CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Trade · History

The International Coffee Agreements

The treaties between coffee-producing and coffee-importing countries, from Brazil’s 1906 price support to the International Coffee Agreement 2022 – what the export quotas did from 1962 to 1989, and why they ended.

Across the chain: History

Source-checkedby Coffearium verification pass· Last source-checked 24 Oct 2026· Prepared with AI assistance· 17 sources (14 tier A)· 2 contested· Updated 24 Oct 2026· Literature last searched 24 Oct 2026
Reading depth
Contents
  • The first International Coffee Agreement (ICA) was done at New York on 28 September 1962; the International Coffee Organization (ICO) administers the agreements.2,1
  • Under the ICAs of 1962 to 1983, a target price or price band was set and export quotas were allocated to producing countries.8,6
  • Quotas operated in 1963–73, 1980–86 and 1987–89 and ended in July 1989, when members could not agree on market shares.6,3,11
  • The agreements of 1994, 2001 and 2007 have no quotas; the ICA 2007 has been extended until 1 February 2028.3,15
  • A 1990 model study found that quotas stabilised world prices but reduced real export revenue for most small exporters.7

What the agreements are#

The International Coffee Agreements (ICAs) are treaties between governments of coffee-exporting and coffee-importing countries. The first was done at New York on 28 September 1962.2 Its first objective was “to achieve a reasonable balance between supply and demand”.2

Each agreement is administered by the International Coffee Organization (ICO).3,16 Its governing body is the International Coffee Council, where exporting and importing members vote.14,13 A new agreement replaced the old one when it expired. There were agreements in 1962, 1968, 1976, 1983, 1994, 2001 and 2007.1

The agreements fall into two eras. From 1962 to 1989 most of them set export quotas to steer the price.8,11 Since 1989 they have had no such tools.3,8 A political economist calls coffee one of the few international markets “ever subject to effective political regulation”.9

Before 1962: Brazil and the Americas#

The idea started in Brazil. On 25 February 1906 the coffee states of São Paulo, Rio de Janeiro and Minas Gerais signed the Taubaté Convention.4 They agreed to buy and hold coffee stocks, financed by loans, to raise international prices.4 A tax on exported coffee was to repay the loans, and new planting was to be discouraged.4 According to the ICO, this first valorisation scheme raised prices at first but encouraged overproduction.3

The Inter-American Coffee Agreement was signed at Washington on 28 November 1940.5 The United States and 14 coffee-producing American republics took part.5 It shared the US market among the producers through quotas, with the largest for Brazil and Colombia.5,3 Its stated aim was “to promote the orderly marketing of coffee”.5

How the export quotas worked#

Under the quota agreements, members set a target price or a price band for coffee. Each exporting country received an export quota.8 The 1962 agreement fixed basic quotas for its first three coffee years, the first of them beginning on 1 October 1962.2

The quotas were adjusted often, according to where world prices stood in the price range.6 When the ICO indicator price fell below the set price, quotas were tightened; when it rose above, they were relaxed.8 Under the quotas, many producing countries had to hold unsold coffee as stocks.6 Exports to importing countries outside the agreement were not restricted.6

The quotas did not run without a break. They were discontinued in 1973 because producing and consuming countries could not agree on price and quota levels.6 They returned in October 1980, were suspended in February 1986 and were reinstated in October 1987.6

19001920194019601980200020202040YearTaubaté Convention(Brazil)19063,4 Inter-American CoffeeAgreementsigned 19405 ICA 19621963–681 ICA 19681968–761 ICA 19761976–831 ICA 19831983–941 ICA 19941994–20011 ICA 20012001–111 ICA 2007 (extended)2011–281,15 ICA 2022 (adopted)adopted 202214,15 Export quotas, firstperiod1963–736,1 Export quotas, secondperiod1980–866 Export quotas, thirdperiod1987–896,3,11
Data table
PointYearSource
Taubaté Convention (Brazil)19061906.153, 4
Inter-American Coffee Agreementsigned 19401940.95
ICA 19621963–681963.5–1968.751
ICA 19681968–761968.75–1976.751
ICA 19761976–831976.75–1983.751
ICA 19831983–941983.75–1994.751
ICA 19941994–20011994.75–2001.751
ICA 20012001–112001.75–2011.081
ICA 2007 (extended)2011–282011.09–2028.081, 15
ICA 2022 (adopted)adopted 20222022.4414, 15
Export quotas, first period1963–731963.5–19736, 1
Export quotas, second period1980–861980.75–1986.16
Export quotas, third period1987–891987.75–1989.56, 3, 11
Coffee agreements from 1906 to 2028. Bars show the years each ICA was in force, from provisional entry into force to expiry; dots show single dates. The lower rows show when export quotas operated.1,3,4,5,6,14,15

Why the quotas ended in 1989#

The quotas ended in July 1989.3,11 According to the ICO’s own history, disputes over market shares divided members most.3 The 1983 agreement was then extended without its economic clauses.3,1

A 2002 study names several further causes.8 More coffee was sold, at lower prices, to countries outside the agreement. Fixed quotas could not follow US consumers as they moved from soluble coffee to ground coffee, which uses more arabica.8 Roasters feared that competitors would buy cheaper coffee from non-members. And the United States no longer saw the quotas as useful in its Cold War policy in Latin America.8

Prices fell after the quotas ended. In real terms, the average indicator price for 1990–93 was 42% of its average for 1985–88.8 The full price record is in Coffee price history.

The agreements since 1994#

No agreement since 1989 has had export quotas.8,11 According to the ICO’s history, the ICA 1994 lacked any mechanism that could directly affect the market.3 The ICA 2007 aims to strengthen the coffee sector “in a market-based environment”.16 It entered into force on 2 February 2011.1 On 27 March 2025 the Council extended it until 1 February 2028.15

A new text, the ICA 2022, was adopted on 9 June 2022.14 Governments may sign it until 30 April 2027 and ratify it until 31 July 2027.14 It has been adopted but is not yet in force: at the Council session of March 2026, members holding two thirds of the votes of each group had not yet ratified it.17,13

As of 2019, the ICO described its work as statistics, monthly market reports, development projects and dialogue with the private sector.16 Its indicator prices are explained in How a coffee price is made.

Reference values: the agreements and their dates#

Instrument Adopted or signed In force Economic clauses Source
Taubaté Convention (São Paulo, Rio de Janeiro, Minas Gerais) 25 Feb 1906 – Stock purchases on loans, export tax, curbs on planting 4
Inter-American Coffee Agreement 28 Nov 1940, Washington – Quotas for the US market 5
ICA 1962 28 Sep 1962, New York Provisional 1 Jul 1963; definitive 27 Dec 1963; to 30 Sep 1968 Export quotas 2,1
ICA 1968 – Provisional 1 Oct 1968; definitive 30 Dec 1968; extended 1973–76 Quotas until 1973; extended without economic provisions 1,3,6
ICA 1976 – Provisional 1 Oct 1976; definitive 1 Aug 1977; to 30 Sep 1983 Quotas from Oct 1980 1,6
ICA 1983 – Provisional 1 Oct 1983; definitive 11 Sep 1985; extended 1989–94 Quotas to Feb 1986 and Oct 1987–Jul 1989; extended without economic clauses 1,3,6
ICA 1994 – Provisional 1 Oct 1994; definitive 19 May 1995; to 30 Sep 2001 None 1,3
ICA 2001 – Provisional 1 Oct 2001; definitive 17 May 2005; to 1 Feb 2011 None 1,3
ICA 2007 – Definitive 2 Feb 2011; extended until 1 Feb 2028 None 1,15
ICA 2022 9 Jun 2022 (Resolution 476) Not yet in force (Mar 2026) – 14,17

“–” means that no source consulted for this entry gives the value. Extensions were made by Council resolutions; the ICO lists their numbers and periods.1

The Inter-American quotas of 1940#

Article I of the 1940 agreement set basic quotas for the US market in bags of 60 kg. They totalled 15,545,000 bags for the 14 producing countries.5 Brazil received 9,300,000 bags and Colombia 3,150,000.5 That is about 60% and 20% of the total (our calculation). In tonnes, the total is about 933,000 t (our calculation, 60 kg per bag).

Quota mechanics in the 1960s to 1980s#

The 1962 text tied basic export quotas to a list of exporting countries in its Annex A.2 According to the ICO’s history, each quota reflected the country’s average exportable production: production less domestic consumption.3 The same history states that the 1968 agreement added “selectivity”, adjusting quotas by type of coffee, and that the 1976 agreement based quotas partly on past exports and partly on stocks held.3

The quotas applied to exports to member importing countries only. A 1989 World Bank paper calls the rest the non-quota market. It consisted of New Zealand, the Soviet Union and the other centrally planned economies of Eastern Europe except Yugoslavia, and the developing countries except Portugal and Greece.6 Exporting members burdened with large stocks reportedly sold there at discounts of as much as 50% to quota market prices.6 Reliable price data for these sales do not exist, according to the authors.6

The suspension of February 1986 was triggered by a sharp price rise, caused by a forecast reduction in Brazil’s 1986/87 crop after a severe drought in 1985.6 According to the ICO’s history, disagreements over market shares delayed the return of quotas until 1987;3 they were reinstated in October 1987.6

Membership#

Membership changed with each agreement. The ICO keeps a dated record of every member’s ratifications, accessions and withdrawals.12 Two examples from it: Switzerland ratified the 1962 agreement in December 1964.12 The United States left in September 1993, acceded again in February 2005, and withdrew from the ICA 2007 with effect from 27 June 2018.12

No membership count per agreement was taken from the record for this entry, because its table layout does not allow a reliable count. For the ICA 2022, the ICO’s list of 24 March 2026 shows the members that had completed ratification holding 28.3% of exporting votes and 16.70% of importing votes, counted for entry into force.13 At the Council session of the following days, the Executive Director reported 28.3% for exporting members and 15.2% for importing members; two thirds of the votes of each group are required.17 The two ICO documents give different importing shares; neither explains the difference. The European Union, with 66.60% of importing votes, was listed as not having signed.13

The ICA 2022#

The Council adopted the ICA 2022 by Resolution 476.14 Resolution 485 of 27 March 2025 extended the deadline for signature to 30 April 2027 and for deposit of instruments of ratification to 31 July 2027.14

The European Commission summarises the new text as follows. It aims to strengthen the global coffee sector and promote its economically, socially and environmentally sustainable development.15 It updates the balance of voting and contributions and integrates the private sector and civil society into the ICO’s work, while keeping the ICO intergovernmental.15 In September 2025 the Commission proposed that the EU sign it.15 At the Council session of March 2026 the EU delegate reported political approval for signature.17

No source consulted for this entry describes any quota or price mechanism in the ICA 2022.

What the ICO does today#

Under the ICA 2007, the ICO describes itself as an authority on coffee statistics.16 It publishes 12 monthly Coffee Market Reports each coffee year and develops and monitors technical cooperation projects.16 A Private Sector Consultative Board advises the Council, and a Consultative Forum on Coffee Sector Finance discusses finance and risk management.16 Promoting coffee consumption is also among its activities.16

What economists found about the quota era#

Price stability. A 1990 model study simulated the world coffee market with and without quotas. It found that the quota system stabilised world prices.7 The 1989 working paper behind it concluded that the quotas kept a floor under prices between October 1980 and February 1986.6 Its simulations also suggest that 1986 prices would have been 24% higher without the quotas of 1981–85, because stocks held under quotas were released when they were suspended.6

Who gained. In the same model, quotas reduced real export revenue for most small exporting countries, while large producers gained.7 Most small countries still gained through lower risk. A quota suspended briefly during a supply shortfall acted like a buffer stock scheme.7

Volatility after 1989. A 2002 study reports monthly nominal price variability of 14.8% in the final eight calendar years of ICO activity. It almost doubled to 37% in 1990–97, a figure the study takes from a 1998 analysis, and rose to 43% in 1998–2000.8 A footnote states that the calculations use data of the Coffee, Sugar and Cocoa Exchange.8 The study does not define the measure.

01020304050Monthly nominal price variability, %Final eight calendar yearsof ICO activity14.8%8 1990–9737%8 1998–200043%8
Data table
Monthly nominal price variability, %Source
Final eight calendar years of ICO activity14.8%8
1990–9737%8
1998–200043%8
Monthly nominal price variability of coffee before and after the end of quotas, as reported in a 2002 study (the 1990–97 value from a 1998 analysis it cites). The study does not state the formula.8

Shares of income. The same study reports estimates from a 1997 study of the share of total coffee income, defined as consumer spending on coffee.8 Producers kept about 20% in the 1970s and almost 20% from 1980–81 to 1988–89. From 1989–90 to 1994–95 their share was 13%, while the share kept in consuming countries rose from 55% to 78%.8 The study calls this a substantial transfer of resources from producing to consuming countries.8 How value is shared along the chain today is in Who earns what.

0255075100% of total incomeProducers, 1971–80about 20%8 Producers, 1980–81 to1988–89almost 20%8 Producers, 1989–90 to1994–9513%8 Consuming countries,1971–80almost 53%8 Consuming countries,1980–81 to 1988–8955%8 Consuming countries,1989–90 to 1994–9578%8
Data table
% of total incomeSource
Producers, 1971–80about 20%8
Producers, 1980–81 to 1988–89almost 20%8
Producers, 1989–90 to 1994–9513%8
Consuming countries, 1971–80almost 53%8
Consuming countries, 1980–81 to 1988–8955%8
Consuming countries, 1989–90 to 1994–9578%8
Share of total coffee income (consumer spending on coffee) kept by producers and in consuming countries, 1997 estimates as reported in a 2002 study.8

Power in the chain. The 2002 study describes the post-agreement chain as buyer-driven, with roasters strengthening their position.8 It also reports that overall coffee quality decreased as roasters used cheaper coffee.8 A 2005 book calls the coexistence of a coffee boom in consuming countries and a coffee crisis in producing countries the “coffee paradox”: what farmers sell and what consumers buy are becoming increasingly different coffees.10

Politics. A 1997 political economy of the coffee trade traces the origins, operations and collapse of the ICO.9 The 2002 study cites it for the change in US Cold War policy that weakened US support for quotas.8 A 2017 textbook chapter states that the quota system achieved a stable price but led to overproduction, and that after its collapse stocks were released quickly and prices fell below production costs.11

Open questions#

  • Income shares after 1995: the shares above end with 1994–95. No study consulted for this entry repeats that method for later years.8
  • The effect of the ICA 2022: as of October 2026 no source consulted for this entry reports its entry into force; the latest ICO record consulted is of March 2026.13,17
  • Non-member sales: their volume and prices in the 1980s are uncertain; reliable price data do not exist, according to the 1989 World Bank paper.6

Key studies#

Akiyama and Varangis, 1990 – The impact of the ICA on producing countries. A global coffee model with a vintage-capital approach to production simulates the market with and without export quotas. Quotas stabilised world prices; most small exporters lost real export revenue but gained from lower risk, while large producers gained.

The World Bank Economic Review 4(2), 157–173 · doi:10.1093/wber/4.2.157

Ponte, 2002 – The ‘latte revolution’? Traces how the global coffee chain changed after the end of the ICA quota system. Prices became lower and more volatile, the share of income kept in consuming countries rose, and roasters gained power in a buyer-driven chain.

World Development 30(7), 1099–1122 · doi:10.1016/S0305-750X(02)00032-3

Bates, 1997 – Open-economy politics. A political economy of the world coffee trade, based on field work in East Africa, Colombia and Brazil. It explains the origins, operations and collapse of the ICO as an international “government of coffee”.

Princeton University Press · press.princeton.edu

Daviron and Ponte, 2005 – The coffee paradox. A book on why coffee booms in consuming countries while producers receive low prices. The authors argue that consumers pay mainly for symbolic qualities and services, not for the material coffee that producers sell.

Zed Books, London · agritrop.cirad.fr

Contested
  • Weight of sales to non-member countries in the 1989 breakdown: a 1989 World Bank paper reports that some exporters with large stocks reportedly sold to non-quota markets at discounts of as much as 50%6, and a 2002 study lists the growing trade with non-member importing countries among the problems of the system8; the ICO’s own history calls these sales an element perceived as unfair but says their volumes were never large enough to affect the price effect of the quotas, and names disputes over market shares as the main issue.3
  • Who gained from the quotas: a 1990 model study found that quotas reduced real export revenue for most small exporting countries while large producers gained, with gains in risk reduction for most small countries7; a 2002 study, citing 1997 estimates, reports that producing countries kept a larger share of total coffee income under the agreements than in the six years after 1989.8 The two measure different things (revenue against a no-quota simulation; shares of consumer spending).
Share or cite this entry

Change log

  • – Volatility figures re-attributed: only the 37% for 1990–97 comes from the 1998 analysis; the study bases its other figures on exchange data. Reason: Wording of Ponte (2002) and its footnote 14.
  • – Quota mechanics re-cited (tightening of quotas when prices fell is from Ponte 2002, not the 1989 World Bank paper); the non-quota market and the cause of the 1986 suspension restated as the World Bank paper gives them; the 1987 delay over market shares attributed to the ICO history only. Reason: The 1989 World Bank paper does not say these things in the cited form.
  • – ICA 2022 status now taken from the ICO Council record of March 2026: not yet in force; two thirds of the votes of each group are required. Reason: ICO document added as source 17.

Sources

  1. A(2011). Dates of International Coffee Agreements. International Coffee Organization.
  2. A(1963). No. 6791. International Coffee Agreement, 1962. Done at New York, on 28 September 1962. United Nations Treaty Series, vol. 469.
  3. BDubois, P. (2013). The International Coffee Organization 1963–2013: 50 Years Serving the World Coffee Community. International Coffee Organization.
  4. ARibeiro, F. (2011). A política econômica e o Convênio de Taubaté na economia cafeeira (1889–1906). Pesquisa & Debate (PUC-SP) 22(1), 75–93.
  5. AUnited States Senate, Committee on Finance (1941). Implementing the Inter-American Coffee Agreement (Senate Report No. 74, 77th Congress, 1st Session). United States Senate.
  6. AAkiyama, T.; Varangis, P. N. (1989). Impact of the International Coffee Agreement’s Export Quota System on the World’s Coffee Market (WPS 148). World Bank, Policy, Planning, and Research Working Papers.
  7. AAkiyama, T.; Varangis, P. N. (1990). The Impact of the International Coffee Agreement on Producing Countries. The World Bank Economic Review 4(2), 157–173. doi:10.1093/wber/4.2.157 · other copies
  8. APonte, S. (2002). The ‘Latte Revolution’? Regulation, Markets and Consumption in the Global Coffee Chain. World Development 30(7), 1099–1122. doi:10.1016/S0305-750X(02)00032-3 · other copies
  9. ABates, R. H. (1997). Open-Economy Politics: The Political Economy of the World Coffee Trade. Princeton University Press.
  10. ADaviron, B.; Ponte, S. (2005). The Coffee Paradox: Global Markets, Commodity Trade and the Elusive Promise of Development. Zed Books, London.
  11. BPerez, J.; Kilian, B.; Pratt, L.; Ardila, J. C.; Lamb, H.; Byers, L.; Sanders, D. (2017). Economic Sustainability – Price, Cost, and Value. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 6, from p. 133. doi:10.1016/B978-0-12-803520-7.00006-2 · other copies
  12. A(2022). Membership of International Coffee Agreements 1962, 1968, 1976, 1983, 1994, 2001 and 2007. International Coffee Organization.
  13. A(2026). Membership of the International Coffee Organization under the International Coffee Agreement 2022, as at 24 March 2026. International Coffee Organization.
  14. A(2025). Resolution Number 485: Extension of the time limits for signature and deposit of instruments under the International Coffee Agreement 2022. International Coffee Council.
  15. AEuropean Commission (2025). Proposal for a Council Decision on the signing, on behalf of the European Union, of the International Coffee Agreement 2022 (COM(2025) 469 final). Council of the European Union, document 12670/25.
  16. B(2019). Progress report on achieving the objectives of the 2007 Agreement (ICC 125-5). International Coffee Council.
  17. A(2026). Decisions and Resolutions adopted at the 141st Session of the International Coffee Council (ICC-141-6). International Coffee Council.

Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.

The literature for this entry was last searched on 24 Oct 2026. A statement that no source consulted for this entry shows something refers to the sources above; newer or missing work may exist.

Suggest a change

On The International Coffee Agreements. Tell us what you would change and why. A source that supports it helps most.