Coffee Monitor: Ethiopia
Independent evidence on Ethiopia's coffee economy · Seattle, Washington
Research
Each paper carries a version number and a revision date. Superseded versions stay reachable. If a figure changes, the change is shown rather than swallowed.
The Observatory is now a research programme as well as a public map. Its operational standard, fixed panel and frozen data releases are published separately so the results can be checked without trusting the interface.
Controlling since 1 September 2026. The rules governing every live country state, including the aggregate-reconstruction test, the external-benchmark rule, the fixed green-coffee export scope, the state one search record and the evidence archive requirements. Summarised on the Method page. Deposited 1 September 2026 at doi.org/10.5281/zenodo.22740331. It supersedes Revisions 2.6 and 2.5 and Addendum 1; the draft numbered 2.6.1 was never operative.
Deposited →A pre-analysis plan for a 51-origin audit of reference-price and farm-gate disclosure. It fixes the questions, the thresholds, the treatment of unresolved cells and the comparative analysis. Deposited at doi.org/10.5281/zenodo.22740569.
The chronology is stated rather than implied: the plan was finalised on 7 September 2026, the registry was frozen on 11 September, and the public deposit followed on 14 September. The deposit therefore came after the freeze, not before it, and that departure is recorded in the plan as Deviation 1. It is not presented as a blind preregistration.
Deposited →A scholarly edition is under back-test and independent validation. Candidate rules and candidate country outcomes do not enter the live map. The controlling version remains Revision 2.7 until a later version is formally adopted.
Validation in progressThe fixed 51-origin panel, the controlling rule book and its hash, the prespecified thresholds and bounds rule, the 2026 cycle figures, and the release verification method. Published so the results can be checked without trusting the interface that displays them.
Open →The farm-gate ledger, the two-window Coffee Income Stabilization Fund, and the case that a record at the port is not a farm-gate account. This September draft supersedes the July text for the purposes of this site. It adds the Ministry of Agriculture's first-party confirmation of 2025/26 export earnings, source-provenance evidence on the widely cited 60 percent farmer-share estimate, and a denominator correction to the rehabilitation programme's 80 percent figure. None of those is treated as a substitute for the missing farm-gate account.
Download →The formally posted version, with the citable DOI. It remains the version of record and is not withdrawn. A consolidated SSRN revision will follow once the outstanding Ethiopian inquiries are answered or formally closed as unanswered, so that the deposited record changes once rather than repeatedly. Cite this one unless you specifically need the September text. ORCID 0009-0009-3189-5920.
Read on SSRN →Ethiopia has built new eyes for its most valuable export. The Coffee and Tea Authority took technical handover of the Ethiopian Coffee Traceability and Management System in March 2026, built to satisfy a European regulation that requires the plot a lot came from to be identified. The question is what the system looks at. A traceability layer that already reaches the plot is the cheapest place a farm-gate price record will ever be built. Published in Addis Fortune (subscription).
Read →The argument made for an international readership. The world's coffee statistics are precise about what leaves the port and silent about what reaches the farm; Ethiopia is the clearest case, and the remedy is a farm-gate record rather than a better export figure. Free to read.
Read →The Ethiopia-specific argument, written for a policy readership: the 3.1 billion dollar export record announced to Parliament says what coffee earned at the border and nothing about what reached the grower, and the Authority's own Coffee Fund will only close that gap if it pays at the farm gate. Published in Addis Fortune (subscription).
Read →The same six-row test applied to a fixed panel of 51 coffee origins. The 2026 research cycle closed on 4 September 2026 with every cell either analytically determined or screened with a named blocker and a stated reopen trigger. The panel is fixed rather than exhaustive: it is the set of origins the Observatory follows, not the set of countries that grow coffee.
Open the Observatory →A pre-analysis plan for the producer-price study, written and registered before the panel results are interpreted, so that the prevalence question is answered against a stated design rather than after the fact. It will be posted here when it carries a registry identifier and timestamp.
Pre-analysis planRevision 2.7 is the controlling text. Later revisions under validation are candidate methodology and are never applied to live determinations; a candidate rule may not change a published state before it is adopted. The Method page carries the public statement of the standard.
Controlling textHow to cite
Mezlekia, Wondwossen. Beyond the Coffee Record: A Framework for Farm-Gate Accounting and Coffee Income Protection. Policy brief, July 2026. coffeemonitor.com/research
Mezlekia, Wondwossen. Beyond the Coffee Record: Farm-Gate Accounting and Income Stabilization in Ethiopia's Coffee Sector. SSRN working paper 7142558, July 2026. https://doi.org/10.2139/ssrn.7142558. ORCID 0009-0009-3189-5920.
To cite the September text specifically, add: revised draft, 12 September 2026, available at coffeemonitor.com/research. The DOI above continues to resolve to the July version of record.
Mezlekia, Wondwossen. "Ethiopia Counts Coffee at the Wrong End." African Arguments, 14 August 2026.
Mezlekia, Wondwossen. "Europe Needs the Plot, Ethiopia Needs the Price." Addis Fortune, 14 August 2026.
Corrections and challenges to any figure in these documents are welcome and are logged in public. See Corrections.