TRENDS IN PRODUCTION, TRADE, AND UTILIZATION

Country Briefs

For this discussion we highlight the seven countries with the highest current area planted to cassava: Brazil, Colombia, Cuba, Haiti, Paraguay, Peru and Venezuela. These countries produce 97% of the region's cassava (Table 3).

Brazil

The region's largest country has been at, or near, the top globally in total cassava production, probably since the crop was first cultivated. Cassava is a major crop in three of the eco-regions: lowland humid North (19% of production); the dry Northeast (46%); and the subtropical South (21%). The picture is completed with minor production in the acid soil, wet/dry savannas of the Central-west (Campo Cerrado) (4%); and in the sub-humid Southeast (9%). National production has varied little over the last 20 years, staying near 20 million tonnes per year. However, in the North (Amazon basin) production has more than tripled in the past 25 years, an outcome of cassava's role in frontier agriculture.

In Brazil bitter and sweet types of cassava are considered essentially different crops - aipim (sweet) and mandioca (bitter). Most of the cassava is of the bitter type (high cyanogenic potential), and is processed prior to consumption. The main product is a coarse, toasted flour (farinha), the principal carbohydrate source of the poor, and a complement to many other dishes. The South leads in starch production for food and industrial use (30 000 tonnes annually), and for on-farm feeding of fresh roots. A nascent animal feed market for dried chips is developing in the Northeast, where growing grains is risky, and shipping grains from the South is costly. Brazil has a strong national research program and a network of state programs working to improve cassava systems. Work is skewed toward the production side, although it has shifted to more emphasis on processing and marketing in the past decade. The National Cassava and Fruit Research Center (CNPMF) in Bahia State holds the world's largest national program collection of cassava germplasm. The National Center for Genetic Resources and Biotechnology (CENARGEN) includes cassava and wild Manihot species within their mandate. Brazil recently has taken a strong leadership role in adaptation of farmer participatory techniques for technology development and diffusion (Pires de Matos, et al., 1997). Some of the world's more advanced cassava agriculture is found in Brazil's southern, subtropical region. Local research and extension programs have been working here since the early 1940s to improve cassava, and the results are illustrated by highly productive systems, with yields averaging 17 tonnes per hectare, and up to 25 tonnes per hectare in intensive systems in Paraná State.

Colombia

Perhaps the Latin American country with the highest agro-ecological diversity, Colombia hosts a wide range of systems for cassava cultivation and utilization. The highest proportion of production (45%) comes from the seasonally dry and semi-arid Atlantic coast region. Another 25% is produced in inter-Andean valleys of the eastern mountain range, and 17% in the central part of the country. The eastern, acid soil savannas (Llanos Orientales) and the high rainfall Pacific coast are minor producers at 9 and 4%, respectively (Balcazar V., 1997). Along with this diversity of environments comes a diversity of biological problems. All but a few of the total range of pests and diseases that affect cassava worldwide are endemic in Colombia. This creates a challenge for growers, but is also an opportunity for researchers to capitalise on hot-spot environments for selecting host plant resistance. Also, many of the natural enemies of pests and pathogens thrive here and can be exploited in research and production. Except for the Amazon and Llanos regions, most varieties are of the low cyanogenic potential type, and consumed fresh. In the mid 1980s, Colombia recognised the potential of cassava as a substitute for imported maize and sorghum in balanced animal feed rations, and began a program involving a range of R&D institutions and farmers' groups. This was the first implementation of the integrated cassava research and development project concept, a development model based on simultaneous work on improving production efficiency, new products and processing methods, and expanded markets. These projects first concentrated on the animal feed market, but later included projects in fresh cassava, starch and flour. The combination of these initiatives contributed to an upturn in production, from 1.3 million MT in 1987 to 2.0 million MT in 1999. Colombia is host country for the International Center for Tropical Agriculture (CIAT), and has both contributed to the global cassava initiatives of this institute, as well as benefited from its presence. As participants in a laboratory for testing of new technology, Colombian farmers have had the opportunity to be early beneficiaries. Some of the original work on basic agronomic practices (stake selection and treatment, planting position, plant density, and herbicides) led to recommendations that were quickly and broadly adopted in parts of Colombia. Because of a collaborative, extensive varietal testing network, Colombian institutions had an advance look at some of the new materials, and released several for farmer use.

Cuba

Cassava production in Cuba follows two very distinct general forms: the large state farms where a high level of technology is applied, and the small private plots that are becoming more common. The state-controlled system allows experiment station technology to be transferred almost immediately to production fields. These farms often have high input systems for cassava, including mechanised land preparation, planting and harvesting; herbicide and fertilizer applications, and even irrigation. Sometimes the inputs were not economical but were subsidized by the state and helped farms meet their production goals. More attention is now given to the economics of production - the use of inputs to produce a profitable output. In spite of this emphasis on technology, however, yield levels have been disappointing, and in fact some of the lowest in the region. Part of the reason is the trend toward growing cassava on ever-poorer soils, and shortages of inputs. Fertilizers and herbicides are often diverted to higher value crops. Most of Cuba's production is used directly for fresh consumption, and most varieties are of the table type. One of the early research successes was to develop a system to extend the period when fresh roots are available on the market, by combining a specific set of varieties with differing maturities, and staggered planting dates. Cuba has been promoting research on use of cassava in animal feed rations, but this is not yet a major market. Cuba's research and extension system has been among the most consistently productive of the region, with a long-term, well-balanced interdisciplinary effort. The program has developed packages of agronomic practices, new varieties, and pest control systems.

Haiti

Haiti is the poorest and most densely populated country of the region, and one where cassava is becoming more important in the diet. Cassava is processed and baked to make the traditional Caribbean form of large flat bread. Most varieties are of the bitter type. Between 1970 and 1995, per capita annual consumption increased from 31.6 to 35.1 kg, while all other countries of the region saw declines. This increasing dietary role is, unfortunately, being driven largely by the effects of land degradation, with more marginal soils brought under cultivation, and degradation of existing cultivated land. Erosion and nutrient depletion are severe on the hillsides and cassava is one of the few crops with dependable yields. In spite of these dire needs in a crop of increasing importance, Haiti has almost no research capacity. The language barrier makes it difficult for Haiti to participate fully in the international networks involved in cassava, or to take advantage of much of the existing research literature. This is a country that desperately needs a substantial R&D effort on cassava, as a means of raising living standards of the rural poor. What little research has been done has been mainly sponsored by NGOs, with a short to medium-term perspective, and insufficient local support.

Paraguay

Paraguay maintains a very strong cultural attachment to cassava as part of a history that goes back to the Guarani Indians' reliance on this crop. Production more than tripled in the period from 1961 to 1999. In the peak years of the late 80s, production reached almost 4 million tonnes. This high level of production, and the relatively strong agricultural sector in general, would seem to favour cassava's increasing move toward use as an industrial raw product, such as animal feed or starch. But since maize is also a major crop here; both compete in several of the same markets. Paraguay is the most rural of South American countries. Over 60% of land area is agricultural, and nearly as many people live in rural areas as in the cities, in contrast to most of South America where 3-7 times more people live in cities. Following from this, it also has the highest share of agriculture in the GDP - 26% (compared to its neighbours, Brazil, 9.6%, and Argentina, 6.0%). Now, however, with one of the highest urban growth rates in South America (4.5% per year) the dynamics of urbanization are likely to drive some of the same trends as elsewhere on the continent - a move toward more industrial uses, with a continuing decline in food uses. The cultural importance of cassava has not translated to a concerted institutional interest. Resources for research have been somewhat irregular, depending very much on the personal interest of individual ministers of agriculture. The extension service has been active both in research and extension, filling a void where no separate research effort was organized for cassava. The 1980s was one of the most active periods, with work on germplasm collection, agronomic practices, pest control and utilization. Recently there is a renewed interest for cassava, especially regarding the potential of value-adding to increase demand and farmer income.

Peru

Most of Peru's production is in the eastern side of the country - in the rainforest, and on the lower slopes of the Andes. The coastal area, with most of the population, relies almost totally on irrigation, and therefore grows higher value crops. As elsewhere in the Amazon basin, cassava is a staple. Production has been relatively stable for the past 35 years, with some rising trend in the 90s. As pressure on land increases, making slash and burn systems more difficult to sustain, those populations near urban markets are looking to intensify and commercialize their agriculture. Some cassava in fact is shipped across the Andes to Lima, mainly for recent immigrants to the city who retain a preference for cassava over more accessible potatoes. Because of isolation of most farms from major markets, future development will need to focus on internal markets close to production areas. Since most is grown in humid rainforest environments, opportunities for dried chips for animal feed are limited, at least with natural drying. Starch and flour are possible options for value-added products. Peru has very limited research capacity in cassava, and does not have any nationally coordinated effort. Projects in processing for animal feed, flour for partial substitution in bakery products, and marketing of treated and bagged fresh cassava have been some of the principal thrusts of recent years.

Venezuela

As a major oil exporter, the share of agriculture in Venezuela's GDP is quite low - only 5.0%. With the global oil boom beginning in the 70s, Venezuela lost some of its motivation to pursue agricultural development, and rural areas not linked to spill-over from oil income suffered the consequences. Land use is also low - about 25% of the total area is in agriculture. Nonetheless, area planted to cassava, and production, have been relatively stable in the past 35 years. This reflects the fact that most farmers who rely on cassava do not have many other options for income. Most of the growing areas are drought-prone (coastal) or have acid, low fertility soils (savannah and rainforest). Various private companies in the past have tried to establish a starch industry based on cassava, but these have either failed outright or had limited success. Currently, other agro industries are applying the lessons of past failures, and are working in a more integrated manner to coordinate production with processing capacity and market demand.

Venezuela had a strong cassava research and training program during the 1960s and 70s based at the Central University at Maracay. This group initiated countrywide work in germplasm collections and evaluation, production practices, and especially developed expertise in the areas of utilization in animal feeding and pest management. Currently, the private sector is sponsoring a modest but effective research program on cassava, aimed mainly at production for starch.

Production Trends

In comparison with Asia or Africa, production trends for cassava in Latin America have been quite stable over the past twenty-five years. Brazil accounts for most of the aggregate variations. This country strongly dominates the Latin American cassava scene, with over three-quarters of the region's current production. Most countries have had gradual tendencies to increase or decrease production, but few have realized dramatic shifts due to major production or market forces (Figures 2 and 3). This is to be expected in the traditional production systems and constrained markets that characterize most of the region.

Figure 2. Aggregate area and production for cassava in Latin America, 1961-1999

Source: FAOSTAT

Aggregate production since the 1960s can be broadly characterized in three phases. Between 1961 and 1972, there was a marked increase in area and production, in response mainly to continued population growth. Area harvested peaked at 2.85 million hectares in 1977. In the period from 1977 to 1984, area planted steadily declined as the full impact of wheat import subsidies and other policy disincentives were translated into reduced consumer demand for cassava. Since the mid-1980s, area has been relatively stable, with only a small decrease to the 1999 level of 2.42 million hectares. Production climbed more sharply than area during the 60s, but then continued roughly in parallel for the next 30 years. The next decade should show increasing yields, as the adoption of productivity-enhancing technology accelerates in the region. The past twenty years show a trend toward wider cyclic variations in area and production as compared to the previous two decades. Specific causes of this fluctuation are difficult to pinpoint, but may be related to uncertainties in the cassava marketplace as agricultural and trade policy in the region undergo reform and adjustment.

Production Systems

Cassava is nearly always part of a farming system that includes other crops or animal components. System characteristics are associated with environmental influences, economic constraints and opportunities, and cultural traditions. Cassava can tolerate long periods of drought after it is well-established, but must be planted during a period of adequate soil water. In most systems, growers plant the crop near the beginning of the rainy season. Cassava is slow to develop a canopy, and early weed control is crucial. This is accomplished mainly by hand weeding, but use of herbicides is making inroads. Farmers rarely apply inputs to control pests or diseases.

Since there is no sharply defined maturity period, harvest may extend over several weeks or even months, determined by utilization systems. As the level of drought and soil fertility or acidity stress increase, cassava tends to become a more dominant component of the cropping system. Thus, in fertile inter-montane valleys of the Andes, cassava is one of many crops. In the semi-arid interior of northeast Brazil, or in the acid soil rainforests, cassava plays a dominant role.

Figure 3. Cassava production trends in major-producing countries of Latin America (logarithmic scale); 1961-1999

Source: FAOSTAT

In some regions there is seasonality to production and harvesting determined by low temperature, drought or excessive rain. In the highlands, plant growth is slow, and the production cycle is typically 18-24 months. Similarly, in the subtropics, farmers often leave the crop in the ground over winter, and harvest after the second growing season. Where rainfall is very low, growth may be so slow that reasonable production is only obtained after the second or third rainy season. In areas of seasonal flooding, harvest may be as early as 5-6 months, since cassava will not tolerate water-saturated soils.

Generally, the more traditional systems are more complex and rely on labour inputs rather than purchased inputs. System complexity has evolved out of the complementary interaction effects of individual components, to provide a balance between stability and productivity. In more modern systems, cultivators incorporate purchased inputs to achieve greater productivity and reasonable stability.

The labour-intensive nature of cassava husbandry is an area of concern as the labour force in agriculture declines. This is most notable in South America, and less so in the Caribbean. From 1970 to 1990, the average number of labourers per hectare of agricultural land in Brazil, Colombia and Paraguay, declined from 0.29 to 0.21. In Haiti and Cuba in the same period, the decline in absolute numbers was the same, but from 0.71 to 0.63 labourers. By comparison, in Thailand, Indonesia and India, the number of agricultural labourers remained stable in the same period, at about 1.25 labourers per hectare.

Products and Markets

About half the cassava produced in the region is used directly for human food, and half for animal feed or industry. This aggregate picture, however, masks the regional variations. In Brazil and Paraguay, the two largest producers, 50 and 65%, respectively, of production is destined for animal feed^[9]. Much of this is used on-farm in non-intensive systems for pigs and chickens. In nearly all other producing countries, the food market predominates, and only 10-20% of production goes for animal feed (Table 4).

The strong regionalization of forms of human consumption appears to be mainly cultural. Certainly there were opportunities for transfer of customs among regions, given the historical levels of communication and commerce. In Colombia, Cuba and Paraguay, the fresh market predominates, and elsewhere it is mainly processed to toasted flour (farinha) or baked into flat bread (casabe in the Caribbean).

On-farm feeding of fresh or dried cassava has a long tradition, but mainly in very non-intensive systems. With rapidly increasing demand for animal products - meat, milk and eggs - cassava is finding markets in balanced rations for animal feeds. The technical details for managing dried cassava in these rations are well established, both in terms of the milling and blending process, and on the animal nutrition side. The main constraints for continued expansion of this market are constancy of supply of raw material throughout the year, stability of quality, and price competitiveness.

Starch is not overall a major product from cassava in the region, but is important in local economies, especially in Colombia (northern Cauca Department), Brazil (South), and Paraguay, and its production is increasing. The two basic forms are native and modified. One of the popular forms of starch modification is fermentation, for a variety of bakery goods. Fermentation and sun drying combine to give cassava starch the capacity to trap air and expand. Baked products have a consistency similar to the gluten-containing wheat flour. The cassava/cheese breads are the most common products from sour (fermented) starch. Native starch is used in an array of food and industrial products: food processing, adhesives, paper and textile manufacturing, and others.

International trade of cassava or its products has been slight up to the present time (Table 4). This is related to a number of factors, including price competitiveness with grains or Asian cassava in international markets, lack of policy to promote exports, and a commercial mentality geared to traditional processing and markets. The region's main exporter is Costa Rica, where exports have risen steadily in recent years, to 100 000 MT in 1997 (FAOSTAT). Remarkably, this is more than all other Latin American countries combined. Nearly all the production goes through a well-organized system to export whole, paraffined or peeled/frozen roots to North America and Europe, to supply growing Latin populations. Other countries, especially Colombia, have taken note of Costa Rica's success in the export market and are now trying to enter with competitive products. Brazil has been a steady but relatively insignificant exporter of cassava flour at 500-1000 tonnes/year.

Not only has the region not developed export markets, but it has imported cassava, mainly from Thailand and Indonesia - up to 19 000 tonnes to Colombia in 1994, and 30 - 35 000 tonnes/year to Venezuela from 1991 to 1993. Even Brazil imported pellets. These imports went almost entirely to the market for balanced feed rations, to substitute for higher internal or imported grain prices or internal cassava chips. To enter international markets, Latin America would have to greatly improve production and processing efficiency in cassava to compete with the relatively lower costs in Asia. A more appropriate goal, and the one most countries are pursuing, is to partially substitute locally-grown cassava for imported grains in animal feed rations, in products for human consumption, and for industry.

Projections

The chief forces likely to set the trends for cassava supply and demand through the medium-range future are already in play to one degree or another. Some of these can be projected with relatively high confidence - the continuation of long-term, steady trends - while others are subject to less predictable variables.

The most thorough recent analysis and projections are those of the International Food Policy Research Institute (IFPRI) (Rosegrant and Gerpacio, 1997). FAO is also in the process of developing projections for cassava, but these are not completed at this writing. IFPRI projects cassava production and utilization in the year 2020 based on a model that takes into account virtually all the world's food production and consumption (International Model for Policy Analysis of Commodities and Trade (IMPACT).

Figure 4. Historical and projected population in rural and urban Latin America

Source: FAOSTAT

Cassava demand is projected taking into account urbanization and the dietary changes brought about by rising incomes. Growth in urban areas shows few signs of levelling off, although the ability of Latin American cities to continue to grow at historical rates could certainly be debated. Given that in aggregate, rural populations have begun a slight downward trend (Figure 4), there is a very real possibility of lower rural-urban migration in coming years, so long as adequate opportunities are available to rural residents. The strong growth in demand for meat will follow from these trends in urbanization and higher income. The study projects a 1.3% per year increase in use of cassava for animal feed in the Americas.

The study does not specifically address the impact of a growing starch market on demand for cassava. Henry and Gottret (1996) are able to find reasonably firm indications of growth potential in Asia, but only rather broad suggestions about the future development of the starch industry in Latin America. While highlighting the growing starch industry in Brazil, Colombia, Paraguay, Ecuador, Venezuela and Argentina, they point out that starch prices in the major producers (Brazil and Colombia), are quite high compared to world starch prices. Average prices in the period 1990-1994 in Colombia, for example, were more than twice those of Thailand ($523 vs. $233 per tonne). Later figures (July, 1997) show Brazilian wholesale starch (export quality) at US$360-400 per tonne, while Thailand maintained prices in the range of $180-$230^[10]. These differences are due mainly to the lower processing costs (efficiency of scale) in Thailand, and to a lesser degree, to lower costs of production. For Brazil to achieve a competitive position is certainly possible, even through application of already-existing technology, but will require a concerted effort.

Production is projected to grow at 0.8% per year to reach 40.5 million MT by 2020 (Table 5). Most of this increase will be from yield improvement rather than area increase. Area planted will increase only from 2.7 to 2.8 million hectares, while yield improves at a rate of 0.86% per year from 12.0 to 14.3 t/ha. This is a reversal of past trends, where production increases have been mainly by increase in area planted.

CONSTRAINTS AND OPPORTUNITIES FOR SYSTEM IMPROVEMENT^[11]

Impact of the External Environment

Policy

Food in Latin America has long been highly political, stemming in part from a history of skewed land tenure systems, and more recently from rapid urbanization, beginning with the post World War II period. Most Latin American governments implemented controls on prices of major urban staples, especially the grains - maize in Mexico and wheat and rice elsewhere. Domestic producers of these grains were supported by price controls, and deficits were made up by subsidized grain imports. Alternative carbohydrate sources like cassava were not included in these support policies, and as a result the price difference between cassava and grains became more pronounced. Consumers responded in a logical way, purchasing less cassava and more grain-derived products. This was most notable in the case of farinha in Brazil, where consumption halved, and wheat consumption doubled between 1960 and 1980 (Table 6). These policies also kept cassava at a disadvantage in the animal feed market, in spite of competitive production costs with local grains. However, this was more of a suppressed opportunity than of a shift in demand as was the case for the food markets.

Much of Latin America borrowed heavily in the 70s and early 80s to subsidize agriculture as well as support infrastructure development. By the mid-1980s, in the midst of a global economic slump, many countries were no longer able to meet their debt obligations. Widespread policy reform in the areas of trade and agriculture was implemented to bring economies back on track. Reduced subsidies and a move toward a more open international trade environment are completely reconfiguring Latin American economies. On the whole, the region is benefiting from these policies in improved economic growth. The longer term implications for equity and environmental impact are still a matter of considerable debate.

Demographics, Income, and Food Demand

Urbanization is a global phenomenon, but nowhere has it been as rapid as in Latin America and the Caribbean. From equal populations in rural and in urban areas in 1961, there has been a steady climb in urban populations with nearly constant levels in rural areas. Currently, three times as many people live in cities as in the countryside (Figure 4). This contrasts markedly with developing Asia, where twice as many people reside in rural areas compared to cities, with the gap narrowing relatively slowly. Populations in rural Latin America have actually begun to decline slightly. The stability of population in the countryside could be seen as positive were it the result of low birth rates, but instead it is largely from rural-urban migration. Governments seem powerless to control the flow. Lack of services in the slums surrounding cities, where most immigrants arrive, does not seem to be an impediment. The solutions obviously have to involve an array of approaches. A key one will be positive incentives to keep people in the rural areas and provide at least some of those opportunities they aspire to by migrating. These include a reasonable income, education for the children, and the basics for a comfortable lifestyle, such as medical care, clean water, and electricity.

Urban dwellers in Latin America are moving quickly toward foods that are more convenient to buy, store and prepare. Daily purchases of raw products at the open market are gradually changing to less frequent shopping for partially prepared foods at the supermarket. The traditional marketing system for fresh cassava does not fit in well with this trend. First, getting the product to the market is expensive because of bulkiness, and because of losses from physiological and microbial deterioration. Homemakers can only store cassava for a few days. Preparation, while it involves only modest inconvenience, may nevertheless be a constraint to cassava consumption. Lynam (1987) noted that cassava is more than five times more expensive in metropolitan areas than the opportunity cost to cassava producers. From 50% to 90% of the consumer price for fresh cassava in urban markets is allocated to marketing services - those costs involved in moving a perishable and bulky product from scattered small producers to convenient locations. The impact on urban consumption is clear, as noted earlier in Table 2.

Figure 5. Animal stocks in major cassava-producing countries of Latin America.

Source: FAOSTAT

On the other hand, cassava flour and cassava bread are both convenient to market and to use in an urban market setting. These products can easily make the transition from a rural to an urban staple, but tend to be less income elastic than competing energy sources.

Incomes in Latin America are rising steadily, although these benefits are not shared equally between urban and rural populations, nor among countries. In spite of the pressure of migrating poor to the cities, overall income is rising faster here, while nearly stagnant among the rural inhabitants. Rising incomes increase the demand for meat, fruits, and vegetables, and, in aggregate, decrease demand for starchy staples. Nonetheless, the very lowest income groups still increase their intake of staples with rising incomes. Cassava is unlikely to make major inroads into substitution for other traditional calorie sources in human diets - maize, wheat or rice. Cultures without a tradition of consuming cassava will not adopt it readily, except in cases of serious calorie deficits. Increased demand, beyond that of population increases, will be mainly for industrial use. There is nonetheless an important growth market in cassava-based convenience foods. The simplest of these is peeled, pieced, packaged and frozen cassava. In Colombia there has been a highly successful introduction of a pureed cassava/potato mixture. These croquettes are especially popular as a side dish in restaurant meals and are widely available in supermarkets.

Starchy staples like cassava may make the most significant contribution to development with a role as a component of animal feed. While cassava products for direct human consumption confront a constrained and possibly declining overall market (except for certain niche products). The animal feed market is growing and dynamic, especially for poultry (Figure 5). This does not automatically provide an entry for cassava into balanced rations - price competitiveness with alternative energy sources is the key.

Trends in Associated Commodities

Cassava in Latin America competes mainly with rice and wheat as a calorie source for human food, with maize and sorghum in the animal feed market, and with maize in the industrial starch market. The total production of these grains, in combination with soybeans, has nearly quadrupled since 1961 - from 44 to 163 million MT in 1997 (Figure 6). Of these, only sorghum has failed to realize steady, substantial increases in production over the past three decades. Like much of the world, Latin America in the past decade moved toward liberalized trade policies that bring internal prices closer to those on the world market. Apart from the Uruguay round of the GATT (now WTO), which provides guidelines for global reforms of tariffs and trade, parts of Latin America entered into regional trade agreements with wide implications for the region. The North American Free Trade Agreement (NAFTA) does not directly involve significant cassava-growing countries, but there is now discussion of extending this agreement south to include other countries. MERCOSUR is a free trade agreement between Brazil, Argentina, Uruguay, and Paraguay, with potential especially to affect cassava's competitiveness for feed and food in Brazil.

Figure 6. Crop production trends in Latin America

^a Most of the wheat produced in Mexico
Source: FAOSTAT

Well before the establishment of trade agreements, grain imports were increasing to meet demand that exceeded the steady increases in local production. The biggest surge in imports occurred in the 1980s and late 90s, with large increases in volumes of maize, rice, sorghum, soybeans and wheat (Figure 7).

Figure 7. Major grain imports to Latin America.

Source: FAOSTAT

In industrial markets, cost, quality, convenience of management, and stability of supply determine level of use of competing raw products. Throughout Latin America there has been a history of combined import restrictions, and subsidies to internal production as a means of promoting rural development. While these policies did succeed in keeping trade balances under control and providing producers a viable income, they were a disincentive to efficiency. Costs of production were high. Subsidies generally were provided for the major crops with export potential or the basic food grains in order to keep food prices low for urban centers.

As subsidies are withdrawn, the first response is increased grain imports. As farmers lose their markets to these imports, they begin to look for ways to improve efficiency and compete with local products at lower cost. This is a daunting challenge, given the long history of agricultural research in developed, exporting countries, and the resources available for efficient, high yield production. Much of Latin America is now in one of the most difficult stages of this transition from highly protected economies to free trade systems. Imported grains freely enter the market, but the national research establishments have not geared up adequately to address the issues of productivity, efficiency, and cost competitiveness.

At the same time this is also an opportunity for the cassava sector. While cassava has generally not employed intensive modern production practices, it also has never been part of a strongly subsidized system. It enters the new trade environment unencumbered by a history of dependence on government support. Farmers are fully aware that their profitability depends on their own skill and resources, and this has been an incentive to efficiency.

Infrastructure

Much of Latin America is faced with daunting obstacles to developing an extensive and efficient infrastructure for transportation, energy and communications. Formidable mountain ranges, vast jungles and savannas, and relatively low population densities mean that infrastructure is expensive to develop, on a per capita basis. In Central America and the Caribbean, countries are generally small and the outlays, as a proportion of small national budgets, can be prohibitive. The World Bank cites infrastructure development as a primary need of the region, and has financed extensive road-building and electricity-generating facilities. Agriculture suffers at both ends, especially in the poorer areas where cassava is more likely to be found. Inputs to production - machinery, fertilizer, chemicals - are expensive or impossible to obtain. Credit for small farmers to purchase inputs is relatively rare. Facilities for commodity and feed storage are inadequate to support efficient and timely transport and use in industries such as that of balanced rations. Getting products to market is also a constraint, especially bulky fresh roots. When these products need to compete against imported grains coming to port cities near major urban markets, cassava is at a serious disadvantage. Latin American economies in general seem to be making good progress, which should allow them to expand investment in infrastructure. In the longer term, it will be a decisive factor in the competitive ability of regional agriculture.

Institutional Resources

During the 1970s, Latin America in general committed to improved agriculture as a strategy for broad development. Many countries sent key scientists or even whole teams for advanced training, and strengthened their research system in expanded and improved facilities. The cassava sector benefited as well from this broad investment in agricultural research. Several countries that previously had no cassava program at all, or very minor efforts, developed national plans for cassava and established research teams to carry them out.

These national programs were complemented by establishment of the International Center for Tropical Agriculture (CIAT) in Colombia. The CIAT Cassava Program became a major institutional force for cassava research and training, and for acting as a convener to bring together national scientists in forums for international exchange and collaboration. The strong interdisciplinary orientation of this program became an operational model for many national programs in the following years. This surge in interest and investment was followed by an economic downturn by the mid-80s for much of the region. This was especially acute for countries that had borrowed heavily, were experiencing runaway inflation, and had difficulty making loan payments. Paring back on government expenditures often hit agriculture hardest, with its declining political power. And within agriculture, the cassava sector had among the least clout. The once-strong or moderate programs of Mexico, Panama, Dominican Republic, Ecuador, Peru and Venezuela were eventually phased out or reduced to very low levels of operation. Currently, the core of countries with strong institutions in cassava research and development is very limited - only Brazil and Cuba retain an interdisciplinary team in the context of a program with national responsibility for cassava research. Cassava programs are plagued by a high turn-over of scientists, although some programs have a very experienced staff. CIAT, in reorganization in late 1996, replaced the Cassava Program (along with the other commodity programs) with a project structure that gives less emphasis to commodity development, and higher priority to integrating commodities with resource management. Hence, this institution's ability to specifically support national cassava programs has been somewhat diluted. On the other hand, CIAT becomes more of a resource for integrating key components into broader agricultural development. During the mid-1980s to early 1990s, CIAT gave a high priority to promoting network development. Several semi-formal and informal networks were formed with missions and activities relevant to Latin America. In reality, most of these networks are a latent resource rather that actual functioning entities. Many depended heavily on CIAT for operational support, and have not been able to find other resources to continue activities.

The Cassava Research and Development Network, while never given a formal network structure, is a broad association of cassava scientists working across all disciplines and areas, linked by a regional newsletter published at CIAT, by attendance at various cassava-related meetings, by communication, by visits, and interchange of technology components.

The Cassava Breeding Network held its first meeting in Cali, Colombia in 1987, and reconvened for triennial meetings thereafter. Cassava does not lend itself well to the types of international variety testing programs that are often a main thrust of breeders' networks. Nonetheless, the interchange of information and germplasm fostered by the network contributed significantly to upgrading the quality and uniformity of genetic improvement activities in the region.

The Manihot Genetic Resources Network (MGRN) was established in 1992 under the auspices of the International Plant Genetic Resources Institute (IPGRI). Latin America, with its position as a center of origin for cassava, clearly should be taking a lead role in assuring the viability and productivity of MGRN. Like for the other networks funding and a diminishing core of cassava scientists are making this nearly impossible. The breeders' network and the genetic resources network have now informally merged, in view of their overlapping functions and interests.

The Cassava Biotechnology Network (CBN) functions globally, and includes active participation from several Latin American countries, especially Colombia, Brazil, Cuba, and Venezuela. This is the only network with strong involvement of developed-country, advanced research institutions. As CBN evolves toward a regionalized structure, the Latin American participants will intensify their contacts and interchange, possibly to collaborate on more region-specific issues.

A Southern Cone Network was established in the late 1980s to address some of the specific problems of subtropical environments, with participation by Paraguay, northern Argentina and southern Brazil. The activities of this network have since been absorbed by the more discipline-oriented networks.

A Post-harvest Network (global) brings together a large group of scientists, mainly from universities and private industry, which previously had little contact with each other. The interchange in meetings and informal communication has been a major contribution to setting the stage for the innovations and initiatives needed to bring expanded market-led benefits to the cassava sector.

In early 1999, several countries came together to launch consortium for cassava research and development, known by its Spanish acronym, CLAYUCA (Latin American and Caribbean Consortium for Cassava Research and Development). Founding members are Colombia, Venezuela, Ecuador, Cuba and Bolivia. This initiative was motivated by a number of factors. The decline in resources of most national programs left large gaps in research. Reorganization by CIAT eliminated the interdisciplinary cassava program, and although considerable research continues, it is less comprehensive than was previously possible. Several key players in the private sector saw the potential of linking their resources with those of the public institutions. Finally, these groups realized that there were many common goals across countries, and that working together to solve common problems would be a productive use of limited funds.

The Resource Base and Production Technology

By far most of the past research investment in cassava has been targeted toward improving production. The rationale is largely an outgrowth of the Green Revolution philosophy and experiences in rice and wheat - an attempt to supply growing markets (increasing populations) with reliable production. Experience with other crops also suggested a need to simultaneously improve both varietal and management components to make overall progress. Initially, the information and experience were too scant to attempt anything like a global quantification of the constraints affecting cassava. In the early 1970s many of the important pests and diseases of cassava had not even been described, and there was little knowledge about cassava's responses to agronomic inputs. Breeding had been rudimentary, with no comprehensive germplasm collections assembled. Over the next two decades, national and international research programs accumulated the basic information to continually make adjustments in their priorities.

The importance of post-harvest constraints did not become fully appreciated until much later. Because post-harvest issues become quickly entwined with broader social and economic issues, and because it is at the interface with initiatives of the private sector, many traditional public research institutions considered this to be outside their mandate. While CIAT had a post-harvest thrust from the outset, it was always a considerably smaller part of the research budget than production.

In 1993, CIAT initiated an exercise designed to quantify production and post-harvest constraints for cassava world-wide. The results are partially presented in Table 7, and these are the principal basis for the following discussion. This constraints analysis represents, in a sense, the accumulated experience of CIAT Cassava Program staff over 25 years of research, travel, and interaction with national program personnel, in addition to the direct input of those national program scientists into the survey. We consider it a highly useful basis for helping establish priorities for production research. It is important to note that the survey does not estimate total yield losses from individual constraints, nor of yield under unconstrained conditions. Rather, the survey asked for estimates of level of yield improvement that could be achieved by applying cost-effective techniques or inputs that farmers might reasonably adopt in the medium-term future. This assumes that most constraints will only be partially alleviated in this time frame, either because of technology that is only partially effective, or the best economic return may be from partial alleviation (such as an economic threshold for pest control). Estimates of yield potential with constraints completely removed would give quite different results, and would have very different interpretation. Continuing refinements to the survey will make the results more accurate and useful.

The 1993 CIAT exercise made a preliminary attempt to quantify post-harvest constraints, and Van Norel (1997) obtained more detailed information with national program visits and interviews in Asia and Latin America. The latter survey, while more detailed, does not include complete geographical coverage, and is somewhat difficult to place in a quantitative framework. Nonetheless, the responses allow for some tentative broad conclusions.

Improvements in practical soil and crop management have the potential to contribute to yield gains of over 50% in Latin America. Adding improvements in yield potential and pest and disease control would more than double current yields. The alleviation of combined pre- and post-harvest constraints could increase economic yield for the cassava sector by 133%, or the equivalent of about 40 million MT. This is intermediate between the 116% gains projected for Asia, and 168% gains for Africa. The differences between the continents are spread across a number of categories, but are most pronounced for the pest and disease constraints. Asia has little potential for increasing yields by controlling the already minimal losses to pests and diseases. Africa has the greatest potential because a relatively narrow range of constraints causes large losses.

Soil Management

Much of Latin America, especially where cassava is grown, suffers from moderate to severe soil constraints. Phosphorus and potassium deficiency, aluminium toxicity and low pH are widespread problems for agriculture. Cassava has a renowned tolerance to low pH and low soil fertility, and is cultivated mainly in these areas. On the better soils, higher value crops prevail. Some of the main exceptions are in sub-humid and semi-arid regions, where cassava holds a comparative advantage more for reasons of drought tolerance. Alleviating fertility constraints with practical and economical inputs could add about 17% to total production, or some 5 million MT of fresh roots.

About half the production area is subject to erosion constraints; the overall yield gains of 11% possible would be achieved by controlling soil and nutrient losses. Short term loss in crop productivity does not adequately highlight the need for concern about soil erosion in Latin America. On balance the weight of evidence points to a deteriorating situation from increasing population pressure on marginal lands.

Crop Management

Cassava's vegetative propagation is a considerable advantage in terms of maintaining varietal integrity, and in providing a large carbohydrate reserve for rapid early growth. On the other hand, propagation from stem pieces introduces some potential physical and biological constraints, often not fully appreciated by growers, since the effects may not be visible on the stakes themselves. Many factors influence quality at different steps, and their effects are often cumulative. Growing condition of the mother plants is the first determinant of quality.

Nutritional status, water stress, and presence of pests or pathogens, all affect quality of planting material derived from these plants. Stake treatment and storage conditions can alleviate or exacerbate these conditions. Finally, handling during preparation and planting can further impact quality. Given the broad array of pests and pathogens in the Americas, the biological component of stake quality tends to be more critical than in Asia, and probably similar to Africa. Yield gains of 13% should be achieved by practices to improve stake quality.

Hand weeding continues as the predominant form of weed control. Competition for labour by other farm operations means that control is often sub-optimal. To expect greater inputs in the form of hand weeding is probably unrealistic for most systems. Farmers already allocate their labour resources efficiently across different farm operations. Higher profitability of cassava cultivation with technology adoption or new market development will probably accelerate herbicide use and mechanized weed control, to add a potential 9% to overall production.

Intrinsic Varietal Traits

During the many centuries of cassava's domestication and evolution at the hands of New World farmers, certain fundamental traits became deeply ingrained in the plant's genetic make-up. These may or may not be the most appropriate for current and future needs of producers and consumers. Yield potential was only one of many traits farmers considered, and not necessarily the most important. In the most traditional of current systems, the criteria for varietal acceptance are virtually unchanged from those of centuries ago. But the increasing commercialization of the crop and application of production technology call for a new perspective on the plant's intrinsic traits, especially yield potential.

The CIAT survey identified low yield potential as the single most significant constraint in Latin America, with the practical possibility to increase yields by 19% over current levels. There are, however, some indications of tradeoffs between yield potential and yield stability for some traits. Strong canopy development requires a relatively high photosynthate partitioning to leaves and stems just for maintenance purposes, which takes away from the potential to increase starch accumulation in the roots. But a strong canopy serves several functions. These plants can compete better with weeds. If pests, diseases or physical damage reduce leaf area, there is a better chance that remaining leaves will continue to sustain the plant. A more robust plant produces more high-quality stems as planting material for the next crop cycle. Hence, creating a more efficient plant in terms of photosynthate partitioning to the economic product - the roots - means that these functions will need to be substituted by management inputs; this should be done at a lower cost than the value of the additional yield gain.

On the one hand, the opportunities to exploit a higher yield potential are very high, since new, efficient varieties currently occupy a relatively small proportion of total cassava area in the Americas. On the other hand, the accompanying management practices to optimize yield expression are also not widespread. Changes in varieties and changes in management will need to proceed in parallel.

Climate Constraints

Drought and low winter temperatures in the subtropics are the two principal climatic variables constraining production. Periodic or extended yield-limiting drought affects about 40% of cassava area in the Americas. At the country level, Brazil has by far the largest cassava area under rainfall constraints, including most of the Northeast, where nearly half of the country's cassava is grown. There are certainly practical limitations to alleviating this stress, with the assumption that irrigation is not economically viable except possibly in limited special situations. This leaves varietal adaptation, and practical management practices to conserve or otherwise manage soil water. The constraints survey suggested a possible 13% yield increase across the affected area, for a total yield advantage for the region of about 5%.

Low winter temperatures certainly constrain production significantly in the subtropics, but there are already varieties well adapted to these systems. Additional tolerance to low temperatures would provide about 13% yield gain (in addition to gains that can come from increasing intrinsic yield potential).

Pests and Diseases

The Americas contrast with Africa not so much in the total yield losses due to pests and diseases, but in the broader diversity of these problems. Asia has few biological constraints, and little yield loss from those that are present, with some isolated exceptions. While no single problem causes high losses across the entire region, the survey suggests that managing the several more important problems could add 28% to current yields. Root rots, bacterial blight and green cassava mite are all widespread and cause the highest yield losses. In general, the arthropod pests are more of a constraint in dry climates or during the drier months of the year, and fungal or bacterial diseases are more severe in wet climates or during the rainy season. The viruses as a group do not follow any general pattern of climatic adaptation.

Given the relatively low figures for potential yield improvements through alleviating constraints to individual pests and diseases, the relative level of research attention may seem surprising. But these numbers must be seen in a broader perspective. The biological constraints are dynamic entities, with potential to move, to mutate and to vary widely from year to year. Any one organism can become a severe constraint, and constant monitoring and preparation are required to reduce future risks. Further, Latin America is the laboratory for Asia and Africa. Most pest or disease problems on those continents originated in the Americas, and many times the solutions lie here as well - in resistance genes, biological control agents, or farmer knowledge and practices about control.

Post-harvest

The post-harvest constraints that are subject to alleviation through research and development inputs include root quality, processing systems, and marketing. In the Americas, the continuing predominance of traditional markets is a serious restriction on farmers' options. Especially, quality criteria must be strictly met in many regions. If those criteria are not met for a given variety, year, location, or harvest period, there may be no alternative markets.

Quality requirements are most stringent for the fresh market, a major sector of the market in Latin America. Typically, consumers have fine-tuned criteria based on a long tradition and familiarity with locally available varieties. These criteria include root form, external and flesh colour, ease of peeling, cooking time, texture, aroma, taste, and possibly others. Criteria are highly regionalized, and variations from these consumer expectations are not readily accepted.

Processed cassava for food has to meet somewhat less stringent quality criteria than for the fresh market. Criteria vary widely across regions and depend on end use. In Brazil's Northeast, for example, varieties for farinha normally must be bitter types, and with moderate starch content. In parts of the Amazon, yellow-fleshed varieties are preferred for farinha or bread-making.

Table 8. Examples of interventions to alleviate poverty and inequity, enhance food security and protect natural resources

Industrial uses generally require high starch content, but are less constrained in most other criteria. In any case, any organization attempting to introduce new varieties, processes or markets must first have a clear picture of any quality constraints that the market may impose.

Broad conclusions from the Latin American post-harvest constraints analysis show:

• Nearly equal gains can be made through R&D investment to decrease processing costs through managing root quality or improving processing technology.

• There is generally higher potential to add value to products, as compared to decreasing processing costs.

• There is a range of about 60-130% total additional revenues possible by improvements in the post-harvest sector - lowest for fresh roots, highest for animal feed, and intermediate for starch and flour.

• Rankings for quality traits by interviewees point out the strong importance of root perishability, with dry matter content also ranked high across regions and uses.

• Constraints to processing appear highly regional, with no predominant factor across products or regions. This is to be expected because of the variations inherent in processing different products. For starch, either water quality or the need for excessive water in processing is a broad concern.

• Price fluctuations were noted as the most important market constraint