Structural Problems of West African Cocoa Exports and Options for Improvement

Structural Problems of West African Cocoa Exports and Options for Improvements

Hans H. Bass

Introduction

Cocoa production and trade has an ingloriously prominent position in Africa's Primary Commodity Dilemma. Adverse price movements for the raw beans as well as high barriers to entry in the markets for semi-processed and final products, along with a history of failed national and international interventions do not offer much hope for the future. Nevertheless, there are some policy approaches to cocoa production and processing, which seem to be more promising than others. In its first part, this paper discusses the (many) constraints and (few) opportunities for West African farmers and companies in the cocoa business; in its second part, it discusses policy options to improve the position of West African countries and firms.

The present paper draws from insights derived from the neo-institutionalist framework of "global commodity chain" (GCC) analysis and the "filiére" approach (see Raikes/Jensen/Ponte 2000; Kaplinsky 2004; de Lattre-Gasquet/Desprèaux/Barel 1998). While conventional trade theory neglects economic power, time, and institutions, in value chain analysis, in contrast, the access to specific technologies, the established market powers, and commercial ties based on the investment of material resources and time are taken into consideration. Not only does this method help to better understand the patterns of international trade, but it also offers new insights into specific policy options for rent shifting or the upgrading of production.

According to the GCC approach, the various phases of physical transformation and service inputs in a commodity chain are typically subject to chain governance with an unequal distribution of power among the firms involved (lead firms, subordinated firms). A main distinction is made between buyer-driven GCCs and producer-driven GCCs. Buyer-driven GCCs have low barriers to entry in the first stages of production, while the buyer firms control processing standards, design, brand names and retailing. Probably without exemption, (tropical) agricultural commodity chains are buyer-driven, and cocoa figures as a prominent case. The filière approach is less concerned with power distribution in vertical integration but with issues of transaction costs. It argues, inter alia, that too much market liberalisation in the various stages of processing can have negative consequences on the efficiency of the whole chain.

1. The cocoa / chocolate commodity chain

The value adding chain in the cocoa/chocolate business can be divided into five stages: beans production (including roasting), grinding and producing semi-finished materials (grinding beans to cocoa liquor, cocoa butter, powder, and cocoa cake), producing industrial chocolate (vending mixes, compounds/fillings), production of consumers' chocolate, and retailing.

1.1 Cocoa Production and Processing

Cocoa is produced in more than 50 countries; 90 p.c. of world production, however, is concentrated in only 10 countries, including Côte d'Ivoire (accounting for 40+ p.c. of the world's cocoa production 2000/01 to 2002/03), Ghana and Indonesia (about 15 p.c. each), Nigeria, Brazil, and Cameroon (about 5 p.c. each) (ICCO 2003, p. 19). In most producer countries, and especially in the African ones, the production is extremely dispersed. Most cocoa farms are very small and are mainly based on family labour. For instance, in Côte d'Ivoire some 900,000 farmers grow cocoa, most of them producing only a few hundred kilos of cocoa per year (ICCO estimates, see FCC 2005).

Raw cocoa is a relatively homogeneous commodity: more than 90 p.c. of world production is in standard beans; the rest, from the superior variety of the criollo cocoa tree, is basis for fine cocoa powder (produced in a number of Asian and American countries, see Annex C of the International Cocoa Agreement 2001: UNCTAD 2001). However, standard beans are of different quality levels depending on both natural conditions and the diligence and knowledge of producers and transporters. Quality differences result in mark-ups and mark-downs of the world market price of raw cocoa beans according to the area of origin. While cocoa from Indonesia or Brazil, for instance, is traded on the London and New York Exchanges with mark-downs of up to 10 p.c. of the world market price, cocoa from Ghana normally generates the highest premium.[1] The quality reputation of a country has therefore been interpreted as a public good to be provided by government's quality controls. However, new transport methods (loose bulk shipment, ship-in-ship transport) and new grinding technologies let this premium erode: The demand for beans, which can be ground without post-processing, decreases as it becomes more important to ship large amounts of beans even if of different size and quality.

Over time, a country's production volume and its share in world production of cocoa may change considerably. Firstly, weather and pest related year-to-year harvest fluctuations are strong (up to ±20 p.c.). In the past, political events and civil unrest (such as in Côte d'Ivoire in 2004) have also contributed to irregular declines of cocoa yields. Secondly, regional production follows pronounced cycles of several years (see Graph 1), reflecting the life cycle of the cocoa tree: Conventional tree types need 5 years before bearing fruit, reach a maximum yield after 11 years, while 20 years after planting yields begin to diminish by 3 p.c. p.a.; hybrid varieties, developed in the 1970s, have a somewhat shorter life cycle. Cycles usually start a few years after significant increases in prices have triggered off new planting. A prominent case in point was the promotion of cocoa planting by national governments, the World Bank and the EU as a lagged reaction to the very high prices in the 1970s (Kofi 1993). In Ghana, for instance, a new production cycle started in 1985, seven years after the world price for cocoa peaked in 1977. This pro-cyclical behaviour was partly responsible for the late 1990s' collapse of cocoa prices. On the other hand, as a reaction to the collapse of cocoa prices in 2000, West African countries simultaneously decided to anti-cyclically promote cocoa planting – a decision which again can trigger off oversupply in 10 years time.

Graph 1: Multi-annual cycles of cocoa production, various countries, in tons, 1971-2002/03.

Thirdly, since the mid-20th-century world cocoa production has exponentially increased (see Graph 2) due to productivity gains, mostly from hybrid trees, and the emergence of new producer regions. Over the past century there seems to be a rotation of major crop growing areas around the equator (see Ruf/Siswoputranto 1995), which in the 1990s had manifested itself in the (re-)emergence of Asia (especially Indonesia) as a cocoa growing area without other areas ceasing their production (Malaysia most recently being one of the rare counter-examples). As more and more countries have taken up growing cocoa, this commodity may, within a few years time, even fit into the general picture of African economies falling behind other regions of the world in its relative share in the export volume of primary commodities (for the general picture see UNCTAD 2003, p.6).

The growth in world cocoa production and the constraint of a relatively stable demand, which is discussed in more details below, contribute to a structural "over-supply" of raw cocoa. Yet, the cocoa market is a "non-dynamic market" in terms of participation in the present expansion of world trade: growth rates of world cocoa exports averaged 1.6 p.c. (1980-2000), as compared to the 8.4 p.c. growth of overall world exports (UNCTAD 2003, p.9-10).

Graph 2: Production of Raw Cocoa (World; o/w Côte d'Ivoire, Indonesia), tons, 1971-2002.

The particular characteristics of the cocoa tree and the limited storage possibilities lead to inelasticity of the volume of raw cocoa supply to changes in labour and fertilizer inputs. Given medium-term oriented decisions as a reaction to substantial price movements, there is no effective instrument to significantly reduce short-term supplies to put an upward pressure on prices. The reduction of labour input, of fertilizers and herbicides, (if used at all), will normally only lead to insignificant yield decreases; lack of care results in quality deterioration. Thus, there is no incentive for the individual farmer to produce less in times of price plunges – a fact which led West African governments to consider crop destruction as an instrument to stabilize prices in 1999/2000 (Maes 2001, p. 92).

The following stages of the value-adding chain "from bean to shelf" are considerably vertically integrated, reflecting static and dynamic economies of scale. A few trans-national trading companies dominate the shipping of raw cocoa (Cargill, Archer Daniels Midland, and Barry Callebaut; see Vorley n.d.). A dozen TNCs controls three-quarters of the world's processing of raw cocoa beans; the Top-Six (Archer Daniels Midland, Cargill, Barry Callebaut, Blommer, Petra Foods, and Nestlé) grind more than 50 p.c. of the global production of raw beans (ICCO 2005).

Today, two third of the processing of raw cocoa beans is done in the consumer countries including The Netherlands (15 p.c. of world grindings), the USA (14 p.c.), and Germany (6 p.c.). Producer countries with a significant share in grinding are Côte d'Ivoire (10 p.c.), Brazil (7 p.c.), and Indonesia (4 p.c.). The grinding in Ghana, Cameroon, and Nigeria together accounts for only 5 p.c. of world production of semi-processed cocoa (all data for 2001/02: ICCO 2005). Nevertheless, in the last few years there has been some small progress for the cocoa producing countries to localize at least the first stage of cocoa processing. The four major European importing countries (Netherlands, Germany, UK and France) in 1991/93 imported only 20 p.c. (3-yrs avg.) of the total weight of cocoa in form of semi-processed goods (cocoa paste and butter), while this share rose to 31 p.c. in the three years 2001/03 (see Graph 3).

Graph 3: Combined imports of Netherlands, Germany, UK, and France in raw and semi-processed cocoa, in ,000 tons, and weight shares of semi-processed, 1991-2003.

Generally, tariff escalation contributes to the unfavourable situation of developing countries in the export of semi-processed and processed agricultural products. For instance, in the European Union import of cocoa beans is generally tariff-free, while tariffs for cocoa paste and cocoa butter from Indonesia were raised at 6.1 p.c. and 4.2 p.c. in 2004, respectively (EU 2005). West African countries, however, are tariff-exempted in the EU due to the Lomé-Côtenou Treaties. The progress in upgrading of cocoa production made in West Africa, however, is even less satisfying considering the fact that grinding facilities in West Africa are almost exclusively owned by the same TNCs, which dominate the international cocoa business, and not by joint ventures or national proprietors (see Table 1) – thus reducing possibilities for technology transfer.

Table 1: West African Cocoa Processing Sites

Seen from the West African perspective, the major producer countries (Côte d'Ivoire, Ghana, Cameroon) have not managed to export more than approximately 15 p.c. of the weight of their cocoa exports in form of semi-processed material (computed with data from UN 2005), which is about 20 to 25 p.c. in terms of value (see Table 2). Although there has been some change in recent years, generally the majority of cocoa exports from West Africa is still in an unprocessed form. The governments through tax incentives encourage local processing. In Côte d'Ivoire, the export tax for beans is 125 CFA francs, for cocoa butter the tax is 109 CFA per kilo, for liquor 60 CFA/kilo and for powder 12 CFA/kilo (UNIDO 2001).

Table 2: Exports of raw cocoa, semi-processed, and processed cocoa products, West African countries, 1997-2003 (mill. US$ and shares)

In the cocoa business, global sourcing is necessary: Most manufacturers of chocolate blend cocoa beans from various origins. Although this may decrease the possibility of taking advantage of specific features of some proveniences, it helps to significantly reduce supply risk and ensures constant standard quality. The main suppliers of the final products are also regularly involved in other segments of food processing, which in turn results in cheaper access to other kinds of ingredients for the final consumer product as well as opening outlets for semi-processed cocoa into the companies' own lines of production. Half a dozen TNCs (Mars, Nestlè, Cadbury, Ferrero, Hershey, Kraft Foods) provide three quarters of the world's supply of chocolate products (ICCO 2005).

In recent years, final demand for cocoa was rather stable in many high-income countries (see Graph 4). On some markets, including Germany, demand even declined in the 1990s (-2.8 p.c. per year; FAO 2003). In general, income elasticity of demand is estimated to be smaller than unity (.92 in the US, .52 in Germany, and .02 in Japan; per capita consumption, estimates for the early 1990's; ICCO 2005). Price elasticity of demand is probably slightly larger than unity in most High Income Countries. Along with market saturation, and low or even negative national income growth, also recent changes in food regulations may have contributed to the present sluggishness of demand in some consumer countries. The EU reduced the requirements for cocoa content of chocolate (EU 2000) and authorizes the addition of a maximum of 5 p.c. of a vegetable fat other than cocoa butter to chocolate. Substitutes, such as shea butter, are generally cheaper than cocoa butter. The directive entered into force in August 2003. It has been estimated (ICCO 2005) that as a result, cocoa demand may decline by up to 200,000 tons per year (i.e. about a fifth of European cocoa import), but it is yet too early to determine the real consequences. However, figures from 2002 and 2003 (see Graph 3) do hint at a decrease in the EU's cocoa imports. On the other hand, the new directive may also open up new possibilities for product innovation, as new features could be added to the final products, and thus increase consumer demand.

Graph 4: Chocolate consumption in important consumer countries, kg per capita (1997 and 2002) and GNP per capita (1999).

The market situation puts a great deal of pressure on existing companies in the cocoa and chocolate business to reduce costs and at the same time increase market share by innovation in products (shapes, packaging), processes (recipes), and markets (esp. the still "unexplored" markets in Eastern Europe and Asia/Pacific). Thus, expenses for research and development as well as brand marketing regularly make up a high percentage of the total expenses of these companies. The high levels of product innovation and marketing expenditure necessary to maintain consumer demand dictate that it is the brand owners, rather than the retailers, who steer the market.

In the chocolate business, knowledge across the value chain is extremely "tacit". Some up-market companies even introduced new products only after having been able to tap adequate human resources (especially Swiss trained specialists). Furthermore, the markets for chocolate are geographically segmented, i.e. there is a "Cadbury World" in the UK, Australia and New Zealand, a "Ferrero / Kraft Foods / Nestlé World" in Continental Europe, and a "Hershey / Mars World" in the US and Canada. In the UK Nestlé, Cadbury and Master Foods (Mars) control 80 p.c., in the US Hershey, Master Foods (Mars) and Nestlé control 75 p.c. of the chocolate confectionary market (ICCO 2005, Cadbury 2003). Three-firm concentration ratios in other major consumer countries are between 60 p.c. (France) and 90+ p.c. (Italy, Ireland) (Kaplinsky 2004). The leading suppliers have shaped consumers' tastes to a degree that makes it very difficult even for the established companies to gain a foothold in the other markets. Recently, especially in up-market niche segments, some gains could be made by exploiting competitive advantages in terms of pure origin chocolates, organic production labels, and fair trade labels – attributes not only applied by niche manufacturers, but also by market leaders (for instance, Barry Callebaut includes fair trade chocolate in its product range).

This said, the "governance" of the cocoa value chain may be described as bipolar: a stronger focus on the grinders, a weaker one on the chocolate manufacturers (Kaplinsky 2004). Vorley (n.d.) describes the cocoa GCC as bottleneck-shaped: 14 million workers in the cocoa sector worldwide and thousands of local collectors and traders form the basis, while the middle part of the chain is either oligopolistic-oligopsonistic or fully integrated, and at the top-end there are hundred of thousands of retailers and millions of consumers.

1.2 Prices and Earnings

Supply and demand conditions as outlined above have led to a downward trend in nominal prices of raw cocoa. Nominal world market prices for cocoa considerably increased during the 1970s, but slumped thereafter, including a three-decade low in 2000 (see Graph 5). Real raw cocoa export prices (i.e. the index of nominal price deflated with the index of export price of manufactures from advanced economies) deteriorated from an all-time high of 387 index points in 1977 to below 100 since 1988, with minima in 1992 (53 points) and 2000 (50 points). This fact lends strong support to the Prebisch-Singer hypothesis of a secular decline in the commodity terms of trade for raw materials for the last third of the 20th century. There was also a considerable degree of price volatility in the raw cocoa market in the 1970s/1980s. However, price instability indices for all raw materials have moved downwards in recent years (see Table 3), probably due to both the already low level of prices and the enhanced use of new financial instruments (hedging and trade on the forward market; see ICCO 2005).

Graph 5: Annual Averages of Nominal Daily World Market Prices* of Cocoa Beans, (London/New York, US cents per pound), 1971-2004 and Deflated Prices (1995 = 100).

Table 3: Price instability indices of agricultural commodities, 1964-2003

The total number of people living from the cultivation of cocoa is estimated to be 3.6 million in Côte d'Ivoire, 3.2 million in Ghana and 1.6 million in Cameroon (ICCO estimates on "workers" involved in cocoa, see FCC 2005) – i.e. between 10 and 20 p.c. of the respective populations lives from the cocoa sector. Cocoa accounts for a large share of total household income among cocoa farmers, usually more than 50 per cent; yet average annual cocoa revenues are low, national averages ranging from US$ 30 to US$ 110 per household member in the four major cocoa producing West African countries (IITA 2002). Therefore, it is no surprise that today poverty is prevalent among the cocoa farmers (quite in contrast to the "golden" 1960s/1970s). For instance, in Ghana one-third of the export crop farmers (i.e. mainly cocoa farmers) are poor in terms of an absolute poverty definition (see TUC-Ghana 2004). The cocoa farmer, however, has often already tapped an even weaker source of value production: child labour (see IITA 2002; Anti-Slavery International 2004).

Not only employment depends on cocoa, but also export earnings (which co-determines import opportunities) – 30 p.c. in Côte d'Ivoire, 20 p.c. in Ghana (computed with data from UN 2005). Thus, ensuring remunerative prices for cocoa, and, considering the pronounced downward trend in the real prices of raw cocoa, the upgrading of West Africa's cocoa exports, is not only a key to immediate poverty reduction but also to economic development in general. Policy options to achieve this goal will be discussed in the following section.

2. Cocoa policies

Analogously to the development policy paradigms regarding the respective roles of state and market, to date there have been three different approaches to national and international regimes for the cocoa sector: (1) pro-interventionist, (2) pro-market, and (3) a "market-informed state" approach.

(1) The pro-intervention approach meant to stabilise and increase farmer's incomes, not least by guaranteeing high export qualities. On the national level, physical interventions by marketing boards (in Anglophone West Africa), and regulative interventions by caisse de stabilisation (in the Francophone countries) can be distinguished. The clearest expression of state intervention in the cocoa sector can be seen (to a certain extent even today) in Ghana. Until 1993 local buying of fermented beans was exclusively done by the para-statal cocoa board (Cocobod), which was responsible for all stages of marketing, and which to date still determines the basic buying price. Since the beginning of liberalisation a dozen private buying companies exist, which so far could attract about one third of the market volume but do not yet export. The main steering instrument of Cocobod is an export tax, which only slightly fell from 50 p.c. in the mid-1990s to 30 p.c. today, and which contributes to a significant difference between farm gate price and f.o.b. export price. Consequently, the Ghanaian cocoa farmer's share of export price remains much lower than the Southeast Asian producer's share, which was 90+ p.c. already in the 1990s (Akiyama / Nishio, 1996). This government intervention can be justified as a move to reduce volatility of income. Although it is difficult to calculate exact figures due to the quick erosion in purchasing power of the Ghanaian currency, the Cedi, in the period 1991-2002 the coefficient of variation of average annual producer prices in US$ seems to have been even higher than the coefficient of variation of world market prices (26 p.c. and 20 p.c., respectively; own computation with data from FAO 2005, ICCO 2005, exchange rates: yearly averages as given by the Bank of Ghana; for a somewhat different view see ul Haque 2004).

On the international level, neither did agreements to limit production or to set up physical buffer stocks enter into force, nor did hopes to form an effective producers' cartel within the framework of a New World Economic Order materialise (Gilbert 1996). An explanation may be that the largest producer country (Côte d'Ivoire) did not have sufficient incentives to join the agreements.

(2) On the other hand, the retreat of the state from the cocoa sector during a phase of low prices along with the liberalisation of the cocoa trade obviously could not increase farmers' income. For the future, it has to be considered that the situation for African exporters will not continue as relatively favourable as it is now, as the preferential treatment for West African exports to the EU vis-à-vis the rest of the cocoa-growing countries will lose its significance in the course of WTO-induced tariff harmonisation.

In addition, the hopes for international capital inflows, which would enable African producers to climb the value chain, has not materialised so far. The high demand for specific technical knowledge and R&D in the business, for economies of scale and scope to reduce costs, and the necessity for marketing power to put the final products onto the supermarkets' shelves and to the consumer – brand aware, price conscious, quality demanding – generally seems to make it difficult for companies from the developing world to enter the chocolate business beyond the first step of grinding beans. However, as long as processing facilities in West Africa are wholly foreign-owned, even here technology transfer is limited. Yet, even attracting FDI to allocate at least part of the grinding in Africa in several cases has produced net losses of resources to African countries due to unfavourable accounting practices of TNCs (see Frankfurter Allgemeine Zeitung, 10.02.2001).

The clearly segmented markets for final products again makes efforts of regional co-operation between chocolate producing enterprises in the Anglophone and the Francophone part of West Africa difficult, as no single target market exists. Also the fact that producers of semi-finished and final products normally need to operate globally in order to reduce the risk of regional harvest vicissitudes and to guarantee a standard quality by blending, tends to limit any regional co-operation and calls for action on a global scope.

The present Sixth International Cocoa Agreement (of 2001), however, has not been able to offer both market-friendly and effective alternatives, as it limits its scope mainly to the promotion of basic research, agricultural and industrial pilot projects, and promotional efforts to increase final demand – a difficult task, as only new market trends in high-income countries, such as demand for multi-sensorial designer food (such as combinations of healthy fish oil and tasty chocolate), the economic growth in middle-income countries such as East European countries, and the catching-up of low-chocolate consuming countries like Japan may increase world-wide cocoa demand on a substantial scale.

(3) As neither pure-market nor pure-interventionist instruments have proven successful, more "market-informed state" solutions seem feasible for West Africa's cocoa sector (see World Bank 2000b for a general overview on this approach). In this view, producer associations and public bodies play leading roles in facilitating new ways for the economy without directly intervening into the markets. Such action in the cocoa sector can include:

+ institutional support to target non-traditional markets by turning a present weakness into a future strength. Rather than extensifying cocoa production (more quantity, less quality – the present strategy in Ghana; see Teal / Vigneri, 2004), which leads to significant enterprise-level and ecological follow-up costs, West African governments should encourage producers to build on the present low yield / high-quality cocoa as a result of the presently low inputs of synthetic fertilizers and pesticides, and add further characteristics, such as bio-organic/eco, pure origin and fair trade certifications. This includes the implementation of quality standards (industrial, ecological, social), resulting in a strong emphasis on serving market niches and a decisive quality emphasis in producer assistance;

+ further encouraging the development of non-traditional uses for cocoa, i.e. non-confectionary uses. As to date there is only one ICCO-initiated pilot plant still in the planning phase for the processing of cocoa by-products, using material from the cocoa pod, which is normally discarded (ICCO 2003), there is much more scope;

+ encouraging the export-crop farmers to diversify into non-traditional products (such as pineapple, mango) to generate risk insurance by diversifying crops. This means that there has to be a substantial speeding up of the little progress, which had been made so far in this respect (see for instance for Ghana: GNDPC 2004, p. 63-4).

The specific characteristics of cocoa, the structural over-supply, the technologies used in processing, and the power distribution along the value-adding chain make it difficult for West African producers to improve their position. Nevertheless, some leeway for action exists.

References