Ghana · Finance, Competitiveness and Innovation (FCI) · MDB Reform Platform
Does the World Bank Learn from Project Failures? (Part 4) — The Jobs Machine That Does Not Create Jobs
14 FCI Projects. $615 Million Committed. 27 Years. Zero Satisfactory Outcomes. 100% of Resources Below the S+ Threshold.
The Bank Should Not Attempt to Buy Reforms
In 1997, the World Bank approved a $30 million credit line for Ghana’s private sector — the SME/Finance project. It was designed to foster enterprise development and generate employment. Default rates were high. The guarantee scheme was never implemented. Assistance to microenterprises was negligible. Both Bank and Borrower performance were rated Unsatisfactory.
IEG’s lesson: “Capacity building in the financial sector is more likely to succeed in the presence of macroeconomic stability, a disciplined financial system, and a business environment conducive to private investment.”
Ghana had none of these conditions. The Bank lent again.
By 2000, the Private Sector Development project was rated Unsatisfactory. IEG wrote: “The Bank should not attempt to buy reforms in a subsector without strong evidence of government commitment.” By 2002, the Non-Bank Financial Institutions project — fourteen components, forty-five sub-components, eleven beneficiary agencies — was rated Unsatisfactory. The divestiture objective was “never seriously pursued.” When undisbursed funds accumulated, the Bank added an HIV/AIDS awareness component entirely unrelated to the project’s objectives. IEG called it “haste makes waste.”
The Bank lent again. And again. And again. Fourteen times over twenty-seven years.
The Complete Record
| # | P-Code | Project | FY | Rating | Inst. |
|---|---|---|---|---|---|
| 1 | P000911 | FINSAC II | 1997 | MS | DPF |
| 2 | P000905 | SME/Finance | 1997 | U | IPF |
| 3 | P000920 | Enterprise Development | 1998 | MU | IPF |
| 4 | P000967 | Private Sector Adjustment | 1999 | MU | DPF |
| 5 | P000960 | Private Sector Development | 2000 | U | IPF |
| 6 | P000943 | Non-Bank Financial Institutions | 2002 | U | IPF |
| 7 | P042516 | Public Enterprise/Privatisation | 2004 | MU | IPF |
| 8 | P069465 | Rural Financial Services | 2008 | MU | IPF |
| 9 | P000970 | Trade Gateway & Investment | 2010 | MS | IPF |
| 10 | P092986 | Econ. Mgmt Capacity Building | 2013 | MU | IPF |
| 11 | P125595 | Ghana PPP Project | 2018 | U | IPF |
| 12 | P145765 | Climate Innovation Center | 2021 | MS | IPF |
| 13 | P161787 | Financial Sector Development | 2024 | MS | IPF |
| 14 | P164211 | Tourism Development | 2024 | MS | IPF |
A note on ratings
This paper uses the S+ threshold (Satisfactory or Highly Satisfactory). MS is not S — it means objectives were achieved with moderate shortcomings. The Bank uses the broader MS+ threshold to report 79–88% “satisfactory outcomes” where S+ shows 30–40%. MS cells are shown in light red. No project in this portfolio is coloured green.
“Quite Possible the Same Results Could Have Been Achieved Without the Project”
The most recent Unsatisfactory — and the most damning. The Ghana PPP Project: a $30 million operation to build the legal and institutional framework for public-private partnerships. The centrepiece: a PPP law. It was submitted to Cabinet four times — in 2014, 2015, 2017, and 2018. It was never enacted. Parliament was dissolved before it could vote. The Ghana Ports and Harbour Authority unilaterally terminated competitive bidding for the Takoradi Port terminal. Multiple TTL rotations contributed to what IEG called supervision inefficiency.
IEG rated efficiency “Negligible.” M&E quality “Negligible.” Then IEG upgraded the Bank’s own Moderately Unsatisfactory self-assessment to Unsatisfactory — because “the ratings in the ICR do not match the narrative and the limited evidence provided.”
The Jobs Machine That Did Not Measure Jobs
The FCI Global Practice exists to create jobs. Approximately 300,000 young Ghanaians enter the labour market each year. Youth unemployment reached 32.5 percent in 2025 — 49.3 percent in Greater Accra. Some 1.34 million young people are not in education, employment, or training. Among those who work, 68 percent are in vulnerable employment. The government has described youth unemployment as a national security risk.
The fourteen projects carried titles explicitly oriented toward employment: ‘Private Sector Development,’ ‘Enterprise Development,’ ‘Competitiveness and Job Creation.’ IEG’s evaluations do not record employment outcomes because M&E systems rarely tracked them. The lessons database mentions employment in only a handful of cases. The jobs machine did not measure whether it created jobs.
Ghana’s non-traditional exports doubled during one of the failed projects — raising counterfactual questions about whether the Bank was needed for the outcomes it claimed to pursue.
The Modality That Worked Was the One the Bank Used Least
IEG’s Ghana CLRV (FY13–18) contains a finding that deserves attention. IFC advisory work on Ghana’s credit reporting system — targeted, specific, market infrastructure — exceeded its MSME lending targets by more than seven times. It was rated successful. A separate IFC SME initiative was rated “highly unsuccessful.”
Targeted market infrastructure advisory — credit bureaux, collateral registries, information systems — can succeed where large sovereign-led reform projects struggle. The modality that worked was the one the Bank used least. The one it scaled was the one that consistently fell short.
After Fourteen Projects: A $500 Million Successor
The Bank’s FY22–26 CPF reflects genuinely more sophisticated diagnostic thinking — correctly identifying infrastructure bottlenecks, governance failures, and political economy constraints. The CPF even states that “the inability of the Government to effect real reform in the energy sector previously undermined the ability of the World Bank to provide critical support.” This is diagnostic learning.
Yet the operational response remains: a $500 million Market Access and Connectivity Project — cross-sectoral connectivity and logistics, led by the Ministry of Roads and Highways. Genuine strategic evolution. But the SSA transport portfolio returns 10.8 percent Satisfactory across 65 projects and $11.2 billion, with 96 percent of resources below the S+ threshold. Ghana’s disbursement ratio collapsed to 2.81 percent in FY2026.
2. Radical simplification. If a project cannot be explained operationally in one page, it is too complex.
3. Geographic concentration. Corridors, cities, export zones. Build functioning economic geography first.
4. Implementation readiness thresholds. No $500M operation without demonstrated capability at smaller scale.
5. Mandatory institutional memory. Lessons-learned gates in the approval process.
6. Measure what matters. Track employment generation, firm creation, productivity change.
7. Finance continuity, not episodes. State capability accumulates through repetition, not episodic interventions.
Series: Does the World Bank Learn from Project Failures?
| Part | Country–Sector | Commitment | Status |
|---|---|---|---|
| 1 | Nigeria Water | $1.8bn | Published |
| 2 | Angola DPF | $2.2bn | Published |
| 3 | South Africa Energy (Eskom) | $9.13bn | Published |
| 4 | Ghana FCI | $615M | This paper |
Companion Papers
The FCI Sector Record — 67 projects, $4.4 billion, 36% S+ across Sub-Saharan Africa. Ghana is the worst single-country record.
The Transport Record — 65 projects, $11.2 billion, 10.8% S+. The foundation that does not hold.
Why the System Does Not Learn: A Game Theory Analysis — The Nash equilibrium of the approval culture.