MDB Results 2026: An Independent Assessment — Day 2, Part 1 of 5
Part 1: The Ship That Does Not Keep Score
The IMF is two buildings. The old one — solid, heavy, the colour of old government, built in an era when institutions were supposed to look permanent and menacing. Exactly the same shape and colour as the FBI building down the street. Ministers tremble when they enter — they are supposed to. Unlike the other building that has a printing press, this one has gold in its basement. It gives it out when it is told to, or in an emergency, if you are foolish enough to take up its offer to help yourself — because when the repayment time comes, you will have no option.
This building is manned by centurions — each in his own window — looking down. Armed with his iPhone and laptop, ready to produce a report at the push of a button. You enter at your own risk. This is not a cruise ship. It is a battleship ready for action at any time – wherever the ship’s owners decide to send it.
To appear more transparent, a new building was constructed adjacent to it. A gleaming glass tower with a nice cafeteria on the first floor, through which you can clearly see the actual centre of power. But every developing country minister knows — every single one of them who has entered this glass building — what you see is a mirage. Things happen here that would shock a magician.
Kofi crossed 19th street and entered the citadel with his genie.
They asked the receptionist for the Annual Report. They were provided it. The genie said “I already have this”. They turned and turned the pages. There were no numbers in it. None. None since 1944.
The lobby was quieter than the World Bank. More purposeful. Less atrium. The passengers here moved with a different kind of urgency — the urgency of macroeconomics. Exchange rates. Debt trajectories. Balance of payments. The urgency of finance ministers calling from crisis countries at two in the morning.
On an upper floor, Kofi turned to the genie since he had a pressing question.
“I thought I saw a genie on the top floor of the last ship,” said Kofi.
“That is a common mistake. But I am the same genie. I live in all the ships. I am made of their reports.”
“Then show me the numbers. Like yesterday.”
The genie’s pages turned. And turned. And turned.
“There are no numbers,” said the genie.
Kofi stared.
“What do you mean there are no numbers?”
“I mean that this ship does not keep score.”
“Show me the table. How many programmes worked. How many did not.”
“There is no table.”
“Show me the decline. Like the World Bank — 81 percent in the 1970s, down to 26 percent by 2009.”
“There is no decline to show. Because there is no starting point. Because there has never been a number.”
Kofi sat down in the corridor. A very serious person walked around him.
“Since when?” Kofi asked.
“Since 1944. In eighty years of lending — Stand-By Arrangements, Extended Fund Facilities, Rapid Financing Instruments — the IMF has never assigned a systematic outcome rating to a single lending programme. Not one. The question ‘did this programme achieve what it was supposed to achieve’ has never been answered with a number that appears in a public database.”
Kofi thought about yesterday. The World Bank’s number was 47 percent. It was not a great number — but it was a number. Someone had been required to produce it.
“Why?” Kofi asked.
“I will tell you. But first let me show you what exists instead.”
“This ship has three things it uses instead of ratings. The first is called an Article IV. Every year the IMF sends a team to every country and writes a report about that country’s economy. It is published. It is serious. It is read.”
“Does it say whether the IMF’s own programmes worked?”
“It does not. An Article IV written one year after a failed programme analyses the same numbers as one written in a country that never had a programme. The language is calibrated carefully. ‘Fiscal consolidation proceeded broadly in line with programme targets.’ ‘Structural reform implementation faced some delays.’ It never says: this programme failed.”
“The second thing?”
“Ex-Post Assessments. For countries that borrow repeatedly — a narrative review. Not a rating. Written by the same staff who designed and ran the programme. No independent validation. No comparable score.”
“The third?”
“The Independent Evaluation Office. Established in 2001, after a decade of Board resistance. It produces genuine, serious, independent analyses. But it evaluates themes — patterns across programmes. It does not rate the Ghana programme individually. In the COVID disbursements, the countries that had the most problematic disbursements were not on the shortlist selected for case studies. The individual programme generates no rating. No record of whether it worked.”
“So the failures disappear,” said Kofi.
“The failures become lessons. Generalised. Attributed to patterns. Never to a specific programme. Never to a specific mission chief. The IEO found the Argentina 2001 programme had systematic design failures. That finding exists in a thematic report. It does not appear as a rating of Unsatisfactory on the Argentina programme record. The programme has no record.”
“But the IMF rates the technical assistance.”
“Yes. When the IMF sends advisors to help a country improve its tax administration or reform its budget office — those activities are rated. The system is called TAEP. It exists. It works.”
He spread a table on the corridor floor.
| IMF Technical Assistance (TAEP-rated) | IMF Lending Programmes (not rated) | |
|---|---|---|
| Annual volume | ~$400–500M in TA expenditure | ~$30–280bn in lending annually |
| Typical exposure | A PFM advisory mission: $50,000–200,000 | A single SBA or ECF: $500M–$10bn+ |
| Sovereign debt consequence | None — TA is grant-funded | Direct: adds to sovereign debt for years |
| Programme completion document | TAEP evaluation report with outcome rating | Does not exist. Not in 80 years. |
| Independent rating | Yes — including IEO oversight | No. IEO evaluates themes only. |
| Public database | TAEP reports published | Does not exist. |
| Accountability consequence | Finding informs future TA design | None. Programme failure leaves no record. |
“So they know how to build a rating system. They built one. It rates PFM advisory missions worth a hundred thousand dollars. And the same institution, with the same people, has never applied that system to a programme worth ten billion dollars.”
“Correct.”
“How much is the technical assistance worth?”
“The annual TA budget is approximately 400 to 500 million dollars.”
“And the lending in 2020 and 2021?”
“Approximately 280 billion dollars.”
Kofi did the arithmetic.
“So the IMF rates the 400 million. And does not rate the 280 billion.”
“A PFM advisory mission worth 100,000 dollars is evaluated more rigorously than a 10 billion dollar balance of payments programme. The institution that holds governments accountable for fiscal transparency generates less transparency about its own programme outcomes than any other major lending institution in the world.”
Kofi stood up from the corridor floor.
“Why has nobody fixed this?” he asked.
“Five reasons. But the most important one I will tell you now. The wonderful people in this building — and they are wonderful people. Brilliant economists. People who came here because they wanted to help. Their entire incentive structure points toward one thing: getting the programme approved, getting the money out, moving to the next country. The mission chief who gets the programme approved and the tranche disbursed gets a good performance review. Nobody’s career at this institution has ever depended on whether the programme they designed, in the country they left, achieved what it was supposed to achieve.”
“It is the same as the cruise ship,” said Kofi. “Everybody is busy making sure the money leaves.”
“At the World Bank they call it disbursement. Here they call it programme support. The outcome is the same.”
The genie was quiet for a moment.
“Let us walk to the Africa Department and let me tell you about a country that received 3.4 billion dollars in emergency financing in April 2020. I will tell you how many people had died of COVID in that country at the moment the money was approved.”
“What country?” Kofi asked.
“Nigeria,” said the genie. “Your mother’s country.”
Kofi was very still for a moment.
“Let’s go. I want to hear the story so I can share it with my mother,” he said.
“Great,” said the genie.
Part 1: The Ship That Does Not Keep Score (this piece)
Part 2: Nigeria and the Africa-Wide Emergency: $280 Billion and Zero Ratings
Part 3: The Cold War Across 19th Street
Part 4: The Homework the IMF Does Rate: AFRITAC and the Score That Has Not Moved
Part 5: Europe’s Chair
Full series at mdbreform.com and mdbreform.substack.com