Part 6: The Coffee Year Closed. The Account Did Not.

 POOR FARMER  |  COFFEE POLITICS

A Record Is Not Enough

Ethiopia's coffee windfall, and what reached the farm gate

By Wondwossen Mezlekia

A postscript to the series, six weeks after Part 5.

 

Part 6: The Coffee Year Closed. The Account Did Not.

Published Friday, August 21, 2026

When Part 5 ran on July 7, Ethiopia's fiscal year had closed and the Prime Minister had told Parliament that coffee exports earned 3.1 billion dollars.

One question was deliberately left open: what would the written close-of-year account show?

The Ethiopian Coffee and Tea Authority had already offered an explanation for the record. Better grades, more specialty coffee, direct sales, new markets and tighter control of diversion had allowed Ethiopia to earn more even as the international reference price fell. I wrote then that the final account should show how much of the result came from price, how much from volume, how much from quality mix, and how much from stock drawdown and coffee carried over from earlier harvests.

More than six weeks later, I still have not found that account.

A volume figure has surfaced, but not where anyone would think to look.

On August 19, Ethiopia Today reported that the Authority had fined exporters five percent of revenue for holding coffee off the market rather than supplying it during the harvest year. In the course of that report, it stated that Ethiopia closed the year with exports of about 430,000 metric tons worth 3.1 billion dollars.

I have not found the figure of 430,000 metric tons in any other published source, official or otherwise.

It may be right. But it is not the close-of-year account Part 5 was waiting for, and there is still no public table against which to check it.

The arithmetic, and what it cannot settle

The previous year's official record was 468,967 metric tons and 2.65 billion dollars. If the new figures are measured on the same basis, export volume fell by roughly eight percent while the average export value rose from about 5.65 dollars a kilogram to 7.21, an increase of roughly 28 percent.

That would explain how Ethiopia earned substantially more while shipping less.

What it does not do is prove the Authority's explanation. A rise of that size could also come from a stronger world arabica market. Without the grade composition and destination mix, there is no way to tell how much came from better coffee and how much came from a higher market. Both can be true.

The hoarding story raises another question: timing. Coffee held back in one period can be sold in another. If stocks cross reporting periods, an annual total cannot tell the difference between a timing change and a change in current production.

When did the coffee move? How did the value change through the year? Which grades and origins accounted for the increase? How much came out of earlier stocks? How does each month compare with the same month a year before? A yearly total answers none of it. One published table of monthly export volume and value would answer most of it.

The National Bank of Ethiopia had already pointed at the same puzzle months earlier. Its March 31 Monetary Policy Committee release said coffee export volume had declined compared with the same period a year earlier. In the same release, the Bank reported significant improvement in exports, particularly coffee and gold, during the first eight months of 2025/26. It gave no coffee volume or value figures to quantify either statement.

That is a small data problem. It should also be a cheap one to solve.

The number still missing

But the larger problem this series uncovered was never the difference between 430,000 and 469,000 metric tons.

It was the number that could not be put on any of the charts.

What was the farmer paid?

Part 5 put the old claim that farmers received roughly 40 percent of coffee-sector revenue against the more recent claim that reform had raised the farmer share to 80 percent. Neither arrived with a published calculation showing the numerator, the denominator, regional differences, marketing channel or production costs.

There have been partial attempts. The Enveritas sustainability map publishes field-level indicators for Ethiopian coffee, and farm-gate price is among them. It appears as a category rather than a value. A category can describe a distribution. It cannot be reconciled against an export receipt, compared month to month, or used to calculate a benchmark.

So the series ended where it began. A record at the port is not a farm-gate account.

That is why the argument did not end with Part 5.

On July 18, Addis Fortune published The Missing Half of Ethiopia’s Coffee Plan. It argued that a national Coffee Fund should be designed around the grower, not around the financing difficulties of suppliers and exporters. The fuller framework is in my working paper, Beyond the Coffee Record: Farm-Gate Accounting and Income Stabilization in Ethiopia’s Coffee Sector, dated July 19 and later distributed through SSRN.

The design is deliberately narrow. Ethiopia needs a recurring regional farm-gate price series, a public estimate of what it costs a representative grower to stay in coffee, verified production and transactions, and payment made directly to eligible growers. One window would support approved rehabilitation and replanting, when a household has to survive while the trees recover. A second would provide temporary support when the regional farm-gate price falls below a published benchmark. Payment stops when the trigger disappears.

The institution that calculates the benchmark should not approve claims. The verifier should not move the money. Payments should reach growers directly. The arithmetic should be reproducible from outside government.

That is not a national coffee price and it is not a guaranteed profit. It is a way of keeping a temporary market or rehabilitation shock from becoming a permanent exit from coffee.

Two other pieces followed the same question. In African Arguments, Ethiopia Counts Coffee at the Wrong End asks what a record export year means to a household deciding whether coffee still earns enough, and often enough, to keep the land. In Hararghe, that decision is already visible in the movement from coffee to khat.

That movement is not new, and this blog has the receipts. In October 2009 I quoted Addis Fortune reporting that farmers with access to transport were dropping coffee in favor of khat, and a former grower in Harari who said export-quality khat brought him 300 to 500 Birr a kilogram while export-quality coffee brought him 25 to 35 Birr. Seventeen years later, a 2025 study of 788 households in East and West Hararghe found that 64 percent had made the same switch. The substitution was documented, then measured. What was never published, in 2009 or in 2025, is the farm-gate price that would have let anyone see it coming.

A second Addis Fortune commentary, Europe Needs the Plot, Ethiopia Needs the Price, asks whether the traceability system Ethiopia is already building for the European Union’s deforestation regulation could also record the first producer transaction. Europe requires Ethiopia to know where the coffee was grown. Ethiopia can use the same architecture to record what the farmer was paid. The missing farm-gate ledger may not require another national information system. It may require adding a small economic record to one already under construction.

Put the questions in writing

It is also why I rebuilt Coffee Monitor.

The point is not to add another set of claims about Ethiopian coffee. It is to keep apart five things that are routinely treated as though they were one: what no one publishes, what institutions collect but do not release, what officials state, what can be calculated from published figures, and what is published as a recurring series with definitions and a stated method.

Those are not the same thing. An announcement is evidence, and it may be authoritative evidence. A public series with definitions and a method can be checked, compared over time, corrected and used to design policy.

The standing questions there are deliberately narrow. A question belongs on that page only when a document, table or definition could close it.

One of those questions is now in writing. On August 17, I asked the National Bank of Ethiopia for the monthly coffee export volume and value figures underlying the 2025/26 record, together with the units, definitions and any revisions. If the table is already public and I have missed it, one retrievable citation closes the question and corrects the record.

That is how this should work. I am not trying to prove that an institution has no data. It may have the data and still not have a public record. The question is what someone outside the institution can retrieve, check and use.

The export question should be the easy one. The hard one remains at the other end of the chain. Even if the reported export volume of 430,000 metric tons is confirmed, even if the 3.1 billion dollars is reconciled exactly, and even if a monthly table shows precisely how Ethiopia earned more from less coffee, none of it tells us what growers in Jimma, Sidama, Guji, Hararghe or Yirgacheffe were paid. That requires the account that still does not exist publicly: a recurring farm-gate price series under a stated method.

Part 1 began with a record and asked what stood behind it. The coffee year has closed. The questions are more precise now. Some can be answered with a table. Others require building the record that should have existed all along.

The standard has not changed.

The port ledger tells us what the world paid.

The farm-gate ledger tells us whether the people who grew it can afford to grow it again.

Until both can be read, Ethiopia's record coffee year remains what it was at the beginning of this series: a record at the port, not an account at the farm gate.

Sources and related work

Part 5: What the Farmer Keeps, Poor Farmer, July 7, 2026: https://poorfarmer.blogspot.com/2026/07/part-5-what-farmer-keeps.html

Window dressing of Ethiopia’s coffee exchange, October 12, 2009: https://ethiopianreview.com/content/11037

Ethiopia Fines Coffee Exporters Over Unlawful Hoarding, Ethiopia Today, August 19, 2026: https://ethiopiatoday.net/ethiopia-fines-coffee-exporters-over-unlawful-hoarding/

National Bank of Ethiopia, Monetary Policy Committee Meeting No. 6, March 31, 2026: https://nbe.gov.et/nbe_news/press-release-monetary-policy-committee-meeting-no-6/

The Missing Half of Ethiopia’s Coffee Plan, Addis Fortune, July 18, 2026: https://addisfortune.news/the-missing-half-of-ethiopias-coffee-plan

Ethiopia Counts Coffee at the Wrong End, African Arguments, August 14, 2026: https://africanarguments.org/2026/08/ethiopia-counts-coffee-at-the-wrong-end/

Europe Needs the Plot, Ethiopia Needs the Price, Addis Fortune, August 14, 2026: https://addisfortune.news/europe-needs-the-plot-ethiopia-needs-the-price

Ethiopian Coffee Map, Enveritas: https://ethiopia-sustainability.enveritas.org/

Beyond the Coffee Record, policy brief: https://coffeemonitor.com/files/Beyond-the-Coffee-Record-Policy-Brief.pdf

Beyond the Coffee Record, working paper on SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7142558

Coffee Monitor: https://coffeemonitor.com/

Standing Questions: https://coffeemonitor.com/questions

 

Poor Farmer  |  Coffee Politics  |  poorfarmer.blogspot.com