Part 5: What the farmer keeps

A Record Is Not Enough

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

By Wondwossen Mezlekia

The fifth and final post in the series.

 

Part 5: What the farmer keeps

The fiscal year has closed, and the Prime Minister says Ethiopia earned 3.1 billion dollars from coffee exports in 2025/26, up from 2.65 billion dollars the year before.

It will be celebrated, and it should be. Ethiopia has announced a five-year plan to raise productivity and push coffee-export earnings toward six billion dollars.

But a record at the port is not a farm-gate account.

That is where this series began, and it is where it ends. Not with the question of whether Ethiopia earned more. It did. The question is whether the grower kept more after the cost of producing, moving, and living through the coffee year.

The world price has already come down from its 2025 peak. Cherry prices rose to levels many washing stations could not finance. Costs remain high. And the one number that would settle the argument, what the smallholder kept after costs, in money that still buys what it bought before the windfall, is still not published.

The clearest sign that a record is not a raise is that washing stations closed during it. Cherry reportedly reached 220 to 250 birr a kilogram in parts of the 2025/26 harvest, about three times the previous season. On paper, a farmer's victory. In practice, the cash to buy cherry rose faster than many washing stations and exporters could finance, and reporting from the harvest described buyers holding back purchases and more farmers drying cherry at home for later sale. A high price is income only when someone can pay it, and the money to move the coffee was the scarcest thing in the chain.

There is a figure from inside the Authority that deserves to be remembered, even if its calculation was never published for public examination. In comments reported by Addis Fortune, Hairu Nuru, then the Authority's Market Information and Regulatory Director, said farmers received only about 40 percent of the total revenue generated by the coffee sector, with the remainder taken by collectors, suppliers, traders, wholesalers, the Exchange, and exporters.

That was not a current farm-gate survey, and I have not found the underlying calculation. It was an official's estimate of how the chain worked at the time. But the fact that such a figure could be stated without a public series behind it is part of the problem. If farmers receive forty percent, or sixty percent, or more than that today, the Authority should be able to show the country the calculation.

To be fair, the Authority has since made a very different claim. In 2025, it told the United States Department of Agriculture that the farmer share had risen from forty percent to eighty percent over five years as reforms shortened the chain and reduced reliance on intermediaries.

That would be important progress. But the calculation has not been published. The country has not been shown what was counted as the farmer's share, which price was used as the denominator, how the number differs by region, grade, or marketing channel, or whether it holds after the cost of producing coffee is deducted.

So the old forty percent and the newer eighty percent leave us at the same place: with a number, but no public ledger behind it.

Until that ledger exists, a record at the port tells us far more about exports than about the farmer's share.

The record was never a smooth climb, either. The ICO composite that averaged 354.32 cents in February 2025 was back to 273.70 by March 2026, 256.05 by May, and around 248 by June. The ICO has suggested that part of the 2024/25 export surge may have been supported by an unusually large drawdown of stocks as exporters took advantage of high world prices. Some of the growth, then, may have been coffee from an earlier harvest sold into a better market. That money was real. It does not recur simply because the headline does.

Put the pieces on one chart and the broad pattern is hard to miss. Through 2024/25, Ethiopia's export earnings moved closely with the world price. The cost of living the farmer carries climbed through the same period and did not come back down. In the lean years, it ran ahead of the sector's own earnings.

The 2025/26 result is the claimed break in that pattern. The Authority says better quality, direct sales, grade selection, market diversification, and tighter control of diversion allowed Ethiopia to earn more even as the New York reference price fell. The final account should show how much of that break came from price, volume, quality mix, stock drawdown, and the sale of coffee harvested in earlier years.

The price boom lifted export earnings; the farmer's costs stayed up

The one line that would settle the argument is the line missing from that chart, what the smallholder kept after costs, in money that still buys what it bought five years ago. It is absent because no one publishes it. A picture with a clean hole where the answer belongs is, in its way, the most honest one I can give you.

The Authority has an explanation for the break in the line, and it deserves a fair hearing. At the press conference announcing the record, its Director General, Adugna Debela, said Ethiopia earned it against a falling market, not because the market carried it there. He pointed to the New York price falling from around four dollars a pound a year earlier to roughly two dollars and forty to two dollars and fifty now. At the same time, he said Ethiopian coffee that had sold near six thousand dollars a tonne the previous year was selling near seventy-five hundred.

He attributed the difference to a move away from commercial and under-grade coffee into specialty and grades one through three; to direct quality-negotiated sales rather than sales tied only to the New York reference; to new buyers in more countries; and to tighter control of diversion and shipment tracking. If the final account supports those claims, it is a real accomplishment. Ethiopia should say so plainly.

But the explanation makes another absence harder to ignore. The Authority can tell the country how it improved the export side: the grades, the markets, the logistics, the buyers, the export price, and the foreign exchange earned. What it still does not publish is the matching account at the farm gate: what the kilo of cherry brought in each producing region, what it cost to grow, what share of the export gain reached the household, and whether that household was better off after costs.

Port accounting and farm-gate accounting are not the same thing. One does not automatically produce the other. But once the export side can be described in this detail, the absence of a regular farm-gate series is no longer a minor omission. It is the number needed to tell the country what the record meant.

Volatility, not the record, is the permanent condition. Colombia is not Ethiopia's twin, but it offers a useful lesson because it built institutions around the fact that coffee growers cannot wait for the market to recover.

The first is a purchase guarantee. Through the growers' federation and its cooperative network, a Colombian producer has a buyer available at a published, market-based price. It does not guarantee a high price. It guarantees that coffee is not left sitting with the farmer because the buyer has run out of cash.

The second is separate: Colombia's Coffee Price Stabilization Fund. It was created to protect producer income when coffee prices fall to exceptionally low levels and the margin turns against the grower. It is not a promise of a permanently high price. It is a mechanism meant to keep a price shock from becoming a household collapse.

Ethiopia does not yet have an equivalent national purchase guarantee or a published coffee-income stabilization mechanism. It has rehabilitation programs, restoration finance, traceability work, and credit for parts of the value chain. Those matter. But they are not the same thing as a buyer with cash at harvest, or an automatic payment when the farm-gate price falls below the cost of staying in coffee.

That is the gap the next reform should fill.

Call it a Coffee Income Stabilization Fund. Build it first from a small, transparent contribution during strong export years, with public and development-partner support available when a severe shock exceeds what the reserve can carry. The contribution must not be passed back to farmers through lower farm-gate prices.

Trigger it only when a published regional farm-gate price falls below a verified regional cost benchmark. Pay registered smallholders directly into verified bank or mobile-money accounts after a legal sale has been confirmed. Cooperatives, washing stations, and licensed buyers may verify the transaction, but they should not control the payment. Publish every trigger, every payment, every region, and every beneficiary category.

Do not use it to subsidize exporters. Do not turn it into another discretionary fund for people with access. Its purpose is narrower: to keep a smallholder from selling cherry in distress, abandoning coffee, or sliding into debt because a price set in New York or London has turned against a household in Guji.

A reserve like this would not eliminate fiscal risk. It would limit it. It would accumulate in stronger years, draw down under published rules in weaker ones, and make clear when a shock is large enough to require additional support.

Ethiopia defaulted on its Eurobond in December 2023 and remains in debt restructuring, although it reached a preliminary bondholder deal on June 29, 2026. That makes transparent, limited, rules-based design essential, not a reason to avoid building a coffee buffer.

None of this is charity, and it is not a subsidy in the lazy sense in which the word is usually thrown around. It is income protection for one of the largest non-state systems of work in the country. When the livelihoods of some fifteen million people depend directly or indirectly on coffee, and most production begins on smallholder farms, a fund that steadies their income when the world price collapses is not a handout to a sector. It is a floor under the largest body of rural producers Ethiopia has. The question is not whether the country can afford to build it. It is whether it can afford to leave that many households exposed to a price it does not set, in a year when that price has already begun to fall.

The Authority is building traceability, and that is good. Traceability tells the state where the coffee came from. A stabilization fund would answer the harder question, the one that surfaces when the market turns, which is whether the person who grew it can stay on the land. And it would force the missing number into daylight, because a fund cannot pay below a cost benchmark that has never been published. The farm-gate series I asked for in the first post is not a separate transparency request. It is the trigger. One reform protects the household, the other proves whether it was protected, and neither works without the figure the Authority still has not published regularly.

The next barrier is already being built, and it has two names that should not be confused. From December 30, 2026, the European Union's deforestation regulation will begin applying to large and medium operators handling coffee, with most micro and small operators following in June 2027. For Ethiopian coffee entering that supply chain, the rule requires traceability to the plot where it grew and due diligence showing compliance.

It is not a rule that certifies each farmer. But the cost of mapping land and proving compliance can still be pushed down the chain to the smallest producer. Separate organic rules can place a different burden on cooperatives. Different rules, same danger: the grower who cannot afford the paperwork is left outside the market. The Authority's national traceability platform, which has already mapped hundreds of thousands of plots, may be the most important infrastructure the sector has built. But mapping is not financing. Unless the cost is shared, traceability becomes one more advantage for whoever already holds the capital.

I have stood on the winning side of a famous coffee victory before. In 2006 and 2007 I was among those who argued that Ethiopia should own the names of its great coffees, Sidamo, Yirgacheffe, Harar, against a company that did not want it to. We won. The names are Ethiopia's, and I am in the record of that fight. Then I watched the part I had not braced for. Owning the name changed less than we had promised, because the links meant to carry the value back down the mountain to the farmer were weak, and a trademark does nothing to strengthen them. The legal right was necessary. It was not sufficient. I learned, in public and a little painfully, that winning the argument and changing the household are two different jobs.

There was an older question underneath that one, and the trademark fight never reached it. Owning a name settles who holds the right. It does not settle whether the people whose labor gives the name its value have any claim on the value built around it. We answered the first question in 2007 and left the second untouched. The record answers it the same way. It proves the name sells. It says nothing about whether the labor behind it shares in what the name earns.

The 3.1 billion dollars is real, and it proves the world will pay for Ethiopian coffee. It does not yet show the grower keeping more of it. The evidence available to the public points to a harder picture: washing stations short of finance, costs that remain high, premium coffee sold at home for foreign currency, new compliance burdens, and farm-gate figures the government still does not publish.

The country is now promising six billion dollars on the strength of a better tree, in a market whose price it does not set. I hope it reaches it. But six billion earned is not six billion kept. A larger record only makes the question larger: who kept it? I will continue counting at the farm gate, the only place the number has ever mattered. I would be glad to be proven wrong in the right direction.

Source note: International Coffee Organization, Coffee Market Reports | Coffee livelihood estimates (about fifteen million livelihoods linked to coffee across the chain; about 5.9 million coffee farmers; smallholders produce roughly 90 percent of output) from Royal Botanic Gardens Kew, USDA, and peer-reviewed studies | Bezawit Huluager, "Authority Relaxes Coffee Export Restrictions," Addis Fortune, posted November 2020 and shown as updated August 9, 2025, quoting Hairu Nuru, then Market Information and Regulatory Director of the Ethiopian Coffee and Tea Authority, on farmers receiving about 40 percent of sector revenue (methodology not published) | USDA FAS, Ethiopia: Coffee Annual (2025, ET2025-0014), carrying ECTA's claim that the share of export price received by farmers doubled over five years, from 40 to 80 percent, on reforms enabling direct market access and reducing intermediaries (methodology not published) | Federacion Nacional de Cafeteros de Colombia, on the coffee purchase guarantee and the National Coffee Fund (FoNC), and the separate Coffee Price Stabilization Fund (FEPC), Law 1969 of 2019 | EU Environment, EUDR FAQs (April 2026) and European Commission, EUDR application dates | USDA FAS, Ethiopia: Coffee Annual (May 2026) | Adugna Debela (PhD), Director General, Ethiopian Coffee and Tea Authority, press briefing on the 2025/26 result, Ethiopian News Agency video, July 2, 2026