Previous Page Table of Contents Next Page


David Hallam[1]

This paper examines the nature, origins and implications of the sharp decline in coffee prices since 1998. The decline is attributed to the significant expansion in global supplies against sluggish demand growth. Recent efforts on the part of producers and exporters to control supply growth or to promote demand growth are reviewed. It is argued that so-called "producer-only agreements" to restrict production or exports are unlikely to succeed because of the difficulties in maintaining the commitment of participants and policing such schemes. The organization of demand promotion is also problematic where stakeholders may see their interests as competing, and the experience with coffee indicates that there is a need to establish clear strategic aims to which all can subscribe. However, in the longer term the tendency towards oversupply in the coffee market can only be addressed by encouragement of diversification out of coffee production at least in marginal areas.

1. Introduction

Although depressed prices have been common to most commodities, much attention has focused on coffee. As the single most important tropical commodity accounting for almost half of total net exports of tropical products, coffee has become emblematic of the problems faced by all developing country agricultural commodity exports. Price falls for coffee have been particularly dramatic: after a brief recovery in the mid-1990s when buffer stocks were finally cleared, real coffee prices had fallen by 2001 to levels lower than ever recorded. In real terms coffee prices today are less than one third of their 1960 level, and for many producers less than the cost of production. According to the International Coffee Organization (ICO), this impacts directly upon an estimated 20-25 million households in coffee-producing countries, and indirectly upon up to a further 100 million engaged in upstream and downstream activities. The wider economic and political implications are clear: as James Wolfensohn, President of the World Bank, noted, "The reduction of coffee prices and also other commodities... is undermining the economic sustainability of countries and millions of families in Latin America, Africa and Asia".[2]

Many different explanations have been proposed for the precipitous decline in coffee prices. These include the emergence of Viet Nam as a major producer and exporter, the depreciation of the Brazilian real, "underconsumption", exploitation of market power by roasters and retailers, technological change in roasting, domestic market liberalization and the abolition of parastatal marketing agencies. In its recent resolution, the European Parliament attributes the crisis to the dismantling of the international coffee agreement and the policies implemented by the World Bank, the International Monetary Fund (IMF) and the World Trade Organization (WTO). But basically the explanation lies in the market fundamentals of supply and demand. While it is tempting to assume that such a precipitous fall in prices must be due to some new factor or some change in market behaviour, according to FAO price determination models the operation of market fundamentals has not changed. Specifically it is the recent rapid growth in global supplies against sluggish demand growth which has led to falling prices, and the low price elasticity of demand means that these price falls are severe.

Suggested solutions to the crisis have been as various as the explanations. These have included supply control, demand promotion, guaranteed prices, product differentiation, support for diversification (and trade liberalization to provide opportunities for diversification), vertical coordination or integration through the value chain, raising the profile of commodity problems in international fora, fair trade initiatives (including obliging the four main coffee roasters to pay a fair price to farmers and end "exploitation"), and even grower support funded by a windfall tax on roasters.

A tendency for expanding supplies to outstrip demand growth on world markets is not peculiar to coffee. The resulting market imbalances coupled with low price elasticities of demand led to the same downward pressure on prices across a broad spectrum of commodities, albeit less dramatically than for coffee. Some of the same solutions, notably demand promotion and supply control, have been implemented or are under active discussion in a variety of other international industry initiatives.

This paper examines the nature of the coffee crisis and discusses industry responses. Specifically, it focuses on the persistent decline in prices and its origins in the tendency for supply on world markets to grow ahead of demand. It considers experiences in internationally coordinated attempts on the part of producers and exporters to influence those market fundamentals by seeking to regulate supply or promote demand. It reviews recent efforts in these directions and examines what lessons can be learned for other commodities.

2. The nature of the international coffee crisis

The collapse in international coffee prices since 1998 is evident from Figures 1 and 2. The average ICO composite price fell by 21 percent in 1999, 25 percent in 2000, and 29 percent in 2001 to reach the lowest annual average since 1971. Apart from the upturn in the second half of the 1990s, prices have trended steadily downwards since the peak in 1977. Since mid 2001 prices appear to have levelled out a little but at very low levels, and this greater stability appears to have continued into the first three quarters of 2003.

Figure 1. Trends and variability in international coffee prices (annual averages)

Figure 2. Trends and variability in real coffee prices (annual average prices deflated by MUV, 1999 = 100)

The recent variability of prices is also apparent in Figures 1 and 2 with the downward trend interrupted by periodic peaks on average every nine or ten years since the maximum in 1977 with the most recent peak in 1998. Variability is important since the duration and amplitude of price movements are relevant to the design of countermeasures to stabilize prices. If shocks are long-lived then the costs of stabilization in terms of storage and financing will probably outweigh any consumption or income benefits. The persistence of shocks in commodity prices has been explored in a recent IMF study[3]. This found that shocks to commodity prices are typically finite in duration but long-lived. For coffee, persistence (measured as the length of time until the effects of a shock decline to half the original magnitude) is at least nine years. This is also evident from Figures 1 and 2. In these circumstances the IMF study concluded that the costs of operating any kind of stabilization are likely to exceed any smoothing benefits. Ironically the International Coffee Agreement was regarded as relatively successful.[4]

Price movements reflect the evolving demand and supply situation. It is clear from Figure 3 that supplies of coffee on the world market have typically run ahead of the growth in demand. Since domestic consumption in producing countries did not expand sufficiently to absorb growing supplies, coffee exports increased. But as developed country markets became increasingly saturated growth in export earnings lagged behind growth in export volumes. The export earnings of coffee producing countries have fallen from US$10-12 billion in the early 1990s to US$5-6 billion currently. However the value of retail sales of coffee has increased over the same period from around US$30 billion to around US$70 billion.

Figure 3. International market balance for coffee

Imbalance in the world coffee market and the consequent low prices were exacerbated by new plantings in Viet Nam, and by an increase in Brazilian exports following expansion of plantings into frost-free areas, productivity improvements and devaluation of the real in early 1999. These supply side developments outweighed the steady increase in global coffee demand. In the ten years to 2000/01 the area under coffee in Viet Nam expanded from 60 578 hectares to 463 450 hectares and coffee bean output increased from 96 000 tonnes to 800 000 tonnes, making Viet Nam the largest robusta producer and second largest coffee producer in the world. The consequent increase in export revenues provided a boost to the country's overall rural economy with multiplier effects on incomes and employment in upstream and downstream activities and leading to significant declines in the incidence of poverty and hunger. However, the subsequent decline in robusta prices, by 39 percent in 2000 and 33 percent in 2001 had a "domino effect" on arabica prices which were already under pressure from the 30 percent increase in coffee exports from Brazil in 1999.

Over the last four years, consumption has remained virtually unchanged, and against this background of saturated markets coupled with low price elasticity of demand, prices tend to decline rapidly and sharply. However, falling prices do not necessarily prompt the expected supply response. The perennial nature of the crop means that adjustment to the scale of production through diversification and exit from the industry is slow: in the short run the price elasticity of supply appears to be very small, around 0.25. It may also be that, as is often argued for perennial crops, supply responses to price incentives are asymmetric: periods of rising prices stimulate new plantings and other fixed asset investments which are not scrapped when prices fall, but rather are simply not replaced when they reach the end of their productive life. Supply responses to falling prices have also been slowed in some cases by national efforts to assist producers, for example through price supports and debt relief. In the short-term adjustments can be made to reduce application of inputs including labour, but creating unemployment and stimulating migration. Reduced labour input through less care of trees and in harvesting also has adverse effects on quality which in turn leads to additional pressure on average price levels. This, together with the fact that much of the expansion in production from Viet Nam was of inferior quality has lowered average quality, posing a threat to the various product differentiation initiatives to develop markets for high-quality "specialist" coffees. However, there are signs that areas planted are being cut back, in Viet Nam and Brazil, for example, and it is partly this which is giving some strength to prices in the last few months of 2003. Elsewhere, the abandonment of farms by smallholders and increased migration to urban centres have been reported to the ICO by Cameroon, Central African Republic, Colombia, Costa Rica, Ecuador, Nicaragua, and Philippines. Colombia further reported coffee land being used for illicit crops. Nevertheless, stocks remain at high levels with an apparent reduction in exporting country stocks being offset by further increases in importing country stocks.

The "coffee crisis" results not only from the price fall but also from the economic importance of coffee in many producing countries. The effects of the fall in coffee prices after 1998 were particularly severe in those countries where productivity growth has lagged behind, and coffee producers faced a tightening price-cost squeeze. However, such has been the extent of the fall in prices that the adverse economic and social impacts have become generalized with declining incomes, increasing unemployment and increasing rural poverty across all producing countries and all production systems. Any gains which might have derived from domestic market liberalization increasing the share of the export price going to farmers have been swamped.

While some traditional coffee exporters such as Brazil have diversified and reduced export dependency on coffee - from more than 40 percent in 1960 to less than 5 percent today - dependency remains a major problem, especially for poor African countries: Burundi derives nearly 80 percent of export earnings from coffee; Uganda and Ethiopia more than 50 percent; and Rwanda slightly less than 50 percent. A number of Latin American countries also have high dependency on coffee, notably Colombia and El Salvador where coffee has accounted for around 15 percent of export earnings, and Guatemala, Honduras and Nicaragua with around 20 percent of export earnings. Export dependency is also reflected in significant shares of employment related to coffee: in Colombia for example 30 percent of the rural population is directly dependent on coffee. Such dependency means that coffee price variations have significant multiplier effects on employment and incomes beyond production itself in related upstream and downstream industries, and across the economy in general.

The economic and social effects of falling coffee prices are documented by a recent ICO survey of producing countries.[5] Nicaragua reported 122 000 job losses, Costa Rica 10 000. In Papua New Guinea employment in the estates sector has fallen by 40 percent. In Ecuador the coffee processing sector is operating at only one third capacity. Almost all countries reported falling incomes and expenditures among coffee-dependent households. An apparently common coping strategy is reduced spending on health and education. In Papua New Guinea 50 percent of parents in the Eastern Highlands had not paid school fees this year. Food security has inevitably been reduced. Increased incidence of malnutrition is documented in Colombia where the number of households in coffee growing areas living below the poverty line increased from 54 to 61 percent between 1997 and 2000. Malnutrition is also reported to be affecting 45 percent of children in the coffee growing areas of El Salvador, where the World Food Programme distributed emergency food supplies to some 10 000 coffee producing families. A March 2002 survey in Viet Nam showed 45 percent of coffee growing families lacking adequate nutrition. A similar picture of impacts on incomes and rural poverty emerges from case studies in Tanzania and Mexico reported earlier by Oxfam.[6]

Declining prices and export revenues also have macroeconomic consequences. Especially in the case of the highly dependent producers/exporters, declining prices and export revenues, and declining incomes in the coffee sector can have an impact on government revenues. Recent research shows this link continues to be particularly strong for African coffee exporting countries in spite of market liberalization, although there is apparently no significant statistical relationship in Latin America.[7] Clearly, the strength of any such effect is likely to reflect the degree of dependency on commodity exports which is typically higher in Africa. However, more anecdotal evidence appears to indicate that the kind of extreme price falls observed for coffee over the last five years do have macroeconomic impacts elsewhere. In the ICO's survey, Côte d'Ivoire, Ethiopia, Nicaragua and Philippines all reported fiscal constraints on the national investment budget. In the case of Nicaragua, the fall in foreign exchange earnings from coffee amounted to around US$300 million between 2000/01 and 2002/03, while the reduction in income tax receipts from the coffee sector is estimated at around US$13.2 million.

In the wide reporting of the coffee crisis, the media was quick to contrast the plight of coffee growers with the apparent buoyancy of sales and profits in the retail coffee markets in importing countries, prompting questions concerning the producers' share in retail prices. Figure 4 gives one illustrative example of marketing margin behaviour as producer prices fell. It does appear that margins were maintained as coffee prices received by growers fell and hence the share of growers in the final retail price diminished. The latter was already small - perhaps between 20 and 30 percent of the retail price of coffee, and only between one and two percent of the price of a cup of coffee sold in a coffee shop. Of course, the green coffee is only one element in the final retail product which includes costs of processing, transport, services and so on as well as the margins taken by firms at the various stages in the value chain. It would not be expected that transmission of prices through the value chain would be perfect in the sense that changes in grower or world prices would be mirrored by equivalent changes in final prices, especially in proportionate terms. The extent to which world price variations are transmitted depends inter alia on market structures at different stages in the value chain, the technology of processing, and the share of the basic commodity in final products. As noted above, the latter is reduced as processing, packaging and services increase in importance. Nevertheless much concern has been expressed at the "fairness" or otherwise of the small share of global coffee income accruing to growers.

Figure 4. Price spread between US retail price and Colombia producer price

The apparent tendency for falling international and producer prices not to be reflected in prices in final markets, and hence for consumer to world price spreads to widen has a further dimension which is that there is some evidence to suggest some asymmetry in price transmission with a tendency for falling world prices not to be passed on but rising prices to be passed on at least to some extent.[8] The implication of this is that final demand does not rise as world prices fall because the price falls are not passed on into final markets. To the extent that retail demand responds to price changes, demand does not rise to absorb increasing levels of supply contributing further to supply-demand imbalance. At the same time, as noted earlier, the characteristics of the product and the production system are such that falling prices do not provoke significant reductions in supply at least in the short-run. The net result is persistence of falling prices.

3. International action on coffee prices

3.1 Supply control

Depressed prices have prompted calls for international action to address problems of market imbalance with proposals both to control supply and to promote demand. For coffee, the Association of Coffee Producing Countries (ACPC) promoted a retention scheme from 1 October 2000 to retain 20 percent of exports to maintain prices above 95 cents/pound and release supplies onto the market when prices exceeded 105 cents/pound. While 19 countries joined, including non-members of ACPC such as Viet Nam, few actually retained any coffee at all: only Brazil, Colombia, Costa Rica and, temporarily, Viet Nam cooperated. Exports and stocks continued to rise, and prices continued to fall.

Analysis based on a simple partial equilibrium model of the world coffee market developed in the FAO Commodities and Trade Division[9] indicates that if 20 percent of exports had actually been retained off the market in 2001, international prices would have been up to 32 percent higher, and the total export revenue accruing to all exporters would have been 5 - 6 percent higher. However, in spite of this apparently large increase in prices, the specified floor price of 95 cents per pound would still not have been reached, so low had prices fallen. In practice few exporters actually committed to retaining any exports. If Brazil, Colombia, Costa Rica and Viet Nam had actually implemented the 20 percent retention world prices would have risen by around 17 percent. However, this would not have compensated for the revenue loss due to the reduced volume of exports, and the revenue accruing to these countries would have fallen by about 6.5 percent. On the other hand, those countries not participating in the scheme and maintaining export volumes would have increased their revenues by 17 percent in line with the price increase. These results are summarized in Table 1.

Table 1. Estimated effects of coffee export retention scheme in 2001


Price effect

Revenue effect

All exporters

+32 percent

+5.5 percent

Brazil, Colombia, Costa Rica, Viet Nam

+17 percent

-6.5 percent


+17 percent

+17 percent

It appears from these results that prices could have been raised significantly in 2001, although not to the target level, even without full participation in the scheme. However, the most active supporters of the scheme would have lost revenues, while the free-riders would have gained. It is not surprising therefore that even those exporters initially declaring an intention to participate withdrew their support. The main difficulty with such schemes is to devise an appropriate institutional structure to maintain general support and compliance and control free-riders, especially where consuming-countries are not party to the agreement, and where financing is uncertain.

Such schemes hark back to the international commodity agreements (ICAs) with "economic clauses" which were widely seen in the 1970s as a solution to the problems of tropical commodities facing weak markets and variable prices. However, at that time support was forthcoming from the importing countries who wished to offset the threat of the use of the newly acquired producer power as revealed in the petroleum and food price peaks reached in 1972-74. Market interventions ended for sugar in 1984, for coffee in 1989 and for cocoa in 1993, while for jute and rubber the arrangements continued until 2000. The ICAs are not now widely regarded as a success, although the coffee agreement did succeed in keeping prices within the agreed range for some time. The coffee agreement also succeeded in raising prices above what they would otherwise have been,[10] and its passing is seen by some as one reason for the coffee crisis. Today existing ICAs focus on measures to improve the functioning of markets, and there is little prospect of the resurrection of their economic clauses.

Nevertheless, interest persists in supply management by producing countries to counter the long-run fall in international commodity prices.[11] The ACPC coffee export retention scheme has already been mentioned, but a similar scheme exists for rubber, and there has been active discussion of the need for such a scheme for tea. These "producer-only agreements" involve export retention or international stock management schemes, or diversion of low quality into alternative uses. However, as the ACPC scheme illustrates, the experience to date has not been encouraging. It seems difficult to maintain the continuing commitment of the parties to the discipline of the agreement, while free-rider problems persist with those suppliers outside. Even so, the issue of market interventions was seriously discussed again at the recent ICO/World Bank round table on the coffee crisis which recorded "A recognition that a totally free market entailed excessive social costs and that some forms of action with an impact on the market might be considered, notwithstanding the fact that finding such a form of action with an agreement between the various parties may be difficult".[12]

In principle, the conditions for a successful - in the sense of raising prices or slowing their fall - producer-only agreement do not appear demanding. The basic requirements are:

The conditions are not prohibitive and the share of trade that a group needs to command (which depends on elasticities of import demand and export supply in non-members) need not be impossibly high to achieve gains in export earnings by withholding some supplies from the market in the short-run. In the longer-run the elasticities rise and with them the critical share required for the successful operation of an agreement, but this should not rule out modestly aimed agreements for a limited number of years. It is not a requirement for an international agreement that it should be designed to last for ever; periodic re-assessments of the membership and tactics make good sense. In any case market intervention cannot be sustained in a one-sided way to counter the tendency for relative commodity prices to decline in the long-run. This can only be achieved by bringing about a permanently improved balance between supply and demand.

The first two conditions are generally relatively easy to fulfil since production of many commodities, although not coffee, is geographically concentrated, and commodity demand is indeed typically inelastic. However, there is a tendency to be overambitious with respect to target prices and to be unwilling to recognize the need to adjust targets in line with changing market conditions, with politics rather than economics governing decisions. There are also difficulties in choosing the currency to denominate the target prices. If the target price is set in US dollars then devaluation of national currencies against this can offset falls in the dollar price. The devaluation of the Brazilian real is one factor which led to growth in world coffee output in spite of falling dollar prices, for example. Above all maintaining commitment, including financial support to establish and implement a scheme, is the most difficult as the experience with the ACPC coffee export retention scheme and the tripartite rubber agreement indicate. The higher the target prices set the greater the incentive for low-cost producers to cheat, and for those outside the agreement to increase their production and market share.

Control of cheating and free-riders is much easier if the agreement has full participation of importers, which, by definition, a producer only agreement presumably does not have. Consuming countries and the multinational trading companies which buy and process many commodities are not likely to favour higher prices, although under the old ICAs importers saw it in their interest to participate. Participation of consumers is not only desirable from the policing point of view, it may also be a legal requirement that importers are represented under WTO rules, although the constraints on WTO members forming producer-only agreements are not entirely clear. Administration of an agreement has also become more difficult after market liberalization which reformed or removed institutional mechanisms for this.

In spite of the continuing interest in such arrangements, it is clear that market intervention of the producer-only agreement type is fraught with difficulties and unlikely to be successful. The ICO has recently launched a Quality Improvement Programme, which although portrayed as a demand enhancement scheme would also have supply side effects by eliminating a certain volume of inferior quality coffee from international markets. The programme was agreed in 2002 under Resolution 407 as the International Coffee Council proposed to prohibit from October 2002 the export of coffee failing to meet specified minimum standards in terms of numbers of defects and moisture content.[13] Exporting member countries are expected to develop and implement national measures to implement the resolution.

If financing does not prove a problem, the scheme should have both demand and supply side benefits since higher quality might be expected to stimulate demand and command a higher price, while the elimination of low quality coffee would reduce overall supply. However, the burden of implementing the scheme will fall most heavily on those producers with the lowest quality at least in the short-run until their quality is improved. In terms of enforcement, coffee failing to meet the specified standards can be refused the ICO certificate of origin by exporting countries, but the cooperation of importers in policing the scheme and informing the ICO of shipments failing to meet the quality standards is purely voluntary. Furthermore, the participation of importers requires their agreement on the quality standards to be specified and a willingness to give up some flexibility in the range of quality entering their blends. More generally, if superior quality is to be demanded and to command a higher price, consumers must be able to recognize quality differences and be willing to pay for them. Quality improvement schemes may therefore need to be supported by educative information and promotion activities.

The impact of the coffee scheme on quality and prices and its costs is to be reviewed in late 2003. However, analysis using the world price determination model referred to above suggests that such a scheme could have beneficial effects. The impact on the demand side of the market is difficult to judge a priori, although estimates of the likely reduction in export volumes might be made. The extent of this reduction depends upon the percentage of production failing to achieve standards. The ICO estimates that around 600 000 tonnes would have fallen below the standard in 2002. However, perhaps 50 percent of this would not have been exported anyway. The model results suggest that withholding this quantity from the market in 2002 would mean that prices would be up to 8 percent higher than they otherwise would have been. This estimate seems in broad accord with the 4.7 percent increase in the average price for 2002 in spite of a 5.3 percent increase in global production. However, as noted above the burden of the scheme will fall most heavily on those producers with the lowest quality. In the case of Viet Nam it is estimated that as much as sixty percent of production in 2001/2002 was substandard.

A similar scheme is under active consideration by the world tea industry. Under this scheme, tea failing to meet ISO standard 3720 would be excluded from the world market. The details, and particularly the policing arrangements, remain to be worked out, and an international working group has been established to develop the proposal. There is some concern that the ISO standard involved may not be sufficiently stringent to be effectively constraining on volumes of inferior quality tea coming on to the world market and hence have little real impact. Most importantly, however, as with the coffee scheme, effective policing to ensure full cooperation is essential. The tea industry is less well-placed to ensure this in the absence of a relevant international body. Furthermore, it requires active participation of importers against a background that inferior quality does find a market.

3.2 Demand promotion

Difficulties in coordinating international action on the supply side have led to interest in demand side measures, and particularly demand promotion. However, from the outset it must be recognized that generic promotion is primarily a means of influencing longer-term trends in demand, not addressing short-run price variability. Nevertheless, there are certain common problems faced by internationally coordinated attempts to regulate supply or stimulate demand. Key amongst these is the need to secure continuing commitment to cooperative activities of participants who may see their interests as competitive. The current ICO promotional work emphasizes activities likely to command general support, and together with the arrangements for finance and organization, reflects a realistic response in the current depressed market conditions. In both respects, the lessons of past experience in coffee promotion are evident. Promotional messages are not the concern here, but rather how to organize and finance promotional programmes where there is not only the familiar free-rider problem, but also where there are conflicts of interest between participants and tensions between generic and "brand" interests. These controversies extend beyond coffee, of course, as evidenced by the current debates and legal challenges to various generic promotion schemes across a range of agricultural products in the United States. In particular the coffee experience offers some insights into how best to deal with the three challenges faced by all promotional programmes: obtaining agreement on programme objectives; generating financial backing for the programme; and sustaining promotional programmes long enough to generate the desired results.

The international coffee industry has a long history of promotional activity, through the Promotion Committee of the ICO which was charged under the terms of the various International Coffee Agreements with the responsibility for undertaking generic promotion for coffee without reference to brand, type or origin. The ICO was most active between 1976 and 1990. During this period over US$43 million was spent on activities and campaigns specifically designed to build the market for coffee. Financial support for market building activities was internationally-based among virtually all coffee producing countries while activity implementation was typically national in conjunction with national coffee associations. The ICO was dominated by Brazil and Colombia, the two primary producers (and campaign financers as monies were raised pro rata by market share). Although monies were raised through a compulsory levy on coffee producing countries that were signatories to the ICO, expenditure was closely scrutinized and members focused on return for investment. Contributions were most difficult for the smaller countries. At the same time, Colombia's contribution was in addition to the monies it was already spending on its national efforts, Café de Colombia. Tight budgets in the late 1980s and early 1990s combined with lack of unanimity among coffee roasters who typically proposed promotion campaign ideas and co-financed them led to the discontinuation of internationally backed generic promotion and emergence of campaigns for specific market segments or national interests.[14]

Stakeholders in the industry must be the primary source of funds for market development activities. There is therefore a close link between the fortunes of the industry, financial support for promotion and the scope and nature of promotional activities undertaken. Promotion has typically been funded primarily by producing/exporting countries, and generally on a pro rata basis relative to market share as in the case of ICO promotional activities. However, at times of protracted low prices the capacity and willingness of producers to fund such programmes may be limited. Consumers/importers may also cooperate in promotion activities in which case there will be a need for a financial formula for cost sharing. Budgetary limitations have led to a need for greater private sector contributions, and the need to attract private sector support is an explicit provision in the most recent ICO promotional plans. Generally speaking, the narrower the financial backing the more focused promotional efforts and greater likelihood of effectiveness. However, the narrower international participation the more limited the funding base and consequently the more limited the possible activity mix and geographical range. It is an ongoing balancing challenge to secure financing and support (especially where each country perceives a commensurate return on expenditure) for promotion campaigns. In such cases high emphasis should be placed on clear targeted campaigns and routine feedback to all participants on campaign progress. Where possible, feedback should identify the return for each country to make it more meaningful to each participant. It also forms an essential tool for participating countries to justify expenditure on generic promotion to their trade and government.

The funding possibilities obviously constrain what can be done and there is little prospect that conventional advertising campaigns to compete with global brands could be mounted or would necessarily find favour with all participants. In the face of financial constraints, including those posed by depressed commodity prices and revenues, promotional activity needs to be carefully targeted and guided by market research. There appears to be much scope for the promotion of demand in producing countries: Brazil successfully raised coffee consumption during the 1990s from around 480 000 tonnes to 750 000 tonnes, and is now second to the United States as a consumer. However, even here, not all market segments offer the same prospects. Increasing consumption among younger age groups in particular poses a particular challenge where per capita consumption is low and heavily advertised soft drinks are the main competitive challenge. Market growth possibilities in the high income developed countries where most coffee has traditionally been consumed are relatively limited: only the specialist coffee markets have seen significant growth recently, again indicating product differentiation as a potentially successful marketing strategy. The same strategic priorities would be appropriate mutatis mutandis for tea. The emphasis must be on general information provision which all participants acknowledge to be of value, or campaigns targeted on specific market segments where those participants with most to gain contribute most. In the ICO's most recent promotional plans the priority is information provision, notably related to positive links between coffee consumption and health, and on targeted markets, especially in producing countries and emerging markets, for which generic programmes need to be developed on a country-by-country basis reflecting the unique characteristics of each market. The expectation is that counterpart funding would be forthcoming, especially for the latter. The basis for promotional activity of all kinds is research and studies related to coffee consumption for which the ICO Promotion Committee also has responsibility. Among producing and consuming countries alike there is common interest in the coordination of a programme of market research to generate an internationally comparable database of information on consumer attitudes and habits, and in the Promotion Committee acting as a clearinghouse for educational, informational and public relations material. Particular points of interest are the challenges of increasing sales to young people lured away from coffee in favour of ready-to-drink, cold beverages, and attacks on coffee on health grounds.

The scope of backing for generic promotion activities is an important determinant of the focus of activities and the marketing message or position. It serves as the basis for setting marketing objectives and targets. If the interests or priorities of backers differ one from another it becomes increasingly hard to secure collective ongoing support, whether that support is conceptual, political or financial. Every marketing proposition must have clear and commonly shared objectives. In the case of Cotton Council International (CCI) responsible for internal cotton promotion, the industry agrees to overall objectives and priorities. Once these are set specific country targets are developed internally as are the activities aimed to address those targets. This information is shared with the industry who agrees to them as a collective approach to promotion. The broader the financial and political support for a generic promotion campaign the bigger the task of gaining support for activity proposals; securing funding; and reporting on how monies were spent and what resulted from that expenditure. Accountability is paramount for generic promotion since all stakeholders need to receive clear, ongoing evidence that monies have been spent efficiently and generated the targeted results. Back-sell communication is essential on at least a semi-annual basis to maintain support for activities that likely take several years to generate significant results.

Regardless of campaign objectives generic promotion campaigns depend on a long-term commitment by backers to allow sufficient time for results to be generated. A minimum of three to five years is a realistic timeframe for most generic promotion activities. Defensive activities that are fundamentally issues management might achieve their objectives in a shorter period. Without exception the longer-term the objectives of the project, the bigger the challenge to maintain political support and campaign funding. Maintaining support over the longer term is a clear challenge for all advertising and marketing campaigns. It can be even more difficult to convey the benefits derived from generic promotion to stakeholders when each has competitive and varied interests. Promotional campaigns need to include demonstrable effective use of funds; commonly agreed goals and realistic targets; specific measurable activities to track campaign effectiveness. In addition, generic promotion campaigns have the challenge of communicating to each stakeholder how those measured results impact the interests of that particular stakeholder to justify ongoing support. However, even the best conceived and managed campaign cannot always anticipate the effect of a changing market and the evolving consumer. For example, Café de Colombia did not expect that the emergence of a strong gourmet following for coffee in the 1990s would undermine its premium position, relegating Colombian coffee to "second best". This single factor caused the Federación Nacional de Cafeteros de Colombia (FNC) to rethink its marketing strategy and the platform for its ongoing promotional efforts for the late 1990s and beyond - despite its success to date in developing a premium position for Colombian coffee over the previous 30 years. An alternative way to view this market development is that the FNC was so successful in its efforts that it set new standards for the industry as a whole which made its current market position redundant.

The experience of the development of ICO promotional campaigns provides some useful pointers to the design and implementation of promotional strategies for other commodities. The difficulty facing collaborative action on the demand side as with action on the supply side is to organize joint activities amongst stakeholders who may see their interests as competing. There is therefore a need to establish clear strategic aims to which all can subscribe, and clear targets based on rigorous market research. Activities in targeted markets need to be adapted to particular local needs in collaboration with local organizations and in these cases those with most to gain should provide the bulk of funding. In periods of declining prices and restricted public budgets it seems inevitable that the private sector must play a greater role in market development activities.

4. Some concluding comments

This paper has looked at the problems facing the international coffee market and specifically at two areas of industry level initiatives to help improve market balance. It is clear that industry level co-ordinated activities are not straightforward to organize, and attention has also focused on a more micro-level at options available to individual producers to secure better returns from the market.

Product differentiation into speciality varieties can achieve premium prices for coffee beans, although such product differentiation opportunities are not open to all. More generally, organic and fair trade products can also command a premium price. The International Trade Centre (ITC) has been active in researching market opportunities in these areas, through the Gourmet Coffee Project, for example.[15] Coffee offers substantial scope for product differentiation in view of its quite different characteristics according to geographic origin. So-called speciality or gourmet coffees often associated with a particular producing region have continued to command a premium and enjoy market growth even when prices in general have been depressed. Of course, not all producers can enjoy the benefits of favoured locations, but alternative bases for differentiation are also possible - environmentally friendly production systems or organic or fair trade, for example. Exploiting such niche markets requires that segments offering higher returns must be identified and targeted, and quality maintained throughout the value chain. It may also be that the market needs to be educated to appreciate and be willing to pay for "speciality" coffees.

In these areas there may be a role for government and for international agencies, but basically they require investment. The problem is who will reap the benefits of any such differentiation - roasters, traders, retailers, governments or growers. A recent study[16] shows that for coffee, while international prices have displayed increasing variability across coffee types as a result of increasing differentiation in final products markets this has not been reflected in prices paid to farmers - in fact variance of grower prices has actually declined. So a growing share of total incomes in the value chain has accrued to economic agents in the importing countries. Fitter and Kaplinsky attribute this to the imbalance in market power between the two ends of the value chain: while coffee growing is typically atomistic following abolition of marketing boards, importing is concentrated with the top five importers accounting for over 40 percent of total global trade, and roasting is even more highly concentrated with the top five roasters in Europe accounting for nearly 60 percent of coffee produced.

In the longer term the tendency towards oversupply in the coffee market can only be addressed by encouragement of diversification out of coffee production at least in marginal areas. Horizontal diversification into alternative crops is the obvious direction, but the important objective is to enhance income and employment opportunities, including outside agriculture. Vertical diversification can be a means of capturing a share of processing and distribution margins which have expanded even as prices of basic products have declined. Such vertical diversification faces tariff escalation and needs to overcome the barriers to entry which are a feature of the concentrated international market.

Ultimately, non-competitive producers must diversify out of coffee production. This will require public assistance to growers to identify market opportunities and to obtain the necessary knowledge, skills and resources to exploit them. Even for competitive producers, price variability and exposure to price risk will remain even if the protracted slumps seen in recent years resulting from global overproduction and stagnant demand may be limited. Attention can then increasingly focus on management of these risks.


Cashin, P., Liang, H. & McDermott, C.J. 1999. How persistent are shocks to world commodity prices? IMF Working Paper WP/99/80.

Fitter, R. & Kaplinsky, R. 2001. Who gains from product rents as the coffee market becomes more differentiated? A value chain analysis. IDS Bulletin.

Gilbert, C.L. 2003. The impact of commodity price variability on government revenues. Manuscript.

International Coffee Organization. 2003a. Impact of the coffee crisis on poverty in producing countries. ICC 89-5.

International Coffee Organization. 2003b. Ideas and initiatives raised at the ICO/World Bank high-level Round Table on 19 May 2003, in search of solutions to the coffee crisis. ICC 88-5.

International Trade Centre. 2000. The Gourmet Coffee Project. May.

Maizels, A., Bacon, R. & Mavrotas, G. 1997. Commodity Supply Management by Producing Countries. A Case Study of the Tropical Beverage Crops. Oxford University Press, Oxford.

Morriset, J. 1998. Unfair trade? The increasing gap between world and domestic prices in commodity markets during the past 25 years, World Bank Economic Review, vol 12 (1998), pp. 503-526.

Oxfam. 2001. Bitter Coffee: How the poor are paying for the slump in coffee prices. Oxfam, Oxford.

Palm, F.C. & Vogelvang, B. 1991. The effectiveness of the world coffee agreement: a simulation study using a quarterly model of the world coffee market. In O., Guvenen, W.C Labys and J-B. Lesourd, eds. International Commodity Market Models. London, Chapman and Hall.

Partial Equilibrium Model of the world coffee market


Behavioural equation/identity


- Production



- Imports



- Ending stocks



- Consumption



- Exports



- Net trade



Each individual exporting country

(20 major exporting countries or regions)

- Production


QP = f(RP(lagged 5 or 3 years), QP(-1), trend,.....)

- Imports



- Ending stocks


ST = f(RP, QC, QP, ST(-1),....)

- Consumption


QC_POP=f(GDPI_POP, RP, QC_POP(-1), trend,.....)

- Net exports



Each individual importing country

(14 major importing countries or regions)

- Production



- Net imports



- Ending stocks


ST = f(RP, QC, QP, ST(-1),....) or exogenous if minor

- Consumption


QC_POP = f(GDPI_POP, RP, QC_POP(-1), trend,.....)

- Exports



Rest of World

- Production



- imports



- Ending stocks



- Consumption



- Exports



Variables are defined according to the following mnemonics. The first three letters refer to the country name according to the code adopted by UNDP. CO stands for coffee. The last letters refer to the variable as follows: QP for production; QC for consumption; QC_POP for per capita consumption; ST for ending stocks; IM for imports; EX for exports; NT for net trade; VST for stock variations; WP for world price; RP for real price; GDPI for Gross Domestic Product Index; GDPI_POP for income per capita; XR for exchange rate; CPI for Consumer Price Index; POP for population.

[1] David Hallam is Chief, Raw Materials, Tropical and Horticultural Products Service, Commodities and Trade Division, FAO.
[2] El Pais, 19 May 2003.
[3] Cashin, Liang and McDermott (1999). Interestingly, this study notes that persistence of shocks is very much less for tea than for coffee - less than one year in fact - with the implication that stabilization might feasibly have net benefits.
[4] Palm and Vogelvang (1991).
[5] International Coffee Organization (2003a).
[6] Oxfam (2001).
[7] Gilbert (2003).
[8] Morriset (1998).
[9] See Annex.
[10] Palm and Vogelvang (1991).
[11] Maizels, Bacon and Mavrotas (1997).
[12] International Coffee Organization. (2003b).
[13] Specifically, exporting members shall not export coffee that: for arabica has in excess of 86 defects per 300 g sample (New York green coffee classification/Brazilian method or equivalent); for robusta, has in excess of 150 defects per 300 g (Viet Nam, Indonesia or equivalent); for both arabica and robusta, has a moisture content below 8 percent or in excess of 12.5 percent measured using the ISO 6673 method.
[14] The cotton and wool industries show a similar experience. In each case there was a significant gap between the contributions of the largest backer and other programme participants, and the disparity between countries' respective contributions led the dominant contributor (the United States in the case of cotton and Australia in the case of wool contributed 40-50 percent of the total budget) to believe that although the total market size may be increasing, other countries stood more to gain by their financial investment and to withdraw their support.
[15] International Trade Centre (2000).
[16] Fitter and Kaplinsky (2001).

Previous Page Top of Page Next Page