened in Tanzania.
The Richmond Power case. What parliamentary accountability looks like when it functions. The contrast with the Eskom case: two energy projects, two different accountability architectures, two different institutional responses. What Tanzania did in 2008 that the World Bank Board has never done in 82 years.
Read the EssayCountry Case Studies — Does the World Bank Learn from Failures?Parts 1–8 · 2024–2026
Nigeria Water — 34 Years. $1.8 Billion. 0.4% Satisfactory. One Household.
Eight projects. Six rated below Satisfactory. One Satisfactory: a $5M community-managed pilot. The eighth — SURWASH, $700M — is rated MU at mid-term. One household with improved sanitation out of a target of 280,000. IEG diagnosed the structural problem in 2006. The design model did not change.
Two IEG PPARs, five ICR Reviews, the SURWASH ISR (June 2025). The $5M pilot that worked and the $700M operation that is following the same trajectory as the six that came before it. Source documents published for independent verification.
Read the Paper Download PDFAngola DPF — 24 Years. $2.2 Billion. Zero Satisfactory. The Same Lesson, Written Three Times.
Six DPF operations. Zero Satisfactory. One Unsatisfactory. One PDO Highly Unsatisfactory. The IEG lesson for the $1.7bn Growth & Inclusion series was recorded word for word from the $450M Fiscal Management DPL that preceded it. $4.34bn total exposure. The pipeline never dried up.
The $450M smoking gun: IEG found it arose from “pressure to design a high-profile program that would garner strong international buy-in.” The Bank then approved $1.7bn more. The GRID DPF ISR (December 2025): High political, macroeconomic, and fiduciary risk. 0% beneficial ownership registered. Resilient & Inclusive Growth DPF ($1.15bn) in the pipeline.
Read the Paper Download PDFSouth Africa Eskom — $9.1 Billion. Rated MU. Nine Consecutive Unsatisfactory ISRs. No PPAR.
The largest single-project failure in the Bank’s Africa portfolio. The Bank approved the $3.75bn loan over the objections of five shareholder countries and an Inspection Panel complaint filed two days before the Board vote. Nine consecutive Unsatisfactory ISRs. $1.7bn disbursed while the project was failing. The RVP who presided over the nine Unsatisfactory ISRs subsequently became Managing Director of IFC.
The Zondo Commission record: R14.7bn in Eskom contracts “afflicted by State Capture.” The Bank conducted Prior Review on every major contract and approved six revised procurement plans during the period the Commission found procurement was being manipulated. The boiler furnace finding. The word “Gupta” does not appear in the 106-page ICR.
Read the Paper Download PDFDRC: The Portfolio That Does Not Deliver — 49 Projects. $6.7 Billion. 6.1% Satisfactory.
49 evaluated projects. $6.7bn committed. 6.1% commitment-weighted S+. The Bank finances the simulation of state functionality in a fragmented rent-distributed political settlement.
The DRC portfolio analysed project by project against the IEG evaluation record, sector by sector. The MTI dominance finding. Why the pipeline persists despite the outcome record. The political economy of lending in a permanent fragility context.
Read the PaperDRC: Inga — The Dam That Does Not Generate. 40 Years. SNEL Bankrupt.
40 years of engagement. ISR collapse trajectory. SNEL bankruptcy. The bankability problem for the new $250M re-engagement. The Bank has never successfully completed a large hydropower operation at Inga.
The full history of Bank engagement at Inga from the 1970s to the present. Why the technical case for the dam remains strong while the institutional case for the Bank’s involvement does not. The DRC Multi-modal Transport case file as a parallel: money deployed after the Bank’s own supervision flagged failure.
Read the PaperGhana FCI — The Jobs Machine That Does Not Create Jobs. $615 Million. 0% Satisfactory.
14 IEG documents across FY1997–2024. $615M. All below S+. The Bank’s diagnostic evolution across three decades has been sophisticated. Its operational model has not changed. SSA FCI transport S+ rate: 10.8%. Seven operational recommendations.
The Ghana FCI portfolio as a case study in the gap between analytical advancement and operational learning. The MSME agenda: what the IEG record says about its delivery. The jobs numbers that were promised and the jobs numbers the evaluations document.
Read the PaperThe Rwanda Model — What the Data Actually Shows
Rwanda is the most cited counter-argument to the platform’s findings. This paper examines what the IEG data shows about the Rwanda portfolio and where the “model” characterisation holds and where it does not.
The Rwanda portfolio evaluated project by project against the IEG outcome ratings. The sectors where Rwanda does outperform. The sectors where it does not. What the Rwanda record actually demonstrates about the conditions under which Bank operations succeed — and what it does not demonstrate about scalability to other contexts.
Read the PaperThe Somalia Exception — Fragile State, Exceptional Constraints, Exceptional Results
Somalia sits at the extreme end of the fragility spectrum. And yet the Somalia portfolio, operating through third-party implementation in an active conflict environment, records 67% S+ — the highest rate of any FCV country in the IDA Africa portfolio. The exception does not disprove the rule. It demonstrates what the rule requires to be overcome.
The Somalia model examined against the IEG record. Why third-party implementation under extreme constraints outperforms Bank-direct implementation in stable middle-income countries. What the Somalia finding implies for the IDA21 RECA architecture and the case for systematising the model.
Read the PaperIFC Accountability — The Commercial Activity Question2 Notes · July 2026NEW
When Accountability Becomes Optional: The IFC Board, the CAO, and the Precedent Set in Cambodia
The IFC Board’s June 23, 2026 determination overturning the CAO’s four-year Cambodia investigation is without precedent in the public record of MDB independent accountability mechanisms. CAO Director General Janine Ferretti resigned the next day. The case documents forced land sales, apparent debt-driven suicides, and harm to Indigenous Peoples across six IFC-backed microfinance institutions.
The CAO found IFC violated its Sustainability Framework during pre-investment due diligence and ongoing supervision. IFC Management refused to submit a Management Action Plan and instead submitted a Management Report rejecting the findings outright. The Board endorsed Management’s rejection. What this means for the new Integrated Accountability Mechanism, the reporting-line architecture that enabled the override, and the financial intermediary scope precedent that now covers IFC’s largest lending channel.
Read the PaperThe Immunity Paradox: How IFC Cannot Be Sovereign and Commercial Simultaneously
IFC tells US courts it is a sovereign institution not engaged in commercial activity — entitling it to immunity from suit under the FSIA. It tells its own Board it is a commercial investor one step removed from end borrowers — entitling it to escape E&S accountability for downstream harm. Both cannot be true. The Board’s June 23, 2026 determination puts both arguments on the primary record simultaneously, in the same case.
What Jam v. IFC (2019) decided and — crucially — did not decide: the commercial activity question was sidestepped, not resolved. Why IFC’s financial intermediary activity is commercial under FSIA. Why the Board’s June 23 Washington decision presents a different factual structure from the Gujarat case. The binary IFC cannot escape: sovereign status requires citizen-facing accountability; commercial status defeats immunity.
Read the NoteThe Architecture Problem: Why Combining Sovereign and Commercial Functions in One Institution Breaks Accountability
On the structural conflict embedded in the WBG’s combination of IBRD, IFC, and MIGA — documented in the Nigeria Power Sector case — and why the IAM merger makes it worse. Connects the Nigeria Power Sector conflict-of-interest findings, the IAM merger, and the Immunity Paradox arguments.
The Zero Club — Sectors Where No Country Achieved Satisfactory OutcomesParts 9–13COMPLETE
The Zero Club — MTI: 14 Countries. 99 Projects. $10.4 Billion. 0% Satisfactory.
MTI is the worst-performing Global Practice in Africa by project count: 12.6% S+ across all evaluated projects. In 14 countries with 2+ evaluated projects, the S+ rate is zero. DPF bifurcation finding: MTI 23.8% vs sector GPs 43.0%. Nash equilibrium explanation for pipeline persistence.
Full Schick 7 PEFA framework analysis. CPIA trend analysis. OBS demand-side data. Niger ICRR deep dive (P175256). The structural argument for why the pipeline never dries up regardless of the outcome record. The conflict of interest at the heart of the MTI model: the practice that owns the instrument, manages the Finance Ministry relationship, and absorbs no consequence when operations fail.
Read the PaperThe Zero Club — Health: 11 Countries. 49 Projects. $2.9 Billion. 0% Satisfactory.
11 countries. 49 projects. $2.9bn. If both MTI/DPF and Health/IPF fail, the problem is instrument-agnostic and practice-agnostic. Kenya (8 projects, $694M, non-fragile, non-MTI) as anchor case: the income-level and instrument explanations do not survive contact with the Kenya health record.
The Zero Club Health analysis placed alongside the Spring Meetings health record (29% S+ across 78 projects, $8.4bn). The convergence of two methodologies on the same finding. What the Kenya health portfolio tells us about the limits of the “strong institutions” explanation for poor outcomes.
Read the PaperThe Zero Club — Transport: The Foundation That Has Never Held.
Countries with 2+ evaluated transport projects and 0% Satisfactory. Four consecutive years of zero Satisfactory transport outcomes in Africa (FY2019–22). The Zero Club transport record placed alongside the sector-wide 4% S+ by commitment finding from the Spring Meetings. The same countries. The same outcome. A different lens.
Transport is the largest sector by commitment in the Africa portfolio. It is also, on the commitment-weighted measure, the worst-performing. The NH28 India case (Unsatisfactory, $615.7M, 99.3% disbursed) from the IBRD Disbursement Disconnect paper as the non-Africa anchor: the transport failure is not a fragility artifact.
Read the PaperThe Zero Club — Education: Human Capital Without Human Results.
Countries with 2+ evaluated education projects and 0% Satisfactory. The Ethiopia $1.334bn zero-Satisfactory anchor case. Four independent evidence sources — HCI+ 2026, IEG 2024 flagship, Rolleston et al. 2025, UNESCO 2026 — converge on the same finding through different methodologies: the Bank gets children into school; whether they are learning inside it is not consistently measured.
The GEQIP paradox: one of Africa’s largest education reform programmes never achieved a Satisfactory rating. Strong systems proved necessary but not sufficient. The four-model synthesis (GEQIP, Kenya SWAp, Tusome, LEAP): learning improves when the intervention reaches the classroom — through structured pedagogy, teacher coaching, accountability for results, and continuous measurement. The MS+ gap: 41 percentage points — among the widest of any sector in Africa.
Read the PaperZero Club — Synthesis: The Cross-Sector Pattern.
The synthesis across MTI, Transport, Education, and Health. What the Zero Club methodology reveals that sector-level analysis conceals: the country-GP intersection is the unit of failure, not the sector alone. 43 Zero Club countries across 14 Global Practices, 419 projects, $29.8bn — and a pattern that is consistent across income levels, fragility classifications, and lending instruments.
The cross-sector finding: DRC appears in five Zero Clubs simultaneously. Countries that fail in one practice fail in multiple. The structural argument: the failure is not sectoral incompetence but an institutional equilibrium in which no actor in the chain has the incentive to halt disbursement regardless of outcomes. The sovereign guarantee as the key parameter that makes the equilibrium stable.
Read the PaperQuality, Disbursement & Ratings ArchitectureParts 14–17 · June 2026
Quality at Entry in IDA — What the Ratings Tell Us
When both QAE and M&E are strong, 73% of IDA projects achieve Satisfactory outcomes. When both are weak — 52% of the portfolio since FY2000 — 5.4% do. Recovery rate from poor design: 10.7% in the 1980s, 0.0% in the 2020s. 735 projects with poor QAE committed $50.6bn at a 3.8% S+ rate. IDA’s own quality system flagged every one before the Board voted.
QAE gradient table (all 6 tiers, 2,141 projects). Recovery collapse by decade. QAE × M&E interaction (four quadrants). QAE by Global Practice. CPIA robustness tests. Speed-quality trade-off. IEG evaluator text analysis (136 projects). The approval culture: Wappenhans to the present. The QAG disbandment in 2014 as the institutional turning point: the only independent pre-Board design review that existed was removed at precisely the moment the coverage evidence made its value clear.
Read the PaperThe IDA Disbursement Disconnect — $158bn Disbursed. $104bn Below S+. 0% ERR Since 2015.
A project rated Unsatisfactory disburses 97.3% of its commitment. A project rated Satisfactory disburses 99.4%. The difference is 1.8 percentage points. $104bn — 65.8% of all disbursements — went to projects that did not achieve a satisfactory outcome. Of IDA Africa projects approved since 2015, not one carries a calculated economic rate of return. The Bank has not merely become disconnected from results. It has become disconnected from appraisal itself.
Disbursement decoupling by outcome band. ERR-coverage collapse (71% in the 1970s → 0% since 2015). The sovereign-guarantee mechanism. IEG’s 2010 cost-benefit evaluation: 82% of TTLs said CBA was never the key funding criterion; 80% said it was sufficient to “tick a box.” The institutional response: Board calls non-compliance “unacceptable” (2010) → OP 10.04 consolidated (2013) → QAG disbanded (2014) → ERR reaches zero (2015). Three Presidents. Fifteen years. The DRC Multi-modal Transport Project case file: $133 million deployed after the Bank’s own supervision rated it Unsatisfactory; suspension proposed twice, declined because it “would have had a negative impact on the Bank–Government relationship.”
Read the PaperThe IBRD Disbursement Disconnect — $554bn Disbursed. $257bn Below S+. 0% ERR Since 2020.
An Unsatisfactory IBRD project still disburses 89% of its commitment. $257 billion — 46.4% of all IBRD disbursements — went to non-satisfactory operations. Of IBRD projects approved since 2020, not one carries a calculated ERR. The income-level explanation does not survive: MENA scores 33.6% S+ on $41bn of lending to middle-income governments — worse than IDA Africa. The Bank has become disconnected from appraisal in both windows at the same institutional moment.
Disbursement decoupling by outcome band (six tiers, 4,535 projects). Duration-outcome relationship: attenuated vs IDA, IBRD borrowers close bad projects faster. Effectiveness lag as predictor. QAE gradient: 73% S+ when adequate, 8.4% when not — a 65-percentage-point gap virtually identical to IDA Africa. DPF: $237bn at 50.8% S+ — lowest instrument rate, 43% of all IBRD disbursements. MENA outlier: 33.6% S+ on $41bn — middle-income governments, no fragility classification. Project size: 261 projects above $500M at 50.2% S+. The four-project exemplar: Eskom, NH28, Vietnam PIR 1, Brazil Env DPL. Russia’s SAL 2 (P050491, $800M, 100% disbursed, Unsatisfactory, FY1998) extending the record to the post-Soviet period.
Read the PaperAAA Credit. Unrated Development Effectiveness. — The Ratings Architecture That Makes Delivery Failure Financially Invisible.
The World Bank holds a triple-A credit rating. It also deploys 46.4% of IBRD and 65.8% of IDA Africa disbursements to non-satisfactory operations. Both statements are true. They are not in conflict. The mechanism is the sovereign guarantee. “Development effectiveness” appears zero times across both Moody’s credit opinions. The IDA21 Deputies Report told 59 donor governments the S+ rate was 91%. The honest figure is 31%. Three times the honest number.
The upstream paper in the Zero Club Series: the institutional architecture that makes the pattern financially invisible. The sovereign guarantee does two things simultaneously — it justifies the AAA and severs development performance from financial consequence. The five-step Caa1-to-Aaa lift table. The Moody’s word count: zero. The management-accounts finding: development outcome risk named in one sentence in both quarterly filings and never quantified — management has already mapped the gap the rating agencies will not. The S&P contradiction: “highly scrutinised for impact” vs ERR coverage = 0%. IBRD RAC 24%→40%, IDA RAC 58%→90% in one revision cycle. The 91% problem in full. Eleven proposals to the rating agencies, World Bank management, and the IDA21 Mid-Term Review. Sources: Moody’s IBRD Credit Opinion (February 2026), Moody’s IDA Credit Opinion (February 2025), S&P Supranationals Special Edition 2025, IBRD and IDA MD&A Q3 FY2026 (March 2026).
Read the PaperNigeria Power Sector — How Not to Do PPPs in Africa’s Largest EconomyCountry Deep Dive · 2026
Nigeria Power Sector — How Not to Do PPPs in Africa: NEGIP & PSGP (P106172, P120207)
The World Bank Group simultaneously served as policy adviser, IFC equity investor, and MIGA/IBRD guarantor in Nigeria’s power sector — a $900 million combined exposure. Result: a $1 billion financial deficit, take-or-pay contracts forcing Nigeria to pay for power it cannot use, and 2,638 MW of hydropower curtailed. The Bank turned from policy adviser to debt collector.
A structural conflict-of-interest case study: when IFC holds equity, MIGA provides political risk insurance, and IBRD backs sovereign payment guarantees in the same sector simultaneously, the independence of the Bank’s policy advice is compromised. The NEGIP and PSGP ICRs document how guarantee monitoring is at odds with broader sector reform dialogue. Draft submitted to the World Bank Group for factual review; no response received.
Read the PaperSector Performance Records — Spring Meetings 2026Seven Sectors · Zero Accountability · April 2026
Zero Accountability — Eleven Findings from the Spring Meetings
$51 billion below standard across seven sectors in Africa. Not a single GP above 38 cents on the dollar. The TTLs who led the worst projects were promoted. The IMF disbursed $170 billion with the instruction to “keep the receipts.” Both institutions collected repayment in full.
The synthesis article for the seven sector records. Sovereign guarantee thesis. Game theory equilibrium. MTI capture. The promotion culture. The IMF parallel. Two reforms: the IDA Challenge Fund and competitive allocation.
Read the ArticleThe Transport Record — The Foundation That Does Not Hold. 4% S+ by Commitment. 96% Below.
65 projects. $11.2bn. 96% below standard by commitment. Four consecutive years of zero Satisfactory (FY2019–22). Twelve countries with 2+ evaluated projects at 0% S+. The largest sector by commitment is the worst by outcome.
The full Africa transport record from the IEG database. Country-by-country breakdown. The NH28 India case as a non-Africa comparator: the transport failure is not a fragility artifact. Five structural failure modes drawn from IEG lesson texts.
Read PDFThe Energy Record — Mission 300. Mission Impossible. 15% S+ by Commitment. 85% Below.
99 projects. $21.5bn. 85% below standard. South Africa $9.9bn zero Satisfactory. Eskom $9.1bn MU. Mission 300 is a commitment to connect 300 million people to electricity by 2030. The evaluated record of the institution making that commitment shows 15% success on $21.5bn.
The full Africa energy record. South Africa as the dominant distortion: $9.9bn at zero Satisfactory. The Eskom precedent in depth. Sub-Saharan energy access gap: what the IEG record says about the Bank’s capacity to close it at the current delivery rate.
Read PDFThe Water Record — Water Forward. Delivery Backward. 25% S+ by Commitment. 75% Below.
53 projects. $5.1bn. 75% below standard. Sharpest FCS inversion of any sector: non-FCS 18.4% S+ vs FCS 40% — fragile states outperform stable states in water. The Nigeria Water case as the anchor: 34 years, $1.8bn, one household.
The full Africa water record. The FCS inversion finding in detail: why Somalia-model third-party implementation produces better water outcomes than Bank-direct implementation in stable contexts. The SURWASH $700M mid-term trajectory.
Read PDFThe Health Record — Health Works Without Accountability. 29% S+ by Commitment. 71% Below.
78 projects. $8.4bn. 71% below standard. Tanzania $1.3bn zero Satisfactory. Mozambique cited repeatedly by Bank management as a health success story: 0% S+ in the IEG record. The gap between the management narrative and the evaluation record is widest in health.
The full Africa health record. The Mozambique contradiction: management’s narrative vs IEG’s ratings on the same projects. Tanzania $1.3bn zero-Satisfactory in detail. What the health record implies for the Bank’s UHC agenda.
Read PDFThe Education Record — Human Capital Without Human Results. 31% S+ by Commitment. 69% Below.
92 projects. $6.6bn. 69% below standard. MS+ gap: 58 percentage points — the widest of any sector in Africa. Projects measure inputs and outputs — schools built, enrolment up, teachers trained — not learning. IEG’s own 2024 flagship: projects with learning indicators receive lower ratings than those without.
The full Africa education record. The Moderately Satisfactory equilibrium: how “partial achievement” absorbs the gap between schooling and learning and renders it invisible in corporate reporting. Ethiopia GEQIP $1.334bn: strong systems, declining learning outcomes. The Kenya evolution: three generations of education lending and what each one learned.
Read PDFThe FCI Record — The Jobs Machine That Does Not Create Jobs. 36% S+ by Commitment. 64% Below.
67 projects. $4.4bn. 64% below standard. Projects named “Competitiveness and Job Creation,” “Private Sector Development,” or “MSME Development” are all rated MU or worse. The FCI practice in Africa succeeds at fewer than two in five operations regardless of how the operation is branded.
The full Africa FCI record. The branding-outcome gap: operations whose titles promise jobs are no more likely to deliver them than operations that do not. The Ghana FCI case (Part 6 of the country series) as the anchor: 14 documents across 27 years, $615M, 0% S+. What the FCI record implies for the IDA21 jobs agenda.
Read PDFThe Agriculture Record — Scale Without Transformation. 38% S+ by Commitment. 62% Below.
98 projects. $8.0bn. 62% below standard. DPF and PforR combined: zero Satisfactory in the Africa agriculture portfolio. The highest-performing instrument is IPF at 41% S+ — the only instrument that requires project-level appraisal.
The full Africa agriculture record. The instrument finding: the two instruments that disburse fastest — DPF and PforR — produce zero Satisfactory outcomes in Africa agriculture. What the agriculture record implies for food security ambitions and the Africa food systems agenda. The Southern Africa drought financing question.
Read PDFThe Knowledge Bank — $1 Billion in Advisory Services and Analytics. No Evaluative Framework.
$1bn per year. No systematic IEG evaluation. Twenty years after Banerjee-Deaton. The same diagnosis. No resolution. The Bank’s advisory and analytical services are the fastest-growing expenditure line and the only major activity with no outcome measurement framework.
The ASA architecture: what gets measured, what does not, and why the absence of evaluation is not accidental. The Banerjee-Deaton critique (2007) and the Bank’s response. The country diagnostic ecosystem: PRSPs, CPFs, CEMs, PFRs, CLRs — what they cost and what evidence exists that they change anything. The case for a systematic ASA evaluation framework built on the same IEG standard as project lending.
Read PDFFCV Strategy — The Delivery Record in Fragile States6 Papers · 2026
Sound Strategy, Broken Platform — Analytical Note on the World Bank FCV Strategy 2026–2030
The strategy is directionally correct. The delivery platform that must implement it records 68.9% of IDA commitments below Satisfactory in FCV contexts. Implementation failure, not strategic design failure, is the central problem. The “broken platform” characterisation is the finding, not a rhetorical choice.
Built on the formal consultation comment submitted to the Bank. Structured around what the comment established, what the new strategy changed, and what remains unanswered. Includes the S+/MS+ methodology box and five-reform prescriptive table. Grounded in the matched IEG dataset.
Read the NoteFCV Strategy Series — Six Papers and Full 130-Page Submission
The full consultation submission: six papers covering the delivery record ($117bn below standard), four root causes, IFC in fragile states (11% S+), IDA21 gaps, and eight recommendations in priority order.
Paper 1: The Strategy Is Right. The Delivery Platform Is Broken. Paper 2: $117bn Below Standard. Paper 3: Four Root Causes. Paper 4: IFC in Fragile States. Paper 5: IDA21 — What It Gets Right and What It Doesn’t Fix. Paper 6: Eight Recommendations in Priority Order.
Series Landing Page Download Full PDFIDA Reform, Private Sector & Lending Instruments2024–2026
IDA at 65: Why Is the World’s Largest Concessional Fund Delivering Satisfactory Outcomes on Only 31% of Its Portfolio?
$117 billion below standard in a decade. 31% Satisfactory. The IDA21 Deputies Report told donors the rate was 91%. It was not. The gap between those two numbers is the governance mechanism that has protected IDA’s delivery failure from accountability for four decades.
From IDA1 ($912 million, 1961) to IDA21 ($100 billion, 2025) — all 21 replenishments, the Deputies process, and the leveraging model. MTI at 11.7% on $26.9bn. DPF in fragile states at 10.8%. Africa receives 70% of IDA resources and records the weakest outcomes. The case for a Challenge Fund for IDA22: 25% of IDA resources competitively allocated to non-Bank implementing partners against independently verified results, evaluated on the same IEG standard as Bank projects.
Read the Paper Download PDFRethinking the IDA Private Sector Window: From Internal Allocation to Competitive Deployment
206 projects. $6.18 billion. 83.5% through IFC-managed facilities with no competitive allocation. IFC both originates and assesses the additionality of its own transactions. Non-PSW IFC commitments in PSW-eligible countries fell during IDA18–IDA20. The window is open, but only one institution holds the key.
The IDA PSW deployed concessional subsidies to catalyse private investment in fragile states. Project-level data shows additionality was realised in only 33% of FCS projects. The PSW replaced IFC’s own capital rather than supplementing it. Six reforms proposed: competitive allocation, results-based LCF, open-access platform, performance-based replenishment share, sector floor, and independent evaluation.
Read the PaperPolicy Without Performance: Isomorphic Mimicry and the DPO Incentive Trap
MTI S+ rate: 27.5%, down from 41% (2005–09) to 17% (2015–19). SSA CPIA flat at 3.1 for 18 years. 11,628 prior actions. The form-function gap: governments adopt the outward forms of reform to trigger disbursements while the underlying administrative capability remains absent.
1,551 evaluated DPF operations. 20 years of IEG lessons the Bank has not absorbed. Five reform proposals. The Angola DPF pipeline as the case study: the same lesson documented by IEG, the same operation approved three times, the same outcome each time.
Read the PaperPaper Triggers by Global Practice: The Evidence Record — Companion Annex
Five GPs. Forty-plus prior action templates. Each with the disbursement trigger, the documented function gap, and the specific IEG project ID and rating. The prior action that triggers disbursement and the evaluation that documents it did not produce the stated objective, side by side.
Read the AnnexDesigned to Fail: The SOML PforR Case Study (Nigeria P146583)
$500 million. The world’s largest PforR at approval in 2015. Rated Moderately Unsatisfactory by IEG, efficiency Negligible. The fiduciary assessment approved the programme knowing procurement fraud had occurred in the implementing ministry.
83% of first-year disbursements — $52.9 million in state performance grants — sat on the balance sheet as a single unaudited line. The management ICR rated it Moderately Satisfactory throughout. Nigeria will repay $387.6 million over 38 years. The Bank earns its spread regardless of outcome. The SOML case as the definitive demonstration that PforR — designed to pay for results — can disburse regardless of whether results are achieved, when the results definition is sufficiently ambiguous and the disbursement-linked indicators sufficiently input-oriented.
Read the PaperGovernance & Institutional Architecture2024–2026
Why the System Does Not Learn: A Game Theory Analysis of the World Bank’s Institutional Equilibrium
The approval culture is a Nash equilibrium. No single actor inside the institution has the incentive to exit unilaterally. The TTL who pushes a failing project through approval is individually rational. The RVP who keeps disbursing during nine consecutive Unsatisfactory ISRs is individually rational. Only the Governors can change the structural parameters.
The formal game theory framing of the argument that runs through the entire platform. Connects the Angola DPF pipeline, the Eskom supervision record, the MTI capture finding, and the promotion culture into a single structural explanation. The sovereign guarantee as the key parameter that insulates every actor from the development outcome.
Read the PaperInstitutional Power Architecture and Portfolio Distortion at the World Bank
MTI: 18.5% Satisfactory globally, 13.5% in FCS. $61.5bn of $88.0bn below standard. Economist Country Directors allocate 1.5–2x more to DPOs. MTI controls the DPF instrument, manages the Finance Ministry relationship, and influences Country Director appointments. It has a structural conflict of interest: it designs the operations, receives no consequence when they fail, and determines whether the next one is approved.
The power architecture that produces the MTI outcome record. The Country Director appointment mechanism. The DPF pipeline dynamic. Why the Nash equilibrium in the game theory paper is stable: MTI has no individual incentive to improve its outcome rate because the pipeline does not depend on the outcome rate.
Read the PaperMIGA — The Guarantor That Cannot Verify Its Own Results2 Notes
The MIGA Record — The Guarantor That Cannot Verify Its Own Results
Africa S+ 72% → 50%. Work quality 56% → 43%. IMPACT launched FY2020 — zero projects evaluated by IEG. MIGA’s website: “every outcome claim clearly linked.” IEG: 69% not tracked. Disclosure Statement: “MIGA makes no guarantee or other promise as to any outcomes.” One document promises. The other disclaims.
The RAP trajectory. The IMPACT framework and the Disclosure Statement contradiction. The foreign investment outcome gap (half rate). The Non-Honoring products (86% to upper-middle and high-income). The financialisation ($7.1bn to reinsurance). First African country at #13. Africa’s share halved: 40% to 20%. Six questions for the Board.
Read the Paper Download PDFMIGA in Africa — 459 Projects. $36 Billion. 20% of Exposure. Africa’s Share Halved.
Power = 36% of Africa exposure. Top 5 countries = 42% of Africa exposure. Same countries where IDA lending has zero Satisfactory in transport and energy. Databases cannot be linked. FCS instrument dormant since FY2017. Not evaluated.
The Africa portfolio from the WBG Guarantees database. Top 10 countries. Power sector dominance. The IEG cross-reference: Nigeria, Kenya, South Africa, Ethiopia — MIGA guarantees on top of zero-Satisfactory IDA records. Does MIGA add demonstrable value in Africa beyond what IDA lending already provides?
Read the Paper Download PDFPEFA & Public Financial Management2 Papers
PEFA & Sub-Saharan Africa
SSA average: 2.3/4.0. Joint lowest globally. Scores highest on budget documentation (de jure compliance). Scores lowest on external audit (functional accountability). The form-function split in PFM reform made visible: countries pass PEFA on the indicators that measure what documents say, and fail on the indicators that measure what institutions do.
46 country assessments across 32 SSA countries, 2016 framework. 13 countries with two or more assessments enabling trend analysis. Schick’s seven core indicators. The basics-first case: the countries that have improved most in SSA are those that addressed basic controls before attempting advanced reforms.
Read the PaperPEFA at the Crossroads: Service Delivery, Core Controls, and the Cost of Losing the Thread
PI-30 External Audit: 1.9 out of 4.0 — the lowest of all 31 indicators. PI-21 Cash Management: 2.2. PI-24 Procurement: 2.1. Every one of Schick’s seven foundational indicators is below the adequate performance threshold in Sub-Saharan Africa. Meanwhile the PEFA Secretariat is promoting PEFA Climate, PEFA Gender, PEFA++, and PEFA SDGs. The framework has lost the thread.
Written by a member of the original PEFA indicator design team. The argument for basics-first: no country has achieved sustained improvement in public services without first establishing the seven core controls. The new variants — Climate, Gender, SDG — ask countries to measure what they cannot yet manage. The cost of losing the thread is paid by the populations who depend on functioning public administration.
Read the PaperFollow the Money — World Bank Procurement: Who Captures the Contracts?2 Papers · June 2026NEW
Follow the Money: $32.5 Billion of World Bank-Financed Contracts in Africa — Who Captures Them and Where the Money Goes
$32.5bn in World Bank-financed contracts in Sub-Saharan Africa. 59% — $19.2bn — left the continent. Sub-Saharan Africa is the only region where the majority of contract value consistently flows outside the borrowing economy. Domestic capture: 41%, against 77% in South Asia and 76% in East Asia Pacific. China’s African share peaked at 30% in FY2024. Chinese subsidiaries registered locally — CGC Nigeria, China Jiangxi (Ghana), Yangtse River — count as domestic. The reported domestic share is an upper bound; the reported Chinese share is a lower bound.
The first systematic mapping of World Bank-financed procurement in Africa by contractor nationality, sector, and fiscal year across FY2020–2026. Country, sector, and contractor-level detail. The six-year trend shows no domestic capture recovery: domestic share ranged from 33% (FY2022) to 47% (FY2023) with no upward direction. At least $740M of nominally domestic African contract value is identifiably Chinese-owned on name alone.
Read the Paper Download PDFFollow the Money: $125.3 Billion of World Bank Contracts — 280,959 Awards, Eight Regions, Six Fiscal Years
The G7 holds the largest blocs of World Bank voting power, nominates its President by convention, and supplies 5.2% of contract value. China holds 5.8% of votes and supplies 17.4% — 3.4 times the entire G7. US construction contracts across all regions over six years: $91M — roughly one mid-size road in Tanzania. On what the Bank spends on itself ($1.4bn in corporate procurement), the order inverts: US 23.7%, G7 41.6%, China 0.3%. The economies that govern the institution capture its operating budget; the lending-side contracts flow elsewhere.
The first complete mapping of World Bank lending-side contract awards across all eight regions and six fiscal years. Domestic capture globally: 56%. Sub-Saharan Africa: 41% — the only region where the majority of contract value consistently leaves the borrowing economy. Outside Africa, domestic capture recovered above 73% by FY2024 after the COVID vaccine procurement dip. Inside Africa, there is no equivalent recovery. Full regional breakdown, six-year trends, shareholding comparison, and the twenty largest contracts globally.
Read the Paper Download PDFExternal Critiques & Institutional Commentary2026
The Survey That Did Not Ask the Right Question — A Response to ODI’s “Reforming MDBs: Perspectives from Client Countries”
73% rate the World Bank as “very effective.” The IEG record: 4% Satisfactory in transport, 15% in energy, 25% in water. $51 billion below standard. The survey did not show respondents the outcome data. If you ask the wrong questions, you get the wrong answers.
650 officials. 125 countries. The most comprehensive MDB client survey ever conducted — and it did not use the most comprehensive MDB outcome data that exists. The report measured perception. IEG measured outcomes. The two have never been placed side by side. This response identifies what the survey gets right, where it stops, and what a third edition should ask.
Read the ResponseEvidence Without Accountability — A Critique of ODI Global’s 2026–2031 Strategy
ODI’s incoming Director admitted the organisation nearly went bankrupt. 78% donor dependency. The strategy proposes to hold MDBs accountable using the same MS+ benchmark the MDBs use to hide their own failures. An accountability organisation that adopts the accountability standard of the institutions it scrutinises has surrendered the argument before making it.
Four-section critique: the financial model, the MS+ benchmark problem, the IDA performance data ODI has not published, and the governance gap. Integrates the IDA performance record ($117bn / 68.9% below-S+), the IFC/MIGA “Successful” scale as evidence that MS+ is not an industry standard, and Charity Commission financial data.
Read the Critique