Auctions and competitions

Reference values: how the auctions work#

The table compares the rules as each source states them. Laws may have been amended since the version read; the date of each version is given.

Ethiopia: from the ECX mandate to the 2017 reform#

According to a 2021 report of the International Institute for Environment and Development (IIED), the ECX was launched to improve agricultural marketing.¹⁰ When coffee trading through the exchange became mandatory is dated differently (see Contested).^9,10

The IIED report names the loss of traceability of individual lots as a side effect of the mandate.¹⁰ The same report finds that the cooperatives’ share of coffee exports did not rise much after the mandate.¹⁰

The 2017 reform was enacted through Proclamations 1050/2017 and 1051/2017.^10,12 Proclamation 1051/2017 was published in the Federal Negarit Gazette on 5 June 2017.¹² A USDA report of May 2018 states that it allowed washing stations, mills and farmers with more than 2 ha of coffee (at least 1 t/ha) to export directly without the ECX.¹¹ Before that change, commercial farms and cooperative unions could already export directly.¹¹ The IIED report describes the result as traceability restored through “vertical integration”.¹⁰ According to the FAOLEX record, buying or selling coffee outside the ECX or other transaction centres set up by the authority remains an offence under Proclamation 1051/2017; the record lists the law as not repealed.¹² Later amendments, if any, were not consulted for this entry.

Whether the exchange linked Ethiopian prices more closely to world prices is not settled. According to a 2015 conference paper, not all coffee regions had become more connected to world markets in the first years after the mandate.⁹

Cup of Excellence rules in detail#

The rules page describes six rounds of cupping.¹ Samples with major defects are eliminated at once.¹

The page also keeps an older wording. It speaks of up to 35 judges and of a minimum of 87 points in the final round.¹ The thresholds were not always this high. In the Honduras auctions of 2005 and 2008, lots that scored below 85 were sold, and in 2012, 2015 and 2018 lots below 87.⁴ When each threshold changed was not stated on the pages consulted.

Median, record and the score–price link#

The organiser’s results allow a simple check of how far the top lot lies from the rest (our calculation from the listed prices).^4,5

Lots include split a/b lots; medians are our calculation. The 2025 Honduras auction pooled three variety categories, and Brazil 2023 three process categories.^4,5 Over these eight Honduras auctions the median rose more than tenfold, from USD 1.60 to 21.60 per lb in nominal terms (our calculation); no adjustment for inflation or for the futures price was made.⁴

Price does not rise strictly with score. In the Brazil 2023 natural (“dry”) category, the coffee ranked second at 90.12 points sold for USD 19.00 per lb, while the one ranked sixth at 88.21 sold for USD 28.40.⁵

What studies find about quality premiums#

A peer-reviewed analysis of Cup of Excellence e-auctions in Central and South America found that clearing prices depend on sensory traits, a third-party quality ranking, country of origin, variety and lot size, and that they were well above commodity prices.¹⁴ In its working-paper version, larger lots sold for less per pound, first place carried the largest premium, and specialty prices moved with the commodity price.¹⁵ The size of these premiums is not given here: the journal text was not open for this entry, and the working paper alone carries no number.

A 2018 peer-reviewed study of the same programme separated “material” attributes, such as cupping notes, from “symbolic” ones, such as origin, variety, processing method and certification. Symbolic attributes explained more of the scores than material ones, and prices were mainly determined by symbolic attributes. Prices also depended on auction conditions such as the number of coffees offered, and first place carried the largest price premium.¹⁶ A later working paper by one of its authors found that, for some ranks, a higher score did not raise the price significantly; it ties the rank premium to how buyers read rankings.¹⁷

No study consulted for this entry measured prices on the Nairobi or Moshi auctions against quality, or how much of an auction price reaches the producer in Kenya or Tanzania. Shares of value along the chain are treated in Who earns what. No study by Teuber on competition prices was consulted for this entry.

Open questions#

• Spill-over to non-winning lots. No study consulted here measured whether a competition raises prices for a region’s ordinary lots in later years.
• Real price trend. The Honduras series is nominal; an inflation-adjusted series, or one set against the futures price at each auction date, was not found in the sources consulted.
• Payment speed and producer share. The rules give payment deadlines but not the share of the final price the grower receives at each national auction.
• ECX after 2017. Later amendments to Proclamation 1051/2017 and the volume still traded through the ECX were not covered by the sources consulted.

Key studies#