III Beverages

International Monetary Fund
Published Date:
January 1988
  • ShareShare
Show Summary Details

This section discusses market developments for coffee, tea, and cocoa.56 In the short run, demand for these three commodities tends to be stable and strongly price inelastic. Supply, however, is subject to cyclical fluctuations owing to the perennial nature of the three crops and the resulting lagged response of production to prices. In addition, the three commodities are vulnerable to weather-related short-run supply fluctuations. Consequently, supply changes tend to be the main determinants of price fluctuations. On the demand side there is little substitution in consumption among the three commodities, except during periods of very large price movements, when there has been evidence of limited substitution between coffee and tea in some consuming countries.

Partly because of the dominant role of supply factors, beverage prices have tended to move independently of prices of other primary commodities. As the weight of coffee in the overall index for beverages is much larger than the combined weights of cocoa and tea, movements in coffee prices have a relatively large effect on the movement of the overall beverage price index. This pattern was evident in 1986 and 1987 when the index of dollar prices of beverages moved in a direction opposite to the index of dollar prices of all commodities (Tables 3 and 38 and Chart 5). In 1986 the index of dollar prices of beverages rose by 15 percent when the index of dollar prices of all commodities excluding beverages declined by 7 percent, while in 1987 the index of dollar prices of beverages fell by 28 percent when the index of dollar prices of all commodities excluding beverages rose by 16 percent. The index of SDR prices of beverages remained unchanged in 1986 and fell by 35 percent in 1987. These price changes were primarily the result of changes in the supply of beverages. The index of world beverage production fell by 11 percent in 1986 before recovering by 17 percent in 1987 (Table 39). Similarly, the index of world beverage supply declined by 5 percent in 1986 before recovering by 8 percent in 1987. The 15 percent increase in the index of dollar prices of beverages in 1986 was largely the result of a sharp drop in world coffee production in the 1986/87 (October/September) coffee year. Similarly, the sharp drop in the index of dollar prices of beverages in 1987 was mainly attributable to a large recovery in world coffee production.

Table 38.Prices of Beverages, 1979–88
YearsIndex of Prices of Beverages1“Other mild” arabica2



Cocoa Beans5

(1980 = 100)(In SDRs)
(1980=100)(In U.S. dollars)
Source: Commodities Division, IMF Research Department.

Chart 5.Prices of Beverages, 1980–87

(In SDRs; indices: 1980 = 100)
Table 39.Movements in the Prices of Beverages and Related Economic Indicators, 1980–87(Annual percentage changes)
Prices of beverages1
In SDRs-13.2-12.57.311.920.9-10.80.0-34.8
In U.S. dollars-12.6-
Consumer price index in seven industrial countries
In SDRs11.413.
In U.S. dollars12.22.6-
Real GNP in seven industrial countries1.21.7-
World consumption of beverages3
Index of consumption1.
World supply of beverages3
Index of production3.79.5-
Index of supply45.412.4-
Index of closing stocks23.933.6-8.8-7.9-4.314.6-14.223.7
Sources: Commodities Division and Current Studies Division, IMF Research Department.


A rapid rise in coffee prices that started in late October 1985 brought to an end a five-year period of relatively stable prices. The dollar price for arabica coffee increased by more than 75 percent over the next four months, while the price for robusta coffee increased by more than 50 percent. Prices of coffee remained relatively high through most of 1986, but by mid-1987 had fallen to low levels similar to those recorded in the early 1980s.

The largest coffee producing countries are Brazil, Colombia, Indonesia, Côte d’Ivoire, and Mexico. In recent years, total annual world coffee production has tended to fluctuate around a level of 90 million 60-kilogram bags (5.4 million tons). Owing to the long gestation period for this tree crop, the time lags between price changes and production adjustments can be quite long. Consequently, the short-term price elasticity of supply tends to be quite low—about 0.06, while the long-term price elasticity of supply is about 0.3. Consumption in producing countries is important only in Brazil, India, Indonesia, and Mexico, and most of the world’s coffee is exported in the form of beans to the industrialized countries of North America, Western Europe, and Eastern Europe. The demand for coffee is not very responsive to price changes; the world short-term price elasticity of demand is about -0.19 while the long-term elasticity has been estimated at -0.23. The income elasticity of demand, on the other hand, is about 0.45 for the world as a whole and is much higher in Japan (2.0), Eastern Europe (1.0), and the EC (0.60).

Export quotas under the 1976 and 1983 International Coffee Agreements had a major bearing on coffee market developments in the 1980s. When export quotas were first imposed in October 1980, the composite indicator price57 of the International Coffee Organization (ICO) remained outside the agreement’s price range of $1.20-1.40 a pound because of relatively large stocks in consuming countries. A frost in Brazil in July 1981 and export quotas were instrumental in keeping prices relatively stable between mid-1981 and October 1985. The disparity between world coffee production and consumption was absorbed by a large buildup in coffee stocks in exporting countries (Table 40).

Table 40.Coffee: World Commodity Balance, 1981/82–87/88(In millions of 60-kilogram bags)
October/September Crop Years
Other countries250.
Exporting countries19.620.218.719.419.519.720.1
Other countries210.010.69.810.610.511.010.9
Importing countries65.066.366.266.465.565.866.5
ICO members356.056.856.255.957.057.556.0
Closing stocks62.064.964.565.368.759.469.1
In exporting countries49.654.351.450.950.043.150.8
Other countries22.227.525.925.
Other stocks12.410.613.114.418.716.318.3
Inventories in importing countries5.
Stocks in free ports2.
Afloat, etc.
Source: Based on statistics of the International Coffee Organization (ICO).

A sharp rise in prices that began in late October 1985 was caused by a prolonged period of dry weather in the major Brazilian coffee producing states of Minas Gerais, Paraná, and São Paulo, which delayed and severely reduced flowering and fruit set on coffee plantations. Despite the arrival of rains in November and early December, prices continued to rise. Estimates by the coffee trade and government agencies indicated that the 1986 Brazilian coffee crop (during the April 1986 to March 1987 Brazilian crop year) was likely to amount to less than one half of the 1985 crop of 30 million bags. Prices, which averaged $1.27 cents a pound in October 1985, rose to $1.41 in November and $1.75 cents a pound in December. Three successive quota increases totaling 5 million bags or about 8.6 percent of the initial 1985/86 quota of 58 million bags did little to check the upward spiral in prices. Large-scale precautionary and speculative purchases resulting from continued concern about the size of the 1986 Brazilian crop helped to drive prices to an average level of $2.04 a pound in January 1986, which was the highest average monthly price since mid-July 1977. Quotas under the International Coffee Agreement were suspended on February 18, 1986, because the ICO had exhausted all its means of defending the ceiling price ($1.40 a pound) of the price stabilization range.

The prospect of a shortage of arabica coffee as a result of the Brazilian drought caused the price differential between arabica and robusta coffees to widen. Precautionary purchases by roasters transferred stocks from producing countries to consuming countries. Substantial exports to nonquota markets at discounts of up to 50 percent had been common during the period when export quotas were in effect, but with the suspension of export quotas, the volume exports to nonquota markets declined sharply, and the differential in export unit values from sales to quota and nonquota markets narrowed considerably.

An increase in export volumes following the suspension of quotas and a lack of demand from roasters in consuming countries who had already built up substantial stocks caused prices to weaken during the latter part of March 1986. Despite the approach of the frost season in Brazil and the increased vulnerability to frost of the drought-weakened trees, prices continued to decline through July 1986 because of unusually warm weather in Brazil and expectations that the long-term effects of the 1985 drought would be quite small. Prices recovered slightly in August when estimates of the 1986 Brazilian harvest were reduced substantially. In the event, the 1986 Brazilian harvest amounted to 11 million bags, as opposed to estimates of 16 million bags in late 1985. Concern about the effects of dry weather on the 1987 Brazilian crop and purchases of coffee by Brazil on the terminal markets caused prices to strengthen further in September 1986 when the ICO indicator price rose to $1.87 a pound.

During the 1986/87 coffee year, the main factors influencing coffee price movements were the outcome of ICO discussions pertaining to the reintroduction of coffee export quotas, the relatively high level of coffee stocks in consuming countries, and the prospects for the 1987 Brazilian coffee crop. At its September 1986 meeting in London, the International Coffee Council failed to agree on a new quota allocation for the remaining three years of the 1983 International Coffee Agreement. Consumers did not agree to a producer proposal to maintain the 1985/86 quota distribution for the 1986/87 coffee year. The relatively high level of market prices at that time may have reduced the urgency to agree on a new quota system. Instead, the Council decided that the Executive Board of the ICO should meet in special session if the 15-day average of the composite indicator price was to fall to $1.45 a pound.

Fears of a second year of drought in the Brazilian coffee belt subsided when an unusually dry spell was broken by normal rains in late October 1986. With ample rainfall from then on, there was a rapid recovery in the coffee trees and in the prospects for the 1987 Brazilian coffee crop, which was then forecast to exceed 30 million bags. Despite a decline in coffee prices from the very high levels in early 1986, import demand remained weak because of a relatively high level of stocks in consuming countries. Weak import demand combined with a favorable supply outlook, caused the composite indicator price to decline from $1.63 a pound in October to $1.30 a pound in December.

A fall in the composite indicator price to $1.45 a pound triggered a meeting of the Executive Board of the ICO in mid-December 1986. The meeting decided only to keep the market situation under review. By early January 1987, when the composite indicator price had fallen to almost $1.20 a pound, officials from eight major exporting countries met in London and decided to convene a meeting of the 50 exporting members of the ICO to discuss matters relating to the reintroduction of quotas. A special session of the Council in February 1987 ended without a consensus on the distribution of quotas. The principal difficulty was the criteria to be used in allocating export quotas. Brazil and a number of larger producing countries were in favor of retaining the old formula under which export quotas were based on historic trade shares of the 1960s and 1970s. Seeking greater access to the types and qualities of coffee that they desired, consuming countries for the first time became more directly involved in the quota negotiation process. These countries were in favor of allocating quotas on the basis of more recent trends in export performance and in the stock levels carried over from the previous marketing year. A group of six producing countries, where production had expanded rapidly in recent years, supported the consumer position.

Following the failure of the special ICO Council to arrive at an agreement, the composite indicator price fell by 15 cents to about $1.00 a pound. Increased supplies from origin and limited demand owing to continued destocking by roasters and dealers caused prices to hover around $1.00 a pound in March and April. In May, though there was no major change in market fundamentals, prices recovered to an average level of $1.11 a pound owing largely to the depreciation of the U.S. dollar. With an increase in coffee supplies resulting from a complete recovery in the 1987 Brazilian coffee crop to 35 million 60-kilogram bags and the question of export quotas being put off until at least the beginning of the 1987/88 coffee year, prices fluctuated between $0.95 and $1.01 cents a pound in June, July, and August.

The prospect of a reintroduction of export quotas from October 1987, the beginning of the 1987/88 coffee year, caused prices to recover from the fourth week of August onward. During the first half of September, Latin American coffee exporters agreed on a quota stance for the ICO meetings, which were scheduled for the third week of September. During the fourth week of September, all the producer members of the ICO, with the exception of Indonesia, agreed on a formula for the reintroduction of export quotas. After protracted negotiations between producers and consumers, on October 5, the International Coffee Council agreed to reintroduce export quotas from October 6 onward. The initial global quota for the 1987/88 coffee year was set at 58 million bags or at the same level as the initial global quota for the 1985/86 coffee year. The Council agreed on an objective to stabilize prices in a range between $1.20 and $1.40 a pound. As the composite indicator price was substantially below $1.20 a pound when quotas were reintroduced, the Council also agreed on transitional arrangements to cut quotas during the first quarter of the 1987/88 coffee year. In a departure from past practice, the Council also agreed on a new formula for the allocation of export quotas during the 1988/89 coffee year. The formula, which takes account of recent trends in exportable production, exports, and of the level of stocks, addresses the concerns of some consuming and producing countries about the current system of quota allocation.

Despite the reintroduction of quotas, followed by a quota cut of 1.5 million bags in early October 1987 and two additional quota cuts of 1.0 million bags each in early January, the 15-day moving of the average of the composite indicator price remained below the agreed floor price of $1.20 a pound until late February 1988.

Responding to the high coffee prices that prevailed during the mid-1970s, world coffee production increased from an average level of 69.3 million bags during the three coffee years 1975/76–77/78 to an average level of 91.5 million bags during the three coffee years 1983/84–85/86. Production of arabica coffee increased by 36 percent during this period while robusta coffee production increased by 27 percent. Regionally, the largest production increases were recorded in South America and Asia. Production in Central America and the Caribbean showed a small increase, while production in Africa remained virtually unchanged. Among the major producing countries, substantial production increases were recorded in Brazil, Colombia, and Indonesia; if allowance is made for the effects of a severe drought in 1983/84, production in Côte d’Ivoire remained virtually unchanged during this period.

The effects of a severe drought in Brazil, which caused total world coffee production to decline sharply from 89.8 million bags in 1985/86 to 77.7 million bags in 1986/87, underscores the key role that Brazil continues to play in world coffee market developments. Owing to a large increase in coffee production and exports from other countries, Brazil’s share in total exports has declined from about 50 percent in the early 1950s to its current level of about 25 percent. At the same time, the cushion to variations in world coffee production provided in the earlier period by the level of stocks in Brazil has been sharply reduced because Brazilian stocks have declined from the equivalent of over one year’s total world coffee consumption in the early 1960s to their current level of about 35 percent of total world coffee consumption. The shift in the focus of Brazilian coffee production from the frost-prone southern states of Paraná and São Paulo to the states of Minas Gerais and Espírito Santo does not seem to have substantially reduced the variability in total Brazilian coffee production. Recent experience indicates that coffee production in Minas Gerais, which currently accounts for about one third of total Brazilian coffee production, is susceptible to both frosts and droughts.

Growth in world coffee consumption has lagged behind the relatively rapid rate of growth in world coffee production, because of the impact of changes in beverage consumption tastes and high retail prices for coffee. Despite promotional efforts, the ICO’s winter 1986 study on coffee consumption indicates that per capita consumption in the United States had declined to 1.74 cups a day as compared with 1.99 cups a day in 1984 and 3.12 cups a day in 1962. The declining trend in U.S. coffee consumption has been slightly more than offset by consumption growth in Western Europe and Japan. Consequently, world coffee consumption has increased, but at an annual rate of less than 1 percent during the period 1980/81 to 1986/87.

The level of world coffee stocks increased sharply from 27.0 million bags in 1979/80 to 41.7 million bags in 1980/81 when export quotas were first imposed under the 1976 International Coffee Agreement. Stock levels remained high during the five subsequent coffee years when export quotas remained in effect. These stocks, which were equivalent to about a year’s world coffee consumption, were largely held in coffee exporting countries. With the suspension of export quotas in February 1986 and the sharp fall in the 1986 Brazilian crop, the level of world stocks fell from 68.7 million bags in 1985/86 to 59.4 million bags in 1986/87.

World coffee production is expected to reach a record 98 million bags in 1987/88 largely because the 1987 Brazilian crop, estimated at 35 million bags, is expected to be about two and a half times as large as the drought-reduced 1986 crop. Since world coffee consumption in 1987/88 is expected to increase by about only 1 percent to 86.6 million bags, world coffee stocks are expected to increase by 16 percent to 69.1 million bags. Despite this large expected increase in stock levels, export quotas under the ICO are expected to stabilize prices near the lower end of the $1.20–1.40 a pound price range specified in the agreement.

After rising to 71.1 million bags in 1985, owing in part to precautionary purchases by importers in the final quarter of 1985, the volume of world coffee exports declined by 9 percent to 64.6 million bags (3.9 million tons) in 1986. Among the primary commodities exported by non-oil exporting developing countries, coffee is the largest single source of foreign exchange earnings. The coffee earnings of these countries increased from $10.8 billion in 1985 to $14.2 billion in 1986 as a result of the sharp rise in coffee prices during 1986. In 1987, despite an expected recovery in export volume to 72.2 million bags (4.4 million tons), export earnings are expected to fall by about 30 percent to $10 billion because of sharply lower prices. In terms of SDRs earnings are expected to decline by 36 percent from SDR 12.1 billion in 1986 to SDR 7.7 billion in 1987 (Table 41).

Table 41.Coffee: Export Earnings, 1984–87
(Values in SDRs)(Values in U.S. dollars)
Earnings (in billions)10.410.712.17.810.610.814.210.0
Developing countries10.410.712.17.810.610.814.210.0
To ICO members19.49.511.
To nonmembers of ICO1.
Volumes (in thousands of tons)
Developing countries4.
To ICO members13.
To nonmembers of ICO0.
Unit values (a ton)2,5302,5003,1201,7802,5902,5203,6602,280
To ICO countries2,6702,5603,1601,8102,7302,6003,7102,310
To nonmembers of ICO21,6401,9002,8001,5401,6801,9303,2801,990
Market prices
ICO daily composite indicator (a ton)
Other milds33,1003,1603,6201,9203,1803,2104,2502,480
Sources: Export data are based on statistics of the International Coffee Organization (ICO); Commodities Division, IMF Research Department for market prices.


Tea prices, which peaked at $4.29 a kilogram in January 1984, started declining sharply during the second quarter of 1985 as a result of an increase in world tea production. Prices were relatively stable in 1986 when they averaged $1.93 per kilogram, but fell further during 1987 to average $1.71 per kilogram for the year.58

The boom in tea prices that started in the fourth quarter of 1983 was caused by a low level of world inventories and concern about prospective supplies owing to low rainfall in Sri Lanka and a drought in southern India. A partial ban on the export of CTC (crush, tear, and curl) teas by India, beginning in January 1984, also contributed to the rise in prices. In the event, world tea production in 1983 was about 5 percent higher than in 1982; production in Sri Lanka declined by about 5 percent, but production in northern India, which increased by 7 percent, more than compensated for a 7 percent reduction in the south Indian crop (Table 42). Prices remained firm during 1986 because of strong import demand to rebuild depleted stocks in consuming countries and because of continued uncertainty about the availability of export supplies from India.

Table 42.Tea: World Commodity Balance, 1981–87(In thousands of tons)
Green tea420480520520550560570
Other countries150150190180190190190
Black tea1,4501,4501,5201,6501,7801,7301,790
Sri Lanka210190180210220210210
Other countries600610640680770760760
Consumption (black and green)1,9201,9302,0302,1602,2602,2802,340
Tea producing countries21,1601,2001,1901,2301,3101,3201,360
Other countries230230290280300320320
Tea importing countries3760730840930950960980
Asian countries200190230260260260270
United Kingdom180170170200180180180
Other countries380370440470510520530
Implied change in total stocks001010701020
Closing stocks in the United Kingdom58675662565460
Sources: Estimates of Commodities Division, IMF Research Department, derived from statistics of the International Tea Committee and the UN Food and Agriculture Organization.

Responding to the stimulus of high prices, tea production increased by 6 percent in 1984; most of the increase in production came from “coarse picking” of more leaves than normal. While the upward trend in production in Kenya was temporarily halted in 1984 because of a drought, production in India and Sri Lanka increased by 10 percent and 17 percent, respectively. Increased production resulting from the high tea prices of 1983 and 1984 and favorable weather in major producing countries brought the tea price boom to an end in 1985, when world tea production rose by 7 percent to a record 2.33 million tons. Tea production in Kenya, which had declined the year before, increased by 25 percent and there were large production increases in a number of the smaller tea producing countries, such as Argentina, Indonesia, and Rwanda. Prices, which averaged $3.46 a kilogram in 1984, declined to $1.98 a kilogram in 1985.

In 1986 world tea production declined by 35,000 tons to 2.3 million tons because of lower black tea production in many major and minor producing countries. Production in India declined by 5 percent to 620,000 tons because of adverse weather conditions in northern India. Output in Sri Lanka declined by 4 percent while production in Kenya declined by 7 percent, also on account of adverse weather conditions. On the other hand, production in China, which consists mainly of green tea, increased by 5 percent to a record 480,000 tons. Prices in 1986 averaged $1.93 a kilogram or about 3 percent lower than in 1985.

The pattern of tea consumption in industrial countries differs from that in developing countries. In major tea drinking industrial countries, such as Australia, Canada, and the United Kingdom, consumption has been declining, while in the developing tea producing countries and developing tea importing countries tea consumption has been increasing rapidly. Between 1961 and 1984 per capita tea consumption declined at an annual rate of 2 percent in the United Kingdom, the world’s largest tea importing country. During the same period per capita consumption increased at an annual rate of 2 percent in Egypt, 3 percent in India, and 5 percent in Pakistan. Import demand for tea increased in 1985 and 1986 because of lower tea prices; purchases by industrial countries are estimated to have risen by about 9 percent in 1986 owing to sharply higher imports by the United Kingdom and the United States. World tea consumption is estimated to have exceeded world tea production by 70,000 tons in 1985, but in 1986 the surplus of production over consumption is estimated to have been only 10,000 tons.

Tea production is estimated to have recovered to 2.36 million tons in 1987, largely because of favorable weather conditions in Kenya. In India the tea crop escaped the effects of drought; ideal growing conditions especially in northern India are estimated to have boosted production to a record 660,000 tons. Tea prices declined sharply from an average of $1.88 a kilogram during the first quarter of 1987 to $1.43 during the second quarter, mainly because of the availability of large supplies of low quality plain teas in London and because of a decision by Pakistan in late March 1987 to suspend purchases of Kenyan tea. During the third quarter, prices recovered slightly owing to greater availability of good quality north Indian and Kenyan teas and a reduction in the supplies of poorer quality grades that had been holding prices down. Prices recovered further during the fourth quarter because of a seasonal rise in demand for tea.

Despite an estimated 3 percent increase in world tea production in 1987, world tea production and consumption were in better balance because the lower tea prices caused an increase in tea consumption in both tea producing and tea importing countries. An implied stock increase of 20,000 tons is estimated for 1987.

The tea price boom caused earnings from tea exports to rise sharply, from $1.7 billion in 1983 to a record $2.5 billion in 1984 (Table 43). In 1985, however, despite an increase in export volume, earnings declined to $2.0 billion. Weak prices caused earnings to decline further to $1.7 billion in 1986 and to an estimated level of $1.6 billion in 1987. Owing to the depreciation of the dollar since the end of 1984, earnings in terms of SDRs have declined much more sharply—from SDR 2.0 billion in 1985 to an estimated level of SDR 1.2 billion in 1987.

Table 43:Tea: Export Earnings, 1984–87
(Values in SDRs)(Values in U.S. dollars)
Earnings (in billions)
Developing countries2.
Volumes (in thousands of tons)9909901,0201,0009909901,0201,000
Developing countries9909901,0201,0009909901,0201,000
Sri Lanka200200210210200200210210
Other countries470450480450470450480450
Unit values (a ton)2,5102,0401,4501,2002,5702,0701,7001,550
Market prices (a ton)13,3701,9501,6401,3203,4601,9801,9301,710
Sources: Export data are estimates of Commodities Division, IMF Research Department, derived from statistics of the International Tea Committee and the UN Food and Agriculture Organization; Commodities Division, IMF Research Department for market prices.


Cocoa is a tree crop, which takes about 10 years to reach maximum yields; the newer high yielding hybrid varieties reach peak yields at about seven years of age. Owing to the perennial nature of the crop and its relatively long gestation period, cocoa producers tend to adjust to prices only after relatively long time lags. Since production in the short run can only be changed by varying the intensity of input use or by varying thoroughness of harvesting, the short-run price elasticity of cocoa supply tends to be quite low.59 In the longer run, production can be more readily varied by increasing or decreasing the area under cocoa cultivation. Consequently, the long-run price elasticity of cocoa supply tends to be higher.60 The relatively high long-run supply elasticity has imparted a cyclical pattern to cocoa production and prices; low production and high prices encourage increased plantings which, after a lag, result in much higher production and lower prices.

World cocoa market developments during the 1980s reflected the lagged response of world cocoa production to the high real cocoa prices in the late 1970s. Production increases, especially in the Côte d’Ivoire, Brazil, and Malaysia, caused world cocoa output to increase from 1.50 million tons in the 1978/79 cocoa year (October/September) to an estimated 2.06 million tons in the 1987/88 cocoa year. Growth in world cocoa consumption, on the other hand, lagged well behind the growth in world cocoa production. As a result, closing stocks are expected to increase from 0.38 million tons at the end of the 1978/79 cocoa year to 0.88 million tons at the end of the 1987/88 cocoa year.

In the period prior to August 1981 when the 1980 International Cocoa Agreement (ICCA) became operational, cocoa market prices were well below both the agreement’s lower intervention price of $1.10 a pound ($2,425 a ton) and the minimum price of $1.00 a pound ($2,205 a ton).61 Buffer stock purchases of 100,000 tons of cocoa between September 1981 and March 1982 failed to raise the price above the lower intervention price. Buffer stock purchases which were suspended in March 1982 due to a lack of funds were not resumed during the rest of the life of the 1980 Agreement.

Improved crop prospects in West Africa and concern about consumer ratification of the 1986 ICCA caused prices to weaken to $2,024 a ton during the first quarter of the 1986/87 cocoa year.62 The 100,000 tons of cocoa purchased by the buffer stock of the 1980 International Cocoa Agreement were transferred to the 1986 ICCA, which entered into force provisionally on January 20, 1987. At its meeting in January 1987 the International Cocoa Council was unable to agree on buffer stock intervention rules. As a result prices continued to fall during the first quarter of 1987. In March, the Council approved a new set of buffer stock intervention rules which permitted buffer stock purchases to commence on May 18 to support the agreement’s lower intervention trigger price of SDR 1,600 a ton. By June 22, when the buffer stock had purchased 75,000 tons of cocoa, buffer stock purchases were suspended pending an International Cocoa Council decision on a revision of the trigger prices for buffer stock operations. Prices averaged $2,046 (SDR 1,577) a ton during the second half of the 1986/87 cocoa year. The International Cocoa Council met in July, September, and December 1987, but failed to reach an agreement on lowering the price range for buffer stock operations. The collapse of the December talks caused cocoa prices to fall sharply to a five-year low of $1,860 (SDR 1,340) a ton. Buffer stock purchases were suspended from late June 1987 until mid-January 1988 because the International Cocoa Council was unable to arrive at a decision on lowering the price range.63

At a special session in mid-January 1988, however, the Council agreed to revise the lower intervention price to SDR 1,485 and lifted the suspension on buffer stock purchases. At the same meeting the Council also agreed to adopt the rules for a scheme to withhold up to 120,000 tons of cocoa from the market. As a result of further purchases by the buffer stock manager, by the end of February 1988 the buffer stock had reached its maximum capacity of 250,000 tons. These actions provided little support to the market because of the prospect of a large surplus of production over consumption in 1987/88. By the end of February, prices had fallen to $1,712 (SDR 1,260) a ton, the lowest level in more than five years.

The largest increases in cocoa production in the 1980s were recorded in the Côte d’Ivoire, Brazil, and Malaysia. Officially sponsored large-scale new plantings with hybrid varieties and rehabilitation programs in the Côte d’Ivoire and Brazil during the late 1970s and early 1980s contributed significantly to increased production in these countries. An increase in the total land area planted to cocoa, from 0.95 million hectares in 1976/77 to 1.35 million hectares in 1985/86, caused production in the Côte d’Ivoire to increase from 230,000 tons to 580,000 tons over the same period. Owing to the efforts of the agricultural extension service and the cocoa research institute, about half of the new area was planted with high yielding hybrid varieties. An increase in planted area in Brazil from 0.49 million hectares in 1976/77 to 0.70 million hectares in 1985/86 was also largely responsible for a 25 percent increase in production in that country. The area planted with hybrid varieties increased from 60,000 hectares to 340,000 hectares during the same period. In addition to a large proportion of hybrid varieties, Brazil also has the youngest stock of trees of all the traditional major cocoa producing countries; about 40 percent of the trees are less than 10 years old. The availability of large areas of undeveloped land suitable for cocoa cultivation and a long tradition of tree crop production enabled Malaysia to exploit rapidly the technological advantages of the earlier maturing and much higher yielding hybrid varieties. The area under cocoa, all of it planted with hybrid varieties, increased from 40,000 hectares in 1976 to 310,000 hectares in 1986. As these new plantings came into bearing and the productivity of the trees increased with maturity, production increased from 12,000 tons in 1974/75 to 160,000 tons in 1986/87. In 1986/87 Malaysia replaced Nigeria as the world’s fourth largest cocoa producer.64

Cocoa beans and the semi-finished cocoa products produced by grinding cocoa65 are largely consumed in the industrial countries and the U.S.S.R. In these countries, chocolate and cocoa-based confectionery products tend to compete with a wide range of other confectionery and snack foods. Despite the availability of these substitutes and a tendency for manufacturers to vary the cocoa content of their products in line with movements in the cocoa price, price elasticity of demand for cocoa is quite low.66 Similarly, the income elasticity of demand for cocoa also tends to be quite low in the major cocoa consuming countries.67 In the period since the 1981–82 recession, though real GDP in the main cocoa consuming countries has grown and cocoa prices in real terms have declined quite sharply in the period since 1983/84, world cocoa consumption has only increased at a relatively modest rate in the period 1984/85–86/87. With world cocoa production exceeding world cocoa consumption in each year beginning 1984/85, closing stocks increased from 460,000 tons at the end of the 1983/84 crop year to 760,000 tons by the end of the 1986/87 crop year (Table 44). Most of the world’s cocoa stocks are held in the form of cocoa beans in the major importing countries. The ratio of total world stocks to world consumption increased from 0.27 in 1983/84 to 0.41 in 1986/87. Owing to a continuing excess of production over consumption, stocks are expected to increase to 880,000 tons by the end of the 1987/88 crop year. Of this amount, the 630,000 tons expected to be held as “free stocks” by the cocoa trade and cocoa end users will have a direct impact on market prices because they are readily available to the market. By the end of February the buffer stock of the International Cocoa Agreement had reached its maximum capacity of 250,000 tons. These stocks would be unavailable to the market unless prices rise high enough to trigger buffer stock sales.

Table 44.Cocoa Beans: World Commodity Balance, 1981/82–87/88(In thousands of tons)
October/September Crop Years
Gross production1,7301,5301,5101,9601,9601,9802,060
Côte d’Ivoire470360410570580580600
Other countries270280290300320330340
Production adjusted for loss in weight1,7101,5101,4901,9401,9401,9602,040
Grinding (consumption)1,5801,6301,7001,8501,8201,8701,920
Cocoa bean producing countries470480510600620580600
European Community570580610640630660680
United States200190210190190220230
Other countries210220230250240250260
Closing stocks790670460550670760880
Commercial stocks690570360450570585630
ICCO buffer stock100100100100100175250
Source: International Cocoa Organization (ICCO), Quarterly Bulletin of Cocoa Statistics (London), various issues.

In dollar terms, earnings of developing countries from exports of cocoa beans and cocoa products increased from $3.6 billion in 1985 to $3.7 billion in 1986 because of a 2 percent increase in export unit value; export volume was virtually the same in both years (Table 45). In 1987 earnings of developing countries from exports of cocoa beans and cocoa products are estimated to have declined to $3.6 billion because an estimated 5 percent fall in export unit values more than offset increased volumes. Owing to the depreciation of the U.S. dollar, the decline in earnings of developing countries from the export of cocoa beans and cocoa products was much larger in terms of SDRs. Earnings declined from SDR 3.6 billion in 1985 to an estimated SDR 2.8 billion in 1987.

Table 45.Cocoa: Export Earnings, 1984–87
(Values in SDRs)(Values in U.S. dollars)
Earnings (in billions)
Developing countries2.
Developing countries0.
Industrial countries0.
Volumes (in thousands of tons)
Developing countries340370380370340370380370
Industrial countries270280270290270280270290
Unit values (a ton)
Developing countries2,6802,6802,2501,9002,7502,7202,6402,460
Industrial countries3,0703,4503,2522,8503,1503,5003,8103,690
Market prices of cocoa beans (a ton)32,3402,2201,7601,5502,4002,2602,0702,000
Sources: UN Food and Agriculture Organization, 1986 FAO Trade Yearbook (Rome) for exports; for cocoa beans, export data include only exports from cocoa bean producing countries; Commodities Division, IMF Research Department for market prices.

Provided weather conditions remain normal, the near-term outlook is for world cocoa production to continue to exceed world cocoa consumption. As total stocks rise to the equivalent of more than four months of consumption, prices can be expected to remain under downside pressure during 1987/88.

    Other Resources Citing This Publication