WorDd Bank §Raf Working Papev No. 32e WccH' ITON. DC. C Prepared by T James Goering Agriculture and Rural Development Department Copyright ©O 1979 The World Bank 1818 H Street, N W Washington, D C 20433, U S A. The views and interpretations In this document are those of t EA and should not be attributed to the World Bank, to its aff iliat organizations, or to any individual acting in their behalf The views and interpretations in this document are those of the authors and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their behalf. WORLD BANK Staff Working Paper No. 324 April 1979 TROPICAL ROOT CROPS AND RURAL DEVELOPMENT Tropical root crops (TRCs) are important sources of calories for low-income groups in developing countries, and have considerable development potential because of their notable ability to convert solar energy to carbohydrates even in difficult agro-climatic conditions where sustained production of traditional food crops is not possible. This paper examines the potential role of these crops as sources of human food, livestock feed and as raw materials for industrial processing and concludes that the most promising immediate possibilities for expanded use are as food and feed. It suggests that the potentials of the TRCs are best realised through continued emphasis on production and utilization research, as well as on extension to transfer improved TRC production technology to farms. Suggestions are made regarding the design of projects which include the TRCs and the role which the World Bank might play in fostering expanded use of this group of crops. Prepared by: T. James Goering Agriculture and Rural Development Department Copyright (a 1979 The World Bank - 1818 H Street, N.W. Washington, D.C. 20433 U.S.A. Background papers for this study were prepared by Truman P. Phillips (University of Guelph, Ontario), David Dapice (Tufts University, Massachusetts) and Douglas M. Forno (East Asia and Pacific Projects Department). It also benefitted from review and comments by staff members of the World Bank, the FAO and agricultural research organizations. TROPICAL ROOT CROPS AND RURAL DEVELOPMENT Table of Contents Page No. I. SUMMARY AND CONCLUSIONS ..................................... 1 A Perspective .......................... ..................... 1 The Tropical Root Crops as Food ............................. 3 The Tropical Root Crops as Livestock Feed ................... 4 Tropical Root Crops and the Starch Market ................... 5 Other Industrial Processing of Tropical Root Crops .......... 6 The TRCs and Project Design ..... ............................ 7 The TRCs and the World Bank ... .............................. 10 II. TROPICAL ROOT CROPS: A PERSPECTIVE ......................... 13 Introduction ................................................. 13 Types of Tropical Root Crops ..... ........................... 15 The Aroids ....... ...................................... 15 Yam . .................................................... 16 Sweet Potato ....... .................................... 16 Potato ................................................. 17 Cassava ......... ....................................... 17 The Research and Development Effort ......................... 21 III. TROPICAL ROOT CROPS AS FOOD ..... ............................ 23 The TRCs and Nutritional Balance ............................ 28 The TRCs and Food Consumption Patterns: Some Case Studies .. 30 Kerala ....................................... ....... 30 Zambia ............................................................ 31 Northeastern Brazil ................. .. ............... 32 TRC Foods in More Highly-Processed Form ................. 32 Some Conclusions ... .. .. ..................................... 34 IV. TROPICAL ROOT CROPS AS LIVESTOCK FEED ...................... 36 Nutritional Considerations .................................. 36 The TRCs as Energy Sources in Compound Feedstuffs ....... .. 36 Imports ............................................ 36 Quality ............................................ 38 The Role of Price .................................. 39 Future Growth of Demand for Cassava in the EEC ...... ... 40 The TRCs as Feedstuffs in Developing Countries ............ .. 42 Some Conclusions ...... .. ... s ............ 43 V. TROPICAL ROOT CROPS AND THE STARCH MARKET ................... 46 Table of Contents - page 2 Pa n Nc,. VI. OTHER INDUSTRIAL PROCESSING OF TROPICAL ROOT CROPS 0000000000 49 Production of Leaf Protein 0.......0o.................0o-oo0 o 49 Starch 000000000000000000000000000o 0000000000000000000oo0o000 49 Fermentation Processes 0 0..00 ..00 . oo000 0.00o. . 0o0o00o00ooo0 51 Sugars .... .... .... ....o .....o ... ....oo -ooo .o .c.- 5R Butanol and Acetone 0... oo 0 ......... 0o.0.0 .. 0 0 0 0 00 00oo 52 Single-cell Protein 0000.0.0.0.00 000.0 0 0.0.000o0 0 00oo0oo 52 VIlo THE TROPICAL ROOT CROPS AND PROJECT DESIGN 000000000000000000 57 The TRCs as Food Crops in Small-farm Development 000o-00o0000 59 Central America 0........... 000 00 00 0 ...... 0 0o0 60 Nigeria 0000000000000000O -O -O -O -O-OO 60 Indonesia 0oo 0 ..00 . 0 0 ..000 .. 00000 00 00 00 00 0 .. 0000000 0 The TRCs and Protein Supplies 000000000000000000oo0o0o00 G TRC Production for Off-farm Sale 0000ooo00ooo0o0 0oo0o0o0 oo0oo 62 The Technology Factor 0... o 0.0 . o0 o-0 o-0o0oo0oo00 00 0 62 The Export Market 0.0000000o000000000o000oo0000000oo00 65 TEXT TABLES AND FIGURE Table 10 Production of Root Crops in Developing Countriesl, D1975 Table 2. Maximum Recorded Yields of Selected Staple Crops Table 3. Tropical Root Crops as a Source of Calories in Selected Countries, 1974 Table 4. Calories Supplied by Cereals and Tropical Root CropsD 1961-65, 1972, 1974 Table 5o Average Nutrients in 100 gm Dry iatter of Green Lezvso9 Beans1, and Fruits Table 60 Food Consumption in Kerala, 1970/71 Table 7. Nutritive Value of Different Cassava Root Products Copared with Sorghum and Maize Table 8. Projected Markets for Cassava Products in Selected EEC Projects in Selected EEC Countries Table 9. Imports of Starch into the United States and Japan, 1970C75 Table 100 Protein Production Directly from Crops and in SinILe-cell Protein Processes Figure 1 An Agro-industrial System for Cassava Table of Contents - page 3 ANNEXES Annex 1. Cassava Toxicity Annex 2. Documentation on Cassava Annex 3. A Note on Small-Farm Cropping Systems in Central America Annex 4. A Note on the Intercropping of Cassava in Northeast Thailand Annex 5. Table 1. World Production of Cassava, 1973, 1975 Table 2. World Trade in Cassava Products, 1970, 1973-75 Table 3. Food Composition of Staples Table 4. Amino Acid Content per 100 Grams Food Table 5. Essential Amino Acids of Some Root Crop Proteins Table 6. Average Monthly Prices for Thai Native Cassava Pellets Table 7. Projected Production of Compound Feed in Selected EEC Countries Table 8. Content of Structural Carbohydrates of Some Tropical Feeds and Forages Table 9. Capital Structure of Small-scale Cassava Starch Plants in Colombia TROPICAL ROOT CROPS AND RURAL DEVELOPMENT 1/ "A friend recently told us that he had discussed the winged bean with an influential Filipino family. 'They were incredulous that such a miraculous plant could exist,' he said. So on a hunch, I took them out back to the servants' quarters. There, climbing along a fence, was a winged bean plant laden with pods. 'But that's just sequidillas,' they said, disappointment echoing in their voices. 'It's only a poor man's crop.' "It is a universal phenomenon that certain plants are stigmatized by their humble associations. For information on a poor man's crop one has to turn, more often than not, to botanists and anthropologists; only they will have taken an interest in the plant. And yet the crop actually may be crucial to the lifestyles--even the survival of-- millions of people." --Dr. Noel D. Vietmayer, United States National Academy of Science I. SUMMARY AND CONCLUSIONS A Perspective 1.01 The tropical root crops (TRCs) are a widely-adaptable group of crops which provide staple food calories to about one-third of the population 1/ This paper is based on numerous research reports and special studies. Heaviest reliance has been on: (1) Proceedings of the Fourth Sympo- sium of the International Society for Tropical Root Crops, CIAT, Cali, Colombia, August 1-7, 1976 (edited by J. Cock, R. MacIntyre and M. Graham); (2) Cassava as Animal Feed, Proceedings of a Workshop held at the University of Guelph, April 18-20, 1977 (edited by B. Nestel and M. Graham); (3) UNCTAD/GATT, Cassava: Export Potential and Mar- ket Requirement (Geneva, 1977); (4) D. Dapice, "Tropical Root Crops in the Diet," Consultant Study Prepared for the World Bank (mimeograph, February 1978); (5) D. Forno, "The Role of the Root and Tuber Crops in the Developing World," World Bank (mimeograph, February 1978); (6) T. Phillips, Tropical Roots and Tubers: Future Markets, Supplies and Uses, Consultant Study prepared for the World Bank (draft), Uni- versity of Guelph, Ontario (February 1978); (7) J. Lynam, "Options for the Latin American Countries in the Development of Integrated Cassava Production Programs," CIAT (mimeograph, February 1978). One of the classic studies and still a useful reference is W.O. Jones' Manioc in - *-' Stanford Food Research Institute (1959). - 2 - in the tropical countries. 1/ Although grown on a wide range of farm sizes, their adaptability and remarkable efficiency in converting sunlight to carbo- hydrate energy makes them of particular interest to small-scale farmers and low-income consumers generally. Production estimates, probably less reliable than those for traditional grain crops, suggest annual TRC production in the range of 170 million tons, equivalent in calorie content to 50 million tons of grain. Occupying some 20 million ha of cropland globally, the TRCs generate an annual farm-gate value which exceeds that for developing country production of either cocoa, rubber, tea or coarse grains and approaches that of rice0 Cassava is the TRC of overwhelming importance, grown in more than 80 countries and accounting for nearly two-thirds of total production0 It is the major focus of this paper0 102 Despite the importance of this crop group, the TRCs have not until recently received much attention from the agricultural research community and development planners0 That deficiency is now being corrected through a sharp increase in research work on the TRCs at the international and national research centers0 At the international centers, the lead in this work has been taken by CIAT in Colombia and IITA in Nigeria0 The potato is the focus of research at the International Potato Center (CIP) in Peru0 National research centers in, e.g., West Africa and India have a long history of research on selected TRCs0 1.03 This paper is, to a significant extent, a review of Ohe "state of the art" of the research and development work on the TRCs, with a view to extending the use of this group of crops in the World Bank's agricultural and rural development lending program0 It is therefore a natural sequel to the Bank's Rural Development Policy Paper0 An underlying premise is that the TRCs, and cassava in particular, have a potential in meeting basic food and energy needs which may not be adequately recognized and is therefore less than fully exploited in rural development programs0 1.04 Most of the TRCs have features which take them uniquely suited for particular roles in rural development programs aimed at resource-poor groups of small-scale farmers0 The aroids or cocoyams, in production terms the least important of the root crops, perform best on good soils with a high water table but, unlike other TRCs, have a marked tolerance to highly saline soils subject to periodic flooding0 The principal edible yams produce large energy yields per ha but require generous moisture and soils of good drainage0 Sweet potatoes are grown on a wide variety of fertile soils for both their roots and leafy tops, which are valuable as livestock feed0 Potatoes are tho leading starchy root crop of the temperate countries and annual production is twice that of all other edible root crops combined0 But because of their limited climatic adaptability, less than 10% of production occurs in develop- ing countries0 Work at the International Potato Center in Peru promises to provide varieties of this particularly nutritious root crop which perform well in wider soil and climatic ranges0 1/ The TRCs included in this paper are: cassava, potatoes, sweet potatoes, yams, and aroids or cocoyams0 - 3 - 1.05 For a variety of reasons, most of the recent upsurge of interest in the TRCs has focused on cassava. It is a major source of calories for some 300 million people, many of whom are rural and poor. It is one of the world's most efficient plant converters of solar energy to carbohydrates, with an ability to store energy at a daily rate per ha which substantially exceeds that of most other traditional high-yielding crops. It is relatively resistant to insects and plant diseases and requires few purchased inputs in traditional production systems. It is adaptable to a wide range of agro- climatic conditions and performs surprisingly well on acidic soils of limited fertility--of which an estimated 1,750 million ha exist in developing coun- tries. This adaptability and cassava's production locus among smallholders makes it well suited for a key role in poverty-oriented rural development strategies. The Tropical Root Crops as Food 1.06 Because of the TRCs' important role as subsistence crops among the rural poor, consumption data on this group of crops tends to be weak. But what cannot be disputed is that the TRCs are an important sources of calories in many of the poorest countries. FAQ estimates for 1974 suggest that the TRCs provide upwards of 20% of total calorie intake for 11 countries and nearly 40% or more of all calories in three (Zaire, Ghana, Togo). Moreover, the TRCs are basic calorie sources in poor regions of other large countries such as Indonesia and Brazil. Evidence suggests that large numbers of low-income groups consume more TRCs as incomes rise, although this food group appears to be approaching an "inferior good" status among middle- and high-income groups. The major use of the TRCs will continue to be as food for low-income consumers, with the most rapid growth in food demand likely to be for the processed less- perishable TRC flours, meals and starches. 1.07 As efficient producers of calories, the TRCs have an important potential in meeting future food needs, particularly of calories but also of other nutrients. When compared with the grains in terms of nutrient content/kg, the TRCs generally rank poorly in terms of protein, carbohydrates, iron, thiamine and iron, but are good sources of calcium and, in the case of sweet potatoes, an exceptional source of Vitamin A. Among the TRCs fresh cassava roots compare favorably with the other root crops in terms of energy content, but rank at the bottom in terms of protein. Cassava roots are gen- erally rich in calcium and ascorbic acid and contain nutritionally-significant quantities of thiamine, riboflavin and niacin. Protein quality of cassava roots is inferior to most alternative carbohydrate sources although the leaves are rich in high quality protein and are consumed by humans in several coun- tries. Among the root crops the potato is distinguished by its high protein content -- roughly equivalent to rice on a dry weight basis and with a protein quality approaching that of beef. It also contains high levels of Vitamins B and C. With high yields and short maturation periods, the potato outranks all major world food crops in protein production per unit of time. 1.08 Where traditional grain crops grow well and good land is readily available, the potential of the TRCs as food crops will remain limited. But in less favored areas the TRCs plausibly have greater potential to alleviate nutritional deficiencies than is commonly believed. Nutritionists have frequently observed that food consumption by most adults (as opposed to children and pregnant or lactating women) in many developing countries tends, on the average, to be roughly balanced in nutritional terms (i.OeO. if calorie intake is adequate, so are most other nutrients). This is not always true across regions or socioeconomic groups, nor is it rarely true among vulnerable groups. But case studies in three countries/areas where TRCs are important in the diet (Kerala, Northwest Zambia, Northeast Brazil) suggest results which are generally consistent with this observation. In these circumstances of balanced, but calorie-deficient, consumption patterns, the most efficient path to nutritional adequacy seems to be to increase the quantity, and promote the effective use, of indigenous foods. Small--farm cropping systems which include the TRCs and groundnuts or beans would be valuable parts of a food and nutrition program in these c'rcumstances0 Careful examination of typical diets for particular age and socioeconomic groups or geographic regions is required before unequivocal recommendations can be made regarding nutrition- ally-adequate consumption patterns and supporting food production systems0 1l09 The special nutritional problems of young children and pregnant and lactating women may be widespread and, in many cases, serious0 But frequently these problems are soluble if nutrition education is combined with increased availability of local foods0 Even in these circumstances, expanded use of the TRCs appear to be warranted, if combined with other foods rich in protein and other nutrients0 The Tropical Root Crops as Livestock Feed 1010 The great attraction of the TRCs as animal feedstuffs lies in their ability to accumulate carbohydrate energy at a rapid rate0 The export of cassava peLlets for livestock feed in West Europe is the most dynamic element in the expanding global utilization of TRC products0 Total shipments in 1978 may approach 5 million tons, reflecting an annual growth rate in recent years of nearly 20%. Exports are dominated by Thailand and, to a much smaller extent, Indonesia0 Thai cassava exports, benefitting from the availability of extensive lands well-suited for cassava production, and government assistance largely through the provision of good physical infrastructure, are now the country's second most important export commodity0 1011 Quality and price competitiveness with alternative energy sources critical determinants of the size of TRC exports to the EEC0 Quality stan- dards exist, although they are frequently ignored by producers and exporters. Price competitiveness is directly dependent on the EEC's Common Agricultural Policy for cereals, under which imports of cassava pellets are subject to a variable levy which keeps the price of pellets low in relation to cereals0 The extent of cassava use in compound feed manufacture depends in part on the type of ration being formulated, as well as price relationships between cassava, other cereal substitutes and the major protein source0 Typically, cassava prices to the compounder must be at least 20% less than cereal and soybean meal prices to ensure maximum substitution0 - 5 - 1.12 The export of cassava pellets to the EEC is likely to continue to expand, although at a growth rate lower than that of the recent past. Among the important political factors which will determine the size of the export market for cassava products is the growing pressure by some EEC members to grant increased protection to community grain producers. Assuming consumption trends of the past and continued price competitiveness, the market for cassava products in the major EEC countries cQuld expand from the present level of about 5 million tons to 7.6 million in 1990. Growing livestock industries in other developed countries such as the USSR and Japan might provide other market outlets for cassava chips or pellets. 1.13 The use of TRCs as feedstuffs in developing countries is expanding, albeit less dramatically than in EEC compound feeds, and, for both technical and economic reasons, appears to offer one of the greatest unexploited poten- tials for increased utilization. The carbohydrates in cassava roots, for example, are highly digestible and provide a valuable complementary energy source to the typical tropical forages which tend to be fibrous, bulky and of limited palatability. Supplementation of tropical forages with cassava products therefore generally has a marked positive impact on livestock performance. Recent work in Venezuela has demonstrated that high yields of good quality protein are obtainable from cassava leaves at reasonable cost for use in livestock feeding. Cassava leaves have been utilized for commercial export as livestock feeds in Thailand. 1.14 In general, cassava products can be successfully substituted for alternative energy sources for different species of livestock in both tropical and temperate countries, although research work is not sufficiently advanced to permit precise specification of rations under particular circumstances. More extensive use of cassava is possible in broiler rations than in laying rations. Life-cycle feeding of cassava in swine rations is feasible, although supplementation with key amino acids, particularly methionine, appears to be essential. The use of cassava as a substrate in the manufacture of single- cell protein is a good possibility in animal feeding. Prospects for using both cassava roots and leaves in ruminant livestock (cattle, sheep, goats, horses) appears to be promising, particularly in combination with urea. Continued research on both production systems for roots and leaves and animal nutrition in TRC feeding programs, now underway at CIAT and national centers, is required to bring several of these possibilities to commercial practice. Tropical Root Crops and the Starch Market 1.15 In terms of physical and chemical properties, TRC starches differ only slightly from those derived from grains. Substitution across a wide spectrum of end uses is possible, implying that quality and price competi- tiveness are basic considerations in determining market shares. Starch imports in the two large markets, the United States and Japan, have been in the range of 100,000 tons annually in recent years, with cassava starch accounting for three-quarters of this amount. The value of total starch imports into these countries of US$30 million in 1975 can be compared with cassava chips exports to the EEC of US$275 million. - 6 - 116 Cassava starch imports into the United States have declined in recent years because of higher freight costs, the erratic nature of import supplies and the increased availability of specialized corn starch modifed to meet particular end uses. This decline seems likely to continue. Prospects of TRC imports into Japan may be somewhat brighter, because of proximity to the major source of supply (Thailand), although the policy of protecting the local Japanese starch industry is likely to limit that grouth. Developing country markets for TRC starch appear to be more promising, particularly in those countries with large textile industries0 The development of modified TRC starches for special uses also might expand the market, although it is not clear that this warrants a high priority among the TRC starch-producing countries. Thailand accounts for about half of the world's cassava/sago starch exports and has large excess production capacity. Because the price of cassava roots is normally governed by the pellet industry in that country, only limited possibilities exist to reduce starch prices to make exports more competitive. l117 Despite this somewhat somber export picture for the TRC starch, local starch industries which supply domestic needs are significant market outlets for TRCs which, in many circumstances, could be expanded. The range of possible production technologies is wide and some economies of scale in starch manufacture seem to operate. The installation oF large-scale starch factories using, e.g., maize as the raw material, appears to make it diffi- cult for smaller TRC-based plants to compete successfully. Where small-farm production and the creation of rural non-farm employment are important develop- ment objectives, deliberate government policy regarding investment strategies and research on processing technologies is likely to be required to improve the quality and efficiency, and thus the viability, of the small-scale TRC starch plant in rural areas0 Other Industrial Processing of Tropical Root Crops 1l18 The efficiency of the TRCs in converting sunlight to carbohydrates and the agronomic potential to greatly increase nutrient yields has led to expanded research on industrial processing of these crops0 A leaf protein concentrate can be extracted from cassava tops using relatively simple "village-scale" technology. Major fabricators of centrifugal process starch manufactuiring' equipment suggest it may be possible to design a medium-scale (throughput of 3-5 tons of fresh root/hr) starch plant which is technically efficient and capable of producing the high-quality product required if TRC starches are to retain their p iotionL in the market0 119 Among the most interesting technological developments in TRC utili- zation are fermentation processes for the manufactuire of iigh-value products such as sugars, ethyl alcohol and single-cell protein (SCP)0 Research on these processes has expanded in recent years in response to sharply higher relative prices for petroleum--the usual feedstock in the manufacture of these products0 However, the economic viability of some of these processes remain to be demonstrated0 Production of glucose sugar from cassava, or other TRC, starch is as feasible technically as from the more common practice of hydrolyzing corn or wheat starch. The major consideration is the relative price of starch from these various sources. A doubling or tripling of cassava yields, as now seems possible, would quickly make the use of cassava starch economically attractive. 1.20 An important question to food planners regarding the TRCs is their potential role in meeting future protein requirements of humans. Production of SCP for animal feed now exceeds well over 1 million tons annually, gener- ally using certain petroleum fractions as the feedstock. The technology is equally suited for organic materials such as the TRCs, provided costs are sufficiently low and raw material supplies are adequate. Typically, raw material costs make up at least 60% of total production costs and are critical to determining economic viability of the process. And because plant costs per unit of output decline sharply as plant size increases, a basic question, in view of the objectives of this paper, is whether efficient, smaller-scale plants are technically and economically feasible. 1.21 Research at CIAT, with support from Canada's International Devel- opment Research Center, is underway to provide these answers. A pilot plant to produce SCP from cassava began operation at CIAT in 1977. In early experimentation, the crude protein content of the dried product was in the 30% range. When fed to laboratory animals, and supplemented with methionine, weight gains exceeded those rrom the feeding of soybean meal protein. An important remaining uncertainty is the possible health risk to individuals from continued exposure to spores of the micro-organism employed in the process. Regular medical exams of research personnel have not revealed any health problem to date. 1.22 Several of the TRCs, and cassava in particular, have been used at various times and places over more than 50 years in the production of ethyl alcohol. The product may be blended with gasoline up to 20% by volume with- out altering the carburetion systems of most engines. Much of the research on the process in recent years has been underwritten by the Brazilian Govern- ment in support of its National Alcohol Program which aims to replace 20% of total motor vehicle fuel consumption in 1990 with ethyl alcohol. Although much of the installed production capacity in Brazil's alcohol industry is based on sugar cane as the feedstock, use of cassava is growing because of its wider adaptability to soil and moisture conditions. In late 1977 PETROBRAS of Brazil started up the world's first large-scale plant to produce alcohol from cassava grown in nearby plantations. Although some 30,000 liters of alcohol were produced to specification during its first test run, problems with the regularity of adequate root supplies have emerged. Also problematic is the relationship between root prices and the government-established price of alcohol. As in all industrialization processes involving the TRCs, high crop (and energy) yields are required to ensure economic viability. The TRCs and Project Design 1.23 The greatest immediate opportunities to expand the use of the TRCs are as sources of human food and livestock feed in agriculture and rural - development projects. Extensive use of the TRCs as industrial raw materials requires lower unit costs of the energy they provide and/or more efficient utilization technologyo An appropriate strategy for rural development must reflect several factors: the need for low unit costs of development programs, the relative abundance of labor supplies among the rural poor, their limited access to good land and their virtual absence of cash reserves. This strategy should also reflect persistent calorie shortages in their diets and the fact that many of the future poor must seek l1velihoods on lands of limited natural rt!rtiltty in a development environment with weak supporting services for agriculture. Improvement in the welfare of this group then requires the introduction of relatively simple low-risk technologies which permit greater returns from the limited resources under their control. The TRCs, by virtue of their importance to smallholders throughout the world, are well-suited to provide benefits related to both agricultural growth and increased income *,pportunities for the poor. But if these benefits are to be fully realized Lhere must be deliberate effort by the governments and donor agencies such as the Bank to influence several factors--ranging from the type of cropping svstem in rural development programs to the nature of TRC production and utilization research and the scale and type of technology in TRC processing0 1.24 Cropping systems which include TRCs and provide adequate income and -i production "mix" to ensure sound nutrition for the farm family are not always easy to design0 Many small-scale farmers, having experienced the vagaries of markets for their cash crops, frequently prefer the security of producing most of their own food even if it means sacrifice of some income0 :irst priority in the design of a cropping system for low-income families may well be to provide a variety of crops to ensure food security, then allocate t.he balance of the holding to cash crops0 The portion of the typical small- holding allocated to basic food production is likely to be small if the TRCs Lrc- included in the cropping program0 With, e.g., cassava yields of 10-20 tons/ha/yr, along with maize yields of 1-2 tons/ha, approximately one-half ha ditvided between these two crops, intercropped with beans, would provide the f,.o.ily's basic calorie and protein needs0 The low-risk nature of cassava production makes this crop an ideal "safety net" to ensure some food supplies in risky project situations0 1.25 To meet family protein needs the use of beans or other protein-rich vegetables, intercropped with TRCs, is possible, although with some reduction in yields. More precise cropping possibilities can be determined only in a location specific context0 CIAT experience with cassava/beans intercropping suggests that to provide each of five adult equivalents with 100 grams of beans/day (over 20 grams of protein) would require only 180 kg of beans per year, or, at 1 ton/ha, 0.18 ha of beans without any reduction in cassava yields. In more favorable areas where root crops with more protein (e.g., potatoes) can be grown, the protein problem becomes less difficult0 The protein content of 1 kg of potatoes and 100 gm of beans is generally adequate for most adults, even i.f cassava supplies the balance of the diet0 Simple fermentation processes for the TRCs, utilization of the leaves in human food or the use of the TRCs in livestock production and fish culture are other ways of augmenting protein supplies in small-farm development programs0 - 9 - 1.26 Because direct consumption of the TRCs by producers is likely to expand only slightly more rapidly than their increase in numbers, improved technologies are needed to increase other forms of utilization. This requires technologies to utilize the TRCs more widely for urban consumption as well as technologies to ensure efficient industrial processing of these crops. Marketing costs, now relatively high for the TRCs vis-a-vis other food groups, can be reduced through improved market infrastructure and transport. Of particular importance for the root crops are improved storage technologies to slow post-harvest deterioration of the fresh product. Further research on the incorporation of TRC meals/flours in composite or fortified flour is a neces- sary part of this overall effort to reduce the cost of calories and increase food nutrients to low-income consumers. These activities can be influenced through careful design of research components in rural development, nutrition or agricultural research projects. To a significant degree, the potential of the TRCs in rural development programs is dependent on the nature of support- ing research and the extent to which results are incorporated in projects. 1.27 Basic choice-of-technology problems also exist for TRCs in non-food ses. At one extreme are smallholder production systems supplying fresh roots to small-scale chipping and drying facilities while at the other end of the technological spectrum are plantation production systems producing fresh roots for large-scale, capital-intensive facilities to produce in- dustrial starch or alcohol. The choice of technology for processing is perhaps the most critical in determining the distribution of benefits from TRC-oriented development programs, since the type of processing technology chosen is an important determinant of production systems for the raw material. Large-scale processing leads to pressures for large-scale production systems, with reduced opportunities for smallholder participation. Small-scale produc- tion and processing technologies in TRC-oriented activities are possible, as evidenced by the cassava/chip industry in Thailand and the cassava flour/starch industry in Northeast Brazil. Technology choices which contribute to broadly- based growth and income distribution objectives clearly are possible, but this requires a clear delineation of technological alternatives and correct specification of the real economic costs of the resources involved. 1.28 The export market for TRC products such as cassava will be deter- mined largely by the extent to which price and quality competitiveness of the TRCs, as compared with other sources of feed energy, can be assured. The need for regular, dependable supplies of pellets, as well as ship cargoes large enough to realize the important economies of scale which exist in shipping and handling, implies that individual export programs should probably be based on annual root production of at least 250,000 tons, equivalent, at 10 tons/ha, to 25,000 ha of crop. This would require about two pelletizing factories, each with annual pellet capacity of 60,000 tons. Total establishment costs for a factory of this size using imported machinery are in the US$500-750,000 range (1977 prices), although locally-fabricated machinery is available and would result in considerable cost savings. Based on recent cereal prices and shpping costs of cassava chips, feasibility planning for cassava exports could prudently use an FOB price for pellets in the range of US$90-100/ton. In a number of countries where the TRCs are staple food items, this would require higher production and lower prices to compete with cassava supplies from traditional exporters. - 10 - The TRCs and the World Bank 1.29 The World Bank, as a development institution concerned with food and energy problems and increasingly involved in the production of these products, has a good opportunity to take an active role in promoting further utilization of the TRCs0 Among the prerequisites for the wider use of the TRCs in meeting future global food and energy needs is further research into the production and utilization of these crops, together with efforts to ensure that research results are speedily incorporated in development projects. 1.30 The most basic requirement for realizing the development potential of the TRCs -- either as human food, livestock feed or industrial raw material -- is to increase crop yields and thereby reduce unit costs of the energy produced, while at the same time increasing the returns to producers. A continued strong focus on production research is therefore essential. Work on this is underway at the international agricultural research centers and at several national research centers0 This work must continually be evaluated as to the extent to which it contributes to each of the potential end uses of TRCs. Bank project staff should be familiar with this research and the possibilities for practical application. Research on small-scale farming systems which incorporate the TRCs could usefully be exyanded aet several locations to focus more directly on unique agro-climatic conffitions. As the work at these centers expands, an emerging problem is the frequent inability of the national extension service to convey research results to farmers in a convincing manner. Already the generation of improved technical practices in TRC production exceeds the ability of many national extension services to disseminate results. The implications for World Bank operations, while not new, are nonetheless important: active support of extension programs as well as production research on the TRCs are vital to the realization of the TRC potential0 1o31 To date, TRCs have been included in only a small number of Bank- supported projects, possibly a reflection of the somewhat negative image which this group of crops has in the minds of some agriculturalists, or because of lack of familiarity with the potential of the TRCs resulting from recent research work0 Closer interaction with scientists at national and international research centers and more active programs of information interchange within the Bank (seminars, information circulars) would help to achieve a wider familiarity on the part of Bank staff0 Experience and information from small-farm cropping systems which include the TRCs are not always readily available to project staff within the Bank0 Thus there is a need to draw on, and facilitate the exchange of, experience from Bank and non-Bank projects, and for the inclusion of pilot components in Bank projects from which experience and information can be obtained0 1o32 Because the TRcs are generally a low-cost source of calories, they merit wider consideration in the Bank's rural development and nutrition projects0 Innovative work in on-going nutrition projects in Colombia and Brazil (composite and fortified flours using cassava; project-funded research and development work on village level processing technology) is providing - 11 - experience which may be applied elsewhere in the world. It is important that this experience be utilized. It is also important that the "nutritional dimension" of small-farm cropping systems in Bank-supported programs be examined as carefully as agronomic and economic considerations during design, preparation and appraisal. 1.33 The low demand elasticities for cassava products consumed directly by humans implies that even modest increases in supplies forthcoming from a production program could push farm prices sharply lower and destroy producer incentive. Expansion of this widely adaptable source of income and employment thus requires the stimulation of markets for TRC products other than fresh roots. The production and consumption of fermented cassava, flour, and other prepared products such as gari and fufu in West Africa and farinha de mandioca in Brazil are widespread in the TRC-production zones and are important rural activities. But basic products such as gari are bulky and perishable and require large amounts of labor in their preparation. Efficient production of these products will require wider use of improved technologies now being used in a few localities. The Bank, through its rural development and nutrition lending programs, could assist in the further development and expanded use of these technologies. 1.34 The Bank could influence technology choice through its industrial lending program. Large-scale equipment to produce TRC starch for industrial use is available, but this is frequently linked to large-scale production systems, thereby limiting the role for the small-scale producer. This approach may be appropriate in some circumstances with growing markets for the output and abundant agricultural land. But alternative systems exist. Medium-scale gari-manufacturing facilities are being brought into use to serve urban areas in West Africa. Research and development work on smaller-scale, village-level processing, particularly for cassava products for human use, is less advanced. The Bank should encourage this type of development through its support of agricultural research, small-scale industry and nutrition lending. It also should play an active role in assisting in the disseminating of research information and experience across projects and countries. 1.35 There has been relatively little experience with non-food uses of the TRCs in Bank-supported projects, yet these uses are among the most promising new markets for this group of crops. One of the most significant is as livestock feed in middle-income developing countries where demand for meat is growing rapidly. Processing and utilization technology is available, although narrowly concentrated in a few producing countries and the EEC. Production and processing technology appear to be well suited to either small or large scale agriculture, although the extent to which production is to be mechanized must be determined on a case-by-case basis. The Bank could use- fully support various projects of different scale and design to produce TRC products for livestock feeds. These could range from individual small farm systems, integrated with village-level drying and chipping/pelleting facilities, to more centralized plantation systems, with or without out- growers. Pilot projects of this nature might be located in underutilized lands of e.g. the Amazon Basin or Indonesia's Outer Islands. Contractual - 12 - arrangements with local feed compounders might be a part of the marketing operation. 136 The use of TRC products as inputs in more sophisticated industrial processing will develop more slowly as costs of TRC energy decline with increased yields and as more efficient processing technology becomes avail- able. But because results to date are promising and because the eventual benefits are potentially large to petroleum-short countries, the Bank should play a role in capitalizing on these possibilities. In the area of utiliza- tion research much of the work is being done by private or quasi-governmental agencies. The Bank should assure that its operations benefit from these technologies as they become available by participting, where possible, in these research activities through joint agreements which may well require partial funding of the work by the Bank0 1.37 Technical and economic uncertainties remain with regard to the use Of TRCs in the production of single cell protein and alcohol. Possibilities for joint funding of utilization research in these areas might also be explored by the Bank. Equally important perhaps is the support of pilot projects based on technology which is now emerging from on-going research efforts. A good example, with large potential benefits to both producers and consumers, is the single-cell protein research program at CIAT. If and when commercial viability of this effort is demonstrated, funding for integrated production and processing schemes could be utilized. An activity of this nature could be a valuable component in an integrated rural development project. Nutri- tional benefits are potentially very large. 1.38 One of the most interesting possibilities for the TRCs is their utilization in the manufacture of ethyl alcohol. The economic viability of this process is likely to be determined largely by the extent to which unit costs of TRC energy can be reduced and whether a breakthrough in more effi- cient fermentation technology can be achieved. Progress is being made in both areas, This process, capable of using very large volumes of TRCs, will almost certainly become more attractive as relative prices of pe.roleum products increase, The optimum scale of operations remains to be determined, although this is likely to be fairly large, requiring a well-organized pro- duction system to provide raw roots to the factory. Commercial pressures are likely to result in correspondingly large-scale production systems in which small farmers play little, if any, role. As an alternative, the Bank could explore possible support for a production system in which smaller farmers provide some of the raw material, possibly through a carefully-manage6 outgrower system. Bank experience with such systems for other crops would be useful in designing such a project. 1.39 In summary: the TRCs, as efficient converters of solar energy to carbohydrates, have a potentially important role in meeting food and energy needs, particularly for low-income groups, That role is likely to become greater as fossil fuel costs rise, and as population pressures push the agricultural frontier to lands of increasingly limited potential in terms of their ability to sustain production of traditional food crops. This potential - 13 - can be unlocked through continued production research aimed at increasing TRC yields, coupled with expanded efforts in utilization research and measures to bring research results into practice through development projects. In each of these activities the World Bank can make a valuable contribution which is central to its interest in food and energy issues. The Bank should continue its support of production research and strengthen its support of TRC utiliza- tion research both at the international and fational levels. It should endeavour to increase familiarity with the TRCs among Bank staff and expand efforts within the Bank to disseminate information and project experience related to this group of crops. The Bank should promote agricultural produc- tion and pricing policies which encourage the efficient production of these crops. It should support measures to strengthen extension services in TRC- producing countries and thereby facilitate the adoption of improved production practices which are already available. And, finally, it should take bold action in support of activities and projects related to the TRCs which may involve technologies which, while not fully proven, hold the promise of gener- ating substantial benefits, particularly for low-income groups in developing countries, but also for the world at large. II. TROPICAL ROOT CROPS: A PERSPECTIVE Introduction 2.01 The tropical root crops in recent years have been "rediscovered" by the agricultural research community and others concerned with the food and nutrition problems of low-income people. Some researchers, of course, have recognized the possibilities of these crops in the past. 1/ Nevertheless, such voices have been in the minority, being generally dominated by those who conceive of the world's food situation as essentially relating to the supply of traditional grains. Such a view is only partially correct and poses the risk of overlooking the potential of a food group which is of great importance to many low-income families. Cash cropping possibilities of the root crops are no less important in some circumstances. 2.02 The paper is heavily descriptive in character, reflecting the view that research work in the last decade on the tropical root crops has opened several new possibilities for development which are not well known. The first major section describes briefly the technical characteristics of the major tropical root crops and indicates the role they play in the world food/feed complex. The next two chapters examine the TRCs as sources of food for humans and feed for livestock. These are followed by a brief chapter on the use of TRCs in the manufacture of industrial starch. The chapter following examines some of the agro-industrial possibilities for the TRCs while a 1/ Messrs. Coursey and Haynes, for example, writing in 1970, suggested that ". . . in the context of providing food for the exploding populations of the tropics, [the root crops] may have greater potential than is commonly appreciated." D. G. Coursey and P. H. Haynes, "Root Crops and Their Potential as Food in the Tropics," World Crops, July/August, 1970, pp. 261-265. 14 - concluding section assesses the role which the TRCs might play in the Bank's agriculture/rural development lending program0 2.03 Historically, governments and research workers have paid relatively little attention to the root crops. The TRCs have frequently been re- garded as inferior food, produced and consumed by a limited number of subsis- tence farmers. They have been thought to occupy a relatively small portion of land in developing countries, and, as in the case of cassava , have been viewed with suspicion by agriculturalists concerned with longer-term effects of this crop on soil productivity. And, until recently, the TRCs have played a very small role in international tradeo 2.04 Several factors account for the growing interest in the TRCso These include a growing appreciation of the large number of poor people who depend on these crops for basic calories0 Global population growth and the sharp relative increase in prices of fossil-based energy have contributed to the interest in the TRCs as food and energy sources0 Recognition of the TRCs' adaptability over a wide geographical area has been a factor0 An important consideration in an "energy conscious" environment is the notable ability of the TRCs to produce and store large quantities of energy per unit area, frequently from marginal lands with highly variable rainfall0 Improved industrial processes to permit efficient transformation of this energy to other agro-industrial products have provided additional possibilities for the TRCs and further heightened interest in them0 2.05 Total production of TRCs in the developing countries is of the order of 170 million tons annually, roughly equivalent in calorie content to 50 million tons of grain0 1/ About a third of the population of the tropical world relies on TRCs for staple food calories and within this population the TRCs are of greatest importance among the lowest-income groups0 The aggregate farm-gate value of TRCs generally exceeds that of developing country produc- tion of either cocoa, rubber, tea, or coarse grains and approaches that of rice. 2/ In the developing countries the TRCs are dominated by cassava (accounting for more than 60% of total TRC production), followed by potatoes, yams and sweet potato, each making up something more than 10% of the total0 Tropical root crops cover more than 20 m ha globally, equivalent to about 7% of the area under foodgrains. 1/ Assuming 100 calories per 100 gm of fresh roots and 350 calories per 100 gm of cereals0 2/ IDRCo Cassava: The Development of an International Research Network (Ottawa, 1976) , po 13. - 15 - Table 1: PRODUCTION OF ROOT CROPS IN DEVELOPING COUNTRIES, 1975 (million metric tons) Sweet Other Cassava Potato 1/ Yam Potato Root Crops 2/ Total Africa 45.1 1.7 18.7 5.2 3.9 74.6 Tropical America 36.9 8.6 0.2 3.9 0.8 50.4 Near East 1.1 3.1 - 0.5 - 4.7 Far East 21.1 6.9 0.1 7.7 0.2 36.0 Other 0.1 - - 0.2 - 0.3 Total (million tons) 104.3 20.3 19.0 17.5 4.9 166.0 (% of total) 63 12 11 11 3 100.0 1/ These figures may be compared with world potato production in 1974 of about 294 million tons. 2/ Includes Xanthosoma, taro, arrowroot, arracocha, yam beans, oca and olluco. Sources: D.G. Coursey, Tropical Root Crop Newsletter No. 9 (1976); IDRC, Cassava: The Development of an International Research Network (Ottawa, 1976), p. 13. Types of Tropical Root Crops 2.06 This paper includes discussion of five main types of root crops: 1/ - Cassava (manihot esculenta) - Potato (solanum tuberosa) - Sweet potato (ipomea batatas) - Yams (dioscorea spp.) - Aroids or cocoyams (xanthosoma spp., colocasia spp.). Potatoes are included in this classification, despite their greater importance in temperate climates and in the consumption patterns of developed countries, because of their potential for expansion to more tropical climates as adapted strains are developed (see para 2.27). 2.07 The Aroids. The aroids, including taro or cocoyam, in production terms the least significant among the TRCs, are perennial plants grown mostly 1/ Of these, only cassava and sweet potato are true root crops, i.e., where the roots are the storage organ. The main edible portions of the other crops are modified stems which function as storage organs or propagules. ° 16 as a staple or subsistence crop which perform best under hot, humid conditions, but which can be grown under a wide range of conditions, ranging from paddy culture to dry upland under irrigation. Best results are obtained on deep friable loam soils with a high water table. The maturation period varies according to variety, but ranges from 6-18 months. Taro yields, of from 7-30 tons/ha, take the form of edible corms which are generally grated or steamed, and sometimes fermented, before consumptiono Unlike most TRCs, some taro cultivars have a marked tolerance to highly saline conditions and can be grown under flooded conditions where there is need to reclaim saline land. Their potential as a calorie source in these difficult soil conditions makes tham a valuable crop in some development programs. 2.08 Yam. The principal edible yams are widely grown throughout the tropics under both rain forest and savannah conditions where there is a combination of adequate moisture and good drainageo Although generally considered to be drought resistant, yams require adequate moisture throughout their growing period0 On heavy soils with poor drainage, the tubers are likely to rot, while on poor soils, the weak root system is unable to obtain sufficient water or nutrients to develop reasonably-sized tubers0 Yams are affected by a variety of pests and diseases and severe losses may occur during storage from various fungal infections0 The average growth period is 8-11 months0 Yields, in the form of edible tubers, range from 5-15 tons/ha, vegetable (boiled, baked or fried), but may be dried and made into flour0 Only a very small portion of world output enters international tradeo In several countries of West Africa, the major production zone, yams are becom- ing increasingly expensive because of their high labor requirements and low average yield0 Due to their high price, they are being partially replaced in consumption patterns by cassava and in some urban areas by rice and wheat0 Replacement by cassava may affect nutrition adversely since yams have two and one-half times as much protein as cassava roots0 2.09 Sweet Potato0 Sweet potatoes are widely grown in tropical, sub- tropical and warm temperate areas under highly contrasting systems of agri- culture, ranging from very intensive,D high-input cropping to subsistence farming0 Varieties vary considerably in their adaptability to soil and other conditions0 They require at least 50 cm of rain during the growing season but can tolerate considerable periods of drought0 However, yields are very much reduced if soil moisture is inadequate during the period when root storage beginso Sweet potatoes can be grown on a considerable range of soils but are sensitive to alkaline or saline conditions and are subject to several insect pests and virus and fungal diseases0 Storage losses due to disease can be very substantial0 2.10 The crop is normally harvested from 3-8 months after planting and yields range up to 50 tons/hao Production is generally used for human food, although large quantities, mainly culls, are used as animal feed0 Sweet potatoes also are used in the production of starch and flour, and less com- monly, alcohol, acetone and lactic acid0 The highly perishable nature of sweet potatoes, together with high production costs, have been a severe limitation to commercial exploitation of the crop in many countries0 In - 17 - Japan, one of the world's largest producers, sweet potato (and derived starch) production are declining because of high production costs and increased availability of lower-priced maize starch. 2.11 Potato. Most potato varieties have very specific temperature re- quirements, thereby limiting the adaptability of this crop in tropical areas. -;.enly distributed rainfall is considered essential and drought, even for short periods, can have serious effects on yields and quality of the crop. Well-drained peat soils are particularly suitable, although potatoes can grow on most soils if drainage is adequate. They are subject to a large number of pests and diseases and prompt control measures are essential to realize eco- nomic yields. Late blight is one of the most serious of all potato diseases (and was responsible for the great potato famine in Ireland during the 1840s). Varieties show varying maturity, ranging from 3-7 months, with 3-4 months common in the tropics. Potato yields vary with variety, length of growing season, climate and soils. With efficient farming methods in temperate climate, yields well in excess of 25 tons/ha are common. Yields half as high are frequent in the tropics. 2.12 Potatoes are the leading starchy root crop of the temperate coun- tries and one of the eight leading staple food crops in the world. Annual production of potatoes is approximately twice that of all other edible root crops combined, although lesls than 10% of output occurs in the developing countries. Potatoes are also commonly used as the feedstock in production of starch, flour and alcohol. They are rich sources of carbohydrates, as well as protein, iron, riboflavin and ascorbic acid. In livestock feeding, the total digestible nutrient content of potatoes per ton is about one-third the value of alfalfa hay and one-quarter the value of a typical average grain mixture due to the high moisture content of potatoes. 2.13 The potential of this crop as an important source of high quality nutrients for humans in the tropics is the major reason for the increased research on this crop, as reflected in the establishment of the International Potato Center (CIP) in Peru. More than 100 million people in developing countries live in climatic zones where the potato grows very well. But a large portion of the population in countries normally considered as tropical or subtropical, such as Pakistan, Mexico, Brazil and Tanzania, live in areas not well adapted to potato varieties readily available today. Thus, the longer-term objective of research at CIP and several national centers is to develop genotypes adapted to extreme conditons and thereby extend potato production into the hot, humid, lowland tropics. 1/ 2.14 Cassava. For several reasons cassava is the TRC that in recent years has attracted the greatest interest of research workers and others 1/ N. R. Thompson, R. I. Wurster, K. D. Sayre, "Utilization of Potatoes in the Tropics," pp. 203-6, in Proceedings of the Fourth Symposium of the International Society of Tropical Root Crops, CIAT, Colombia, 1-7 August 1976. IDRC Publication 080e. - 18 - concerned with rural development. It is by far the most important of the tropical root crops with annual production exceeding 100 million tons grown on some 12 million ha. It is produced in more than 80 countries, although fewer than 20 account for 90% of production and 6 produce two-thirds of global output. It is a major source of calories for some 300 million people in developing countries, many of whom are in the lower socioeconomic classes0 It is a low-risk crop with annual yield variations typically less than half the yield variation of basic grains. An important factor behind current interest is cassava's remarkable capacity to produce and store large amounts of carbohydrate energy, frequently under adverse environmental conditions. It is recognized as one of the world's most efficient plant converters of solar energy to carbohydrates, with an ability to store energy at a daily rate per unit area of land which substantially exceeds that of other high-yielding crops such as maize and rice0 1/ Table 2: MAXIMUM RECORDED YIELDS OF SELECTED STAPLE CROPS Crop Tons/Ha/Year Calories/Ha/Day (000') Cassava 71 250 Maize 20 200 Sweet Potato 65 Ro0 Rice 26 R76 Sorghum 16=18 140O158 Wheat 12 220 Banana 39 80 Source: CoAo Vries, et.al0, 1976. Netherlands Journal of Agricultural Science, 15:241=248; reported in Barry Nestel and James Cock, Cassava: The Development of an International Research Network, IDRC 059e, po 140 Sugar cane is not included in Vries' list, although it produces very large quantities of calories per ha per day0 2.15 A major factor behind the extensive production of cassava through- out the world is its adaptability to a wide range of soil and moisture conditions0 It can be successfully grown in areas with rainfall ranging from 50 cm per annum to over 250 cm0 Except at planting, cassava can withstand periods of prolonged drought and is therefore a valuable crop in regions of 1/ K0 F0 Gregory, "Cassava as a Substrata for Single-Cell Protein Production: Microbiological Aspects," pp. 72-78, in Cassava as Animal Feed; Proceed- ings of a Workshop Sponsored by the IDRC and the University of Guelph, April 18-20, 1977. - 19 - low or uncertain rainfall. 1/ While light, sandy loams of medium fertility give the best results, the crop can be grown successfully on soils ranging from stiff marine clays with a pH of 8-9 to sands or loose laterites with a pH of 5-5.5. It produces some crop on acidic soils of limited fertility, of which some 1,750 million ha exist in developing countries. When grown on clay soils, the plant produces stem and leaf growth at the expense of roots--a situation in which carbohydrate yields are unusually low. Saline and swampy soils are unsuitable. 2.16 Although cassava has a remarkable ability to grow on soils of limited fertility, it removes large quantities of nutrients from the soil. Research suggests that with fresh root yields of 40 tons/ha and 50 tons/ha of tops, annual nutrient removal is of the order of 285 kg of nitrogen/ha, 60 kg of phosphorous, 380 kg of potassium and 160 kg of calcium. 2/ The require- ment for potash appears to be particularly critical. If this nutrient .is not available in adequate quantity, yields are sharply reduced; starch content is low; and the content of potentially toxic cyanide compounds is increased. Proper fertilization is therefore essential to prevent serious soil depletion and to maintain cassava yields over extended periods. 2.17 Despite a common view to the contrary, cassava is susceptible to various pests and diseases, although it demonstrates greater resistance than most other TRCs and nearly all grains. Bacterial plight and mosaic are the important diseases. The latter is particularly serious in Africa. Stem borers, weevil borers, spider mites and scale insects are troublesome in many production zones. 2.18 The maturation period for cassava is 9-24 months, depending on the variety, climate and soil conditions. A few quick-growing varieties can be harvested in 6-7 months, but optimum yields are obtained only after 9-12 months. When used for human consumption, the roots are normally harvested within 12 months, but if required for processing, they may be left to reach full maturity after 18-24 months. The root can be left in the ground for extended periods without serious deterioration, a factor which adds to the utility of the crop 1/ This ability to produce large quantities of carbohydrates has earned the title "famine crop" for cassava. The title may be well-deserved. During the 1943 drought-induced famine in India, more than a million people reportedly died in Bengal (a state where cassava was not grown), while there were almost no starvation deaths in Kerala, where cassava was grown extensively. Reported in University of Georgia, A Literature Review and Research Recommendations on Cassava; AID Contract No. csd/ 2497, March 1972, p. 12. 2/ A.J. De Boer and D.A. Forno, "Cassava: A Potential Agro-Industrial Crop for Tropical Australia," The Journal of the Australian Institute of Agricultural Science, pp. 241-251 (December 1975). 20 - in some development programs and enhances its value where security of food supplies is an important consideration. Yields show wide variations around a global average of about 9 metric tons of fresh roots per ha. i/ 2019 Production of cassava in 1975 exceeded 100 million tons (fresh roots) of which more than 40% was produced in Africa, about 30% in South America and the balance in Asia (Annex 3, Table 10 World Production of Cassava, 1973, 1975)o By a substantial measure, Brazil is the largest pro- ducer, accounting for more than a quarter of global production, followed by Indonesia (12%), Nigeria (10%), and Zaire (9%). Smallholders account for most global production0 Aggregate figures1, which are subject to considerable error, suggest that cassava production since the early 1960s expanded by about 2.5% p.a0 2/ 2o20 Cassava, like several other TRCs, is utilized in three major ways: (1) for consumption as food by humans; (2) as animal feed; and (3) as a raw material in the manufacture of various industrial products, of which starch is the most important0 Roughly estimated, these global proportions are 95, 4 and 1% for all TRCs and 91, 8 and 1% in the case of cassava. 3/ Considerable ',ariation in these proportions exists among countries0 2.21 As a human food item, cassava is consumed as a boiled or roasted vegetable or as pastes, meal and flour0 The better known preparations in- clude gari, and fufu in West Africa, farinha de mandioca in Brazil and a similar product in Indonesia. In sonme counries, as in West Africa, virtually all cassava is consumed as human food; in others (e.g., Thailand) almost all production is utilized in livestock feeding0 2.22 The animal feed market is a small1, but rapidly growing, outlet for cassava products0 Sizeable quantities are used internally for this purpose in producing countries, although the most dynamic growth is in the export market0 Most of this takes the form of cassava pellets shipped to the EEC 1/ Reported yields from the Malagasy Republic are indicative: Zone Strain Type Yield (tons/ha) Dry, low-lying Local 7-8 Dry, low-lying Selected 12-35 Humid, low-lying Local 9-25 Humid, low-lying Selected 28-66 Medium altitude (300-900 m) Local 4-20 Medium altitude (300-900 m) Selected 30-80 High altitude (300-1,300 m) Local 412 High altitude (300-1,300 m) Selected 12-25 2/ Cassava as Animal Feed, opo cit., po 96. 3/ T. Phillips, Tropical Roots and Tubers: Future Markets, Supplies and Uses (draft), opo cit., po 55o - 21 - countries for use in the manufacture of compound animal feeds. In the 1970-75 period, world trade in cassava products trebled from US$107 to $306 million, with dried roots/tubers in various forms accounting for 90% of the total (Annex 4, Table 2, World Trade in Cassava, 1970, 1973-75). 2.23 The starch market is complex and among the slowest growing outlets for cassava. Competition with sta\rches from other sources such as maize and potatoes is keen. The two major consumers of TRC starch are the United States and Japan, although in these markets cassava starch is only a small and declining proportion of total starch utilization. This decline reflects the difficulty of providing a steady supply of high-quality starch and the growing price disadvantage of TRC starches from developing countries, due in part to higher transport costs vis-a-vis those for starches from locally- available raw materials. The continued development of specialized maize-based starch in major importing countries has further eroded the TRC starch market. Small, but growing, outlets for TRC starch exist in several developing coun- tries and in specialized end uses. 2.24 Two basic considerations in the utilization of cassava (and most other root crops) are the relative perishability and bulkiness of the fresh roots. Harvesting is laborious and transportation of roots is costly and can be done efficiently only if good roads link production areas with areas where the product is consumed or processed. Typically, utilization systems are based on weight-reducing processing techniques near the point of production, as evidenced by the extensive development of village-level processing activilities in, e.g., West Africa and Brazil. In large-scale industrial use, these attributes of bulkiness and perishability pose particular problems about the organization of input supply and transportation to the factory. Furthermore, because cassava is propagated vegetatively, planting materials are bulky and difficult to store. In consequence the dissemination of disease-free, improved varieties is more difficult than in the case of crops multiplied from seeds. The Research and Development Effort 2.25 While traditional grain crops have been subjected to intensive re- search over a long period, and thereby have reached a larger potential of their yield potential, genetic research on the TRCs is at a relatively early stage. Largely for this reason, the potential is very good for continued substantial yield increases among the TRCs. Since about 1970, research on the TRCs (including potatoes) has increased sharply from around US$0.5 million to more than $5 million at present. 1/ A major factor in this changing research effort has been the empharsis given to the TRCs by the International Agricul- tural Research Centers (particularly CIAT, IITA and CIP) and several national research organizations,:some of which have a long history of work on this group of crops. 1/ This history of this research effort for cassava is well-described in B. Nestle and J. Cock, Cassava: The Development of an International Research Network, IDRC-059c, 1976. 22 2026 The principal objective of cassava research at CIAT ic to develop cassava production technologies which require relatively low levels of pus° chased inputs, adapted to a wide range of ecological conditions and suitable for use on both large and small farms0 1/ IITA's root and tuber improvement program puts top priority on cassava, followed by yam, sweet potato and coco- yamo The primary objectives of this res arch program are to minimize factors that currently limit production, to develop plants adopted to wore advanced farming systems and to improve product quality for human consumption. A major focus at IITA is to identify resistance to African mosalac (unique to cassava in Africa and India) and cassava bacterial blight and to combine such resistance with a high-yielding capability. 2o27 The basic mandate of CIP is to improve the yield and yield stability of the potato in developing countries and to develop varieties than can be efficiently grown both in highland cold regions and in the humid tropics. CIP organizes its research around the collection1, maintenance and utilizEtion of genetic material, control of diseases, viruses and pests, adaptation to environmental stress, post-harvest technology, improvement of nutritional quality, and seed production technology. Work is underway on the control of fungal, bacterial and virus diseases and nematode pestso A major focus to date has been improved seed production technology0 Current work on the use of the botanical potato seed as a means of propagation offers the promise of sharply reduced production costs and will greatly facilitate the dissemination of improved varietieso 2.28 The work at CIAT and IITA on cassava agronomy is beginning to pay off in terms of higher yields0 In Colombia, the national average yield is about 8 tons/ha but local lines grown with improved cultural practices and minimal inputs produced 16 tons/ha while the best CIAT lines with improved cultural practices averaged 28 tons/ha0 2/ Ongoing breeding and selection work promises further sizeable yield increases0 CIAT agronomists now con- sider it technically possible to raise cassava yields in Latin America by 25-100% with a few recommended practices requiring cash outlays of no more than US$3/ha0 Research on cropping systems with beans and cassava has shown it possible to produce as much as 85% of normal monoculture yields for both crops when the two were planted on the same dateo Research on cassava utili- zation is less advanced1, and it seems likely that more work in this area will be required to best utilize the potential of this crop in rural development0 IITA's work on cassava has advanced to the point where high-yielding, disease- resistant lines are available for distribution to farmers through national programs0 1/ Ibid., po 160 _ 2/ CIAT, Program and Budget Proposal, 197879D, Po 80 - 23 - III. TROPICAL ROOT CROPS AS FOOD 3.01 For many in the developing countries, the TRCs are a basic, if not the dominant, source of staple calories. The table below, derived from the 1974 Food Balance Sheets of FAO, indicates several countries whose populations consume an important fraction of their calorie supplies in the form of TRCs. Because the data refer to national averages, some countries, e.g., India, Zambia, Ecuador, Sri Lanka, are omitted, despite the great importance of TRC consumption in certain regions within those countries. Even within the countries listed, interregional variation can be quite high, as in Brazil, Indonesia, or Nigeria. Because national food balance data typically are weak, referring only to averages and abstracting from consumption variations due to income, ecological factors or cultural differences, the information in the table is suggestive, not definitive. But what cannot be disputed is that the TRCs are important food items in many of the poorest countries. 3.02 Because of their adaptability and efficiency in conversion and storage of solar energy, the TRCs generally are basic sources of low-cost calories. Data from Nigeria, which are probably indicative of relative calorie costs from various food staples in other important TRC-producing countries, suggest that calories from cassava are frequently only 25-50% as costly as calories from traditional grains and pulses. 1/ On the other hand, the TRCs are not economical sources of protein, but when grown in combination with efficient protein sources such as pulses, offer a relatively low-cost means of meeting both protein and calorie requirements. 1/ The following data relate to retail prices in Ibadan, Nigeria in March 1973. Index of Prices Per Unit of Food Calories Crude Protein (Price of Gari=1-0) Gari 100 100 Melon seed 435 24 Cowpeas 227 16 Maize 108 39 Rice 242 25 Yams 132 22 Source: Federal Office of Statistics, Lagos and V.A. Oyenuga, Nigeria's Food and Feeding Stuffs (Ibadan, Nigeria: Ibadan University Press), adapted from J. April, G. Hersh, D. Rogers, C. Slater Cassava's Role as a Food Staple: A. Cross Disciplinary Systems Analysis of Nutrition Problems in a Cassava Dependent Culture, USAID, Preliminary Report, September 1974, Ch. III, p. 19. Table 3: TROPICAL ROOT CROFS AS A SOURCE OF CALORXES IN SELECTED COUNTRIES, 1974 P Ecant AvergSeam of Total Total Celoria Caloriao Calorieo Country P ltion GNP per it' Cons on from TRCI/ from TRCs (midol915 (mgket pricaos psi capita millions) 1975) paz day) Zaire 24.7 140 19880 1906o 56 Ghana 9.9 590 29320 870 38 Togo 2o2 250 29220 850 38 Ivory Coast 6.7 540 29650 820 31 Nigeria 75.0 340 2C,OO 720 35 Mozambique 9.2 180 1 ,980 720 36 Angola 5.5 370 2,020 700 35 Rwanda 4.1 100 29090 630 30 Benin 3o1 130 29010 590 29 Tanzania 14o7 170 29000 540 27 Cameroon 7.4 280 2,370 530 22 Paraguay 2.6 580 29720 450 17 Madagascar 8.8 200 2D390 370 15 Bolivia 5.6 360 R,850 290 16 Guinea 5.5 130 29000 290 14 Uganda 11.6 230 2D100 300 14 Peru 15.4 760 2D360 320 13 China (PRC) 8228 1380 29330 270 12 Indonesia 132.1 220 29130 250 12 Kenya 13.4 220 29120 200 9 Brazil 107o0 19,030 29520 230 9 Jamaica 2.0 1,110 29660 230 9 Belize 001 670 29440 200 8 1/ Figures rounded to the nearest 10o Source: Calorie consumption data are from FAO. Population and income figures are from World Bank Atlas 19770 - 25 - 3.03 The orthodox view is that rising incomes among most consumers result in some shift away from the TRCs to other staple food items such as the grains. Indeed, demand studies suggest that in many countries this is the case, par- ticularly among middle-income countries such as several in Latin America. However, as a generalization, it appears that income elasticities of demand for TRCs are: (i) low but positive among most low-income countries; and (ii) moderately negative among middle-income countries. Income elasticity coeffi- cients for different income groups within a country could be expected to show the same pattern. 1/ The net overall impact of income growth on developing country consumption patterns appear to be slight. FAO estimates suggest that the portion of total calories supplied by the TRCs declined less than one percentage point in the past 10-15 years. 3.04 Cross elasticities of demand among the cereals and TRCs are high and subsitution frequently is easy. 2/ However, there tends to be a general hierarchy of preferences. Cereals are usually preferred to root crops while wheat and rice are preferred to the coarser grains. Among the TRCs, consump- tion of taro and yams sometimes increases with rising incomes and there may also be strong local preferences for sweet potato and cassava. 3.05 The role of the TRCs in meeting future food needs depends in part on what those needs are perceived to be. The prospect that the TRCs might provide a larger share of dietary needs of many in the LDCs is viewed with concern by some nutritionists. Calorie and protein deficiencies are assumed as the major problems, with the protein shortage generally considered to be the most intractable. Despite general appreciation of the efficiency of the TRCs in producing calories, there is little enthusiasm for this food group as producers of protein and other nutrients. This lack of enthusiasm for the TRCs is frequently shared by economic planners in developing countries and others concerned with food and nutrition policy. Many feel that the TRCs are an inferior food group and that a major objective of food production policy should be to aim for something better. To these individuals the TRCs are considered largely as animal feed or only marginally-palatable food for starving people, not the basis for planning for food security and nutritional balance. 1/ Recent analysis in Indonesia suggests that cassava products will continue to be a basic source of calories even with moderate per capita income growth. Careful studies using 1976 Socio-economic Survey results indi- cate that the income elasticity coefficient for cassava remains positive for two-thirds of Indonesia's population. One-third of the population has an estimated expenditure elasticity greater than 0.55 for fresh cassava. 2/ The cross elasticity of demand indicates how purchases of a particular product are influenced by price changes of competing or substitute products. - 26 3006 These concerns require closer examinationo Two comparisons are useful: (i) the nutrient content of the TRCs vis-a-vis the grain staples; and (ii) the nutrient content of the major root crops. In terms of nutrient content per kilogram of fresh root, the root crops generally compare poorly with the grains in terms of protein, carbohydrates, iron, thiamine and iron, but are good sources of calcium and, in the case of sweet potatoes, an excep- tional source of Vitamin A0 (See Annex 4, Table 3, Food Composition of Staples.) Among the TRCs, fresh cassava roots compare favorably with other root crops in terms of food energy and carbohydrate content, but rank at the bottom of this list in terms of protein contento Cassava roots are relatively rich in calcium and ascorbic acid and contain nutritionally-significant amounts of thiamine, riboflavin and niacin0 The nutritional merits of cassava, when compared with other tropical root crops, are even greater when yields of Table 4: CALORIES SUPPLIED BY CEREAL AND TROPICAL ROOT CROPS, 1961-65D 1972, 1974 1961-65 1972 - 1974 Developing Countries (Total Calories) 2,106 2,165 2,170 Africa, Total calories 2D049 2,087 2,078 From cereals 999 1,032 1,023 % of total 48.7 4904 49.3 From roots and tubers 464 445 448 % of Total 22.6 2103 21.6 Latin America, Total calories 2,349 2,470 2,491 From cereals 954 973 1,018 % of total 40.6 3904 50.9 From roots and tubers 177 179 159 % of Total 7o5 7.2 6.4 Near East, Total calories 2,299 2,424 2,457 From cereals 1,488 1,511 1,537 % of total 64.7 62.3 62.6 From roots and tubers 56 52 52 % of Total 2.4 201 2o1 Far East, Total calories 2,019 2,054 2,053 From cereals 1,350 1D415 1,422 % of total 6609 686 . 70.7 From roots and tubers 72 64 67 % of Total 3.6 301 3.3 Source: Monthly Bulletin of Agricultural Economics and Statistics, FAO, May 1977. - 27 - calories per ha per month or per man-day are considered. It has been esti- mated that 1 kg of fresh cassava could contribute 66% of the calcium re- quirements of an active adult living in the tropics, as well as 125% of the required iron, 58% of the thiamine, 83% of the riboflavin, 40% of the niacin and 1600% of the Vitamin C. 1/ However, much of this vitamin and mineral content is lost in traditional processing methods. The protein quality of cassava roots is inferior to most alternative carbohydrate sources and is particularly deficient in methionine and tryptophan relative to the content of other root crops. Annex 4, Tables 4 and 5, examine the protein quality of the TRCs and compare their essential amino acid content with a high quality reference protein. Again, cassava roots rank poorly in comparison with other TRCs. 3.07 Although the root or tuber is generally considered the predominant food source of the TRCs, plant leaves also are important in some circumstances. Cassava leaf is consumed by humans in Zaire, New Guinea and Central Java and is particularly important during times of food shortage. Cassava leaves are rich in protein, with a crude protein content ranging from 20-35 percent. 2/ The amino acid spectrum of the leaves is as good or better than that of soy- bean meal. Also noteworthy (See Table 5) is the high carotene content of the cassava leaf, a factor which explains the relatively low incidence of xero- phthalmia (Vitamin A deficiency) in low income areas where the leaves are consumed by humans. Table 5: AVERAGE NUTRIENTS IN 100 GM DRY MATTER OF GREEN LEAVES, BEANS, AND FRUITS Protein Calcium Iron Carotene Vitamin C gm mg mg I.U. mg Green leaves 1/ (amaranth, cassava, ipomea) 30 1,300 20 6,200 810 Beans (mung type) 27 160 7 150 0 Fruits 1/ (banana, orange, papaya, mango, pineapple) 5 140 3 4,500 315 1/ Data represents an average of all crop leaves or fruits mentioned. SOURCE: H.A.P.C. Oomen, "Vegetable Greens: A Tropical Underdevelopment", in Chronica Holticulturae, 4 (1): 3-5 (1964). 1/ Jones, Manioc in Africa, op. cit., p.315. 2/ University of Georgia, "Biochemical properties and Nutritive Value of Cassave," in A Literature Review ...," op. cit., p. 119. 28 3008 The potato tuber is distinctive among the root crops in terms of its relatively high protein content and its high levels of Vitamins B and C. Given its relatively short maturation period and high yields, it offers the potential of making a major contribution to human nutrition in the areas to which it is adapted. This will be broadened with identification of more adaptable varieties. Recently, potato clones maturing in 60 days have been grown in the humid lowlands of Peru. Potato protein quality is good and clones with 10-12% protein (dry weight basis) have been identified in the CIP germ plasm bank. I/ This compares with about 7% protein content for milled rice. The potential for breeding potatoes with even higher protein content is considered excellent0 Clones with 20-25% of protein dry matter are known. With early-tuberizing varieties, 3-4 crops are possible in one year, thereby tripling or quadrupling traditional annual yields per ha and providing a continuous harvest to avoid complex storage problems0 On this basis the potato outranks all major world food crops in terms of protein production per unit of time. Only three short season crops--soybeans, peas, and beanso- exceed the potato in this regard0 2/ The TRCs and Nutritional Balance 3.09 Given the World Bank's concern with global food supplies, it is appropriate to explore what impact increased development emphasis on the TRCs might have on existing nutritional problems0 The most pervasive of these prob- lems is increasingly recognized as a shortage of calories, although deficiencies in other nutrients may be important in localized situations and for particular groups such as children and pregnant women0 Nutritionists have frequently observed that food consumption by adults (as opposed to children and pregnant or lactating mothers) in many developing countries tends, on the average, to show nutritional balance in the sense that if per capita calorie supplies are adequate, supplies of most other nutrients will also be sufficient0 3/ This is not always true0 But a good starting point is to examine what present nutrient intake levels are, what they would be if "scaled up" to adequate calorie levels, and what diets would look like if only TRCs were used to increase caloric levels0 1/ Based on the mix of amino acids, with egg protein equal to 100, potato protein ranks at 710 This may be compared to pork at 84, beef at 80, rice at 57, wheat flour at 52 and navy beans at 47. 2/ Thompson, Wurster and Sayre, "Utilization of Potatoes in the Tropics," opo cit., po 206. 3/ The nutrition problems of children and pregnant or lactating mothers are important and frequently are severe in low-income areaso But often the intra-family distribution of food, more than its overall supply, is critical0 Changing family behavior in this regard is difficult, but may be easier if total food availability of the proper type is increasedo - 29 - 3.10 Against this view is the widespread impression that consumption patterns based heavily on TRCs almost always tend to be deficient in protein, with particularly adverse results for children. And, indeed, it is demon- strably true that if a child eats mainly cassava or other high-calorie, low-protein food, he is almost certain to suffer from protein deficiency. However, experience in several regions of developing countries suggests that TRCs can be quite important on average in consumption and yet child mortal- ity, one measure of nutritional adequacy, may be low. In Kerala State, for example, average TRC consumption exceeds 740 (1970/71) calories/day, although infant mortality has been among the lowest in the poorer LDCs. 3.11 In such circumstances of balanced, but calorie-deficient, consump- tion patterns, the most efficient course of action to improve nutrition would seem to be to concentrate on improving the quantity and promoting the effec- tive use of indigenous foods which serve most of the population quite well. In Africa it has been suggested that a mixture of 70% cassava flour and 30% groundnut flour would be superior to the commonly-used maize flour in terms of available protein, caloric density and niacin. 1/ The practical implication of this is that the addition of a handful of groundnuts to the daily diet of those over 6 months of age in nutritionally-deficient areas of Africa would solve a large portion of the nutritional problems which exist. 2/ The sug- gestion is that this small modification of TRC-based diets in much of Africa would make those diets nutritionally adequate. And, as Latham points out, 50 gm of groundnuts per day for each child would require only 0.03 ha--an amount of land available to most African farmers. 3.12 The balance sheets referred to above in general support the obser- vation that diets in countries with moderate to high TRC consumption are not seriously deficient in protein. Excluding Zaire, the data relate to the African nations where TRCs are very significant in the diet. Caloric intake from TRCs ranged from 500-900, making up 20-40% of total calorie consumption. Seven of these 10 countries exhibit per capita caloric consumption levels under 2,200 per day (the mid-range of many estimates of calorie requirements) and only one has over 2,400 calories. Yet even in this group, daily protein intake is under 40 gm in only one case and under 50 in three. In the second group of 12 countries, TRC calories account for 8-16% of the total. In 5 of these countries, total caloric intakes in the year of survey were under 2,200, but none had daily protein intakes under 40 gm and only 3 were under 50 gm per day. Again, consumption patterns in this second group of countries sug- gest that moderate consumption of calories from TRCs may be consistent with adequate protein consumption. 3.13 Also of note is the tendency to consume, in per capita terms, not more than 1,000 calories/day from TRCs. Even data on low-income regions 1/ James McDowell, "In Defense of African Foods and Food Practices," in The Tropical Doctor, January 1976, p. 37. 2/ Ibid., citing Latham in Human Nutrition in Tropical Africa, p. 182. - 30 - within a country, such as Kerala, Brazil's Northeast and the poor districts of Jogjakarta, support the view that daily per capita intakes of calories from TRCs rarely exceed 1,000o This natural upper limit to calorie consump- tion from this source, as might be expected from considerations of human stomach capacity and palatability, should be taken into account in the design of cropping systems for small farmersO This fact should also permit easier resolution of dietary deficiencies since one-half to two-thirds of total calories would come from other sources with nutrients that cassava or other TRCs may lack. The TRCs and Food Consumption Patterns: Some Case Studies 3.14 More detailed case studies provide additional information on the question of nutritional adequacy in countries where TRCs are important in consumption. Three examples, one from each major developing continent, are chosen: Kerala in :ndia, rural northwest Zambia and northeast Brazil. In the Kd.Aal and Zaa.., -es onlv calories and proteins are considered; in Brazil the adequacy u ucher nut,:e=tcs also is f ;min. : In each caze, average daily intake of calories from TRCs is 50j-7500 3.15 Kerala. The south Indian state of Keral2 '- i population in excess of 20 million. Its ciaracter-stLcs of low i' mes, in combination with relatively Low infant mortalitv. high average 'uls of educational achievement and 'ocngevity has made it a subject of tonsiderable study. However, the relacively high level of sJcial services in the state, high literacy rates, unsually varied diets, 2tc0 make Kerala somewhat atypical of other low-income regions or countrieso A recent UN study, concluding that food balance sheet data were more reliable than sample surveys, compiled the information on food consumption given in Table 6. Table 6: FOOD CONSUMPTION IN KERALA, 1970/71 (average daily per capita value) Total Food Items Consumption Calories Protein (gm) (gm) Rice 289 1,000 18o5 Cassava (Tapioca) 474 744 3.3 Coconuts 60 267 207 Fruit 87 68 07 Fish 41 46 803 Milk 30 23 100 Meat 5 6 lol Oil 24 212 - Sugar 25 100 - Subtotal 2,466 35.6 All Other 53 202 Total 2,519 37.8 SOURCE: United Nations, Poverty, Unemployment and Development Policy: A Case Study of Selected Issues with Reference to Kerela; ST/ESA/29, pp. 180-91 (1975). - 31 - 3.16 The data suggest that calories appear to be adequate, while pro- tein may be deficient (assuming the protein quality is 60-70% that of egg protein). 1/ One question is what adjustments in the diet might be made to provide adequate protein. Reduced cassava consumption could be suggested, although an important consideration is that it is a very cheap source of calories--typically about 40% the cost of calories from rice or sugar. 2/ Chickpeas usually cost at least 50% more than rice on a calorie basis but have three times as much protein. Reducing rice consumption by a third and making this up with increased consumption of chickpeas would reduce calories slightly but increase protein by 6.5 gm (and provide an improved amino acid mix). One conclusion is that even when protein is short, it may be wise to retain the efficient calorie source and substitute other foods for more efficient protein sources. The practicality of this point would, of course, be determined by such considerations as taste and agronomic possibilities. 3.17 Zambia. This case refers to information from the 1971 Zambia food survey pertaining to the rural northwest region where TRC consumption is unusually high. 3/ The typical diet is extremely varied, including maize, sorghum, cassava, sweet potatoes, pulses, leaves, fruits, fish, game, insects and birds. Summary data indicate that average daily per capita consumption of calories totaled just over 1,700 while intake of protein was nearly 45 grams. This combination of general protein adequacy, combined with a calorie deficiency, is notable. From the nutritional viewpoint, higher consumption of TRCs would be justified even though they already contribute about 540 calories per day. For the typical adult the most pressing need is more calories, although a modest protein increment would increase the safety 1/ However, analysis of amino acid composition indicates that all amino acids were supplied in excess of requirements for adults. Indeed, the safety margin was 2:1 (twice as many essential amino acids available as needed to utilize the 30 gram/day requirement), except for the sulphur- containing group was was 50% in excess of requirements. This suggests that indicated protein intake levels may be approximately adequate for most members of the population. Infants, children and pregnant and lactating women are likely to be notable exceptions. Local consumption studies are required to provide more definitive evidence. 2/ It has been pointed out that in Kerala cassava is normally eaten with fish in a combination generally considered to provide an unusually at- tractive protein-calorie balance. See Davidson R. Gwatkin, "Nutrition Plap ing and Physical Well-Being in Kerala and Sri Lanka," A Paper Pre- pa ted for the Interciencia/AAAS Symposia on Nutrition and Agriculture: St ategies for Latin America, February 13-14, 1978. Other factors are also present in Kerala--good health care, sanitation, and high green and leafy vegetable consumption relative to the rest of India. The main point is that high TRC consumption is consistent with healthy children. 3/ Published by the FAO in 1974; ESN:DP/ZAM/69/512. - 32 - margin. The young and pregnant/lactating mothers would, of course, require both additional calories and protein. 3.18 Northeastern Brazil. This survey, taken in the 1960s, examines the adequacy of calories, proteins and selected other nutrients and relates consumption to income. 1/ The major finding of the study, aside from a pervasive shortage of Vitamin A due to the exclusion of fresh vegetables from the survey, was that the major nutritional problem of the poor is not enough to eat rather than the wrong types of foods0 While calorie consumption always rose sharply with income--from inadequate levels among the poor (bottom 5%) to adiequate levels among the middle-income class (40-60 percentile) to excessive among the rich (top 3%)--the ratio of nutrients per 1,000 calories often changed little or not al all by income classes0 Even though TRCs provided over 30% of total calories for all income groups, the poorest income class in the rural Northeast was consuming more th-n enough protein, but only 70% of the calorie requirements. 3.19 This unusually detailed study among the poorest in a poor region indicates that in general the quality of the typical diet is good, despite the high TRC intake0 Among this group of rural poor, intake levels of protein, phosphorous, iron, thiamin, and niacin were adequate. Vitamin A intake was deficient, as was riboflavin, although to a lesser extent0 Only these two nutrients would remain in short supply if increased cassava alone were used to provide the additional needed calories0 2/ In sum, nutritional adequacy seems to require more of the same food mix, rather than sharply altered consumption patterns0 The remaining deficiencies, such as Vitamin A, could be addressed at low cost through home gardens or the supplementation of cooking oils0 TRC Foods in More Highly-processed Form 3.20 The TRCs also have important nutritional roles in the production of composite flours and are already used in this manner in a number of 1/ Survey data were published in 1970 by the Getulio Vargas Foundation0 Subsequent analysis of the data is reported in the American Journal of Clinical Nutrition, Vol, 30, June 1977, ppo 955-964. 2/ The only other minor problem with the quality of the diet for this group was that the methionine (one of the essential amino acids) score was low and that supplementation might be warranted0 However, any reasonably balanced expansion of this diet would eliminate this problem, since protein provides 14% of all caloric energy, as compared with a require- ment of less than 9%. - 33 - countries. 1/ While the major motivation for wider use of composite flours is to reduce foreign exchange expenditure on wheat imports, other benefits would be expected. The substitution of wheat imports by flour components derived from local sources would stimulate agricultural production, local processing and employment. If the indigenous product is widely grown, as is generally the case with the TRCs, income distributional benefits could result as well. If production and processing are efficient, a lower cost food staple could be made available to consumers. When combined with programs of nutrient fortifi- cation, the wider use of composite flours would also generate nutritional benefits. 3.21 Research and development work on composite flour using root crops and other local products is perhaps most advanced in Colombia. Based on initial research in 1971-72, it was concluded that while rice and maize flours are the preferable ones for use as non-wheat components in composite flours, cassava flour and starch also have good technical possibilities. The pilot work demonstrated that the production of bread from wheat flour diluted up to 30% with non-wheat components is possible on a commercial scale. But the large-scale introduction of such flours would require a concerted effort by both the public and private sectors to ensure the wide-spread availability at attractive prices of non-wheat raw materials. While cassava flour and starch demonstrate good baking qualities, the capacity in Colombia of producing adequate quantity and quality of these cassava products is severely limited. The ongoing World Bank nutrition project will, among other things, support the development of village-level processing technology to overcome this problem. 2/ Expanded cassava production and lower prices would be required if this type of composite flour were to be economically-attractive to millers and consumers. At relative prices prevailing in the early 1970's in Colombia, the price of composite wheat flour which included cassava would be higher than that of a pure wheat flour. 3.22 Potentially large foreign exchange benefits could result from successful introduction of composite flours which use local products. Nigeria may be a good example, given the importance of the TRCs in that country and rapidly-growing food imports. Recent work shows that bread of superior 1/ The Composite Flour Program was initiated in 1964 by FAQ to determine the possibilities of using materials other than wheat in the manufacture of bakery, confectionary and pasta products. Composite flours are defined by FAQ to include all mixtures used in the manufacture of these products, either as additives to wheat flour or in complete replacement. The program was conceived against the background of rapid growth in consumption of wheat products in developing countries and the consequence of growing wheat imports and foreign exchange expenditure for basic food products. Much of the analytical work for the program has been done by the Tropical Products Institute (TPI) in the United Kingdom. See Economic Aspects of the Composite Flour Programme, TPI, London, June 1975. 2/ World Bank, Colombia: Appraisal of An Integrated Nutrition Project, Report No. 1583-CO, (August 29, 1977). - 34 - nutritional quality could be obtained by commercial baking methods with composite flours of 75% wheat flour, 20% cassava starch and 5% soy flour. The use of chemical binders would permit a further increase in the cassava content. If 10% of the wheat flour consumed in Nigeria in 1974 had been replaced by cassava starch, the value of wheat imports could have been reduced by at least U.S.$5 million. This would have required some 21,000 tons of cassava meal, equivalent to 60,000 tons of fresh roots. Total production of cassava in 1975 was in the range of 10 million tons0 3.23 In Brazil work on the fortification of cassava flour has been underway for several years0 By mid&1978 a commercial firm was selling some 50 tons/month of cassava flour fortified with up to 15% of a processed soy producto Although good market acceptance of the fortified product was reported and increased production is planned, the retail price wass about twice that of the traditional cassava flour, thereby putting the more nutri- tious product beyond the reach of many low-income consumers0 Some Conclusions 3o24 Several conclusions are suggested regarding the TRCs and nutritional improvement: - A high intake of calories from TRCs is common among the low-income countries and among poor classes within those countries0 In many circumstances this consumption pat° tern is likely to be associated with a reasonably well- balanced diet, which while frequently deficient in calo- ries, has a proper protein/calorie ratio0 This is gen- erally true because most groups limit the amount of TRC calories in the diet and have significant amounts of cereals, legumes, fish and/or other animal products as well; - Calorie shortfalls among the poor appear to be the most serious0 but not the only significant nutritional problemo In most cases, the caloric shortfall should be met by increasing intakes of a variety of commonly-eaten foods, of which the TRCs can be an important group; - The special nutritional problems of the young and preg- nant and lactating women appear to be widespread and, in many cases, serious0 But in some cases these problems are soluble if effective nutrition education is combined with the increased availability of local foods0 Even in these circumstances, expanded use of TRCs appear to be warranted, if combined with other foods rich in protein and other nutrients0 - The use of the TRCs in composite and fortified flours may be among the most effective means of nutritition - 35 - intervention in areas where these crops are basic food staples. Expanded use of the TRCs in this form will require lower production and processing costs than now exist in many countries. - Clearly, careful examination of typical diets for par- ticular age and socioeconomic groups or geographic regions is required before unequivocal recommendations can be made regarding nutritionally-adequate consumption patterns and supporting food production systems. Rules of thumb are difficult to find. But low-protein food alone should not be encouraged if this would move overall protein content of the diet below 8-10%. As the demand for food grows in areas with limited agricultural potential, there is reason to believe that the TRCs, and cassava in particular, will remain important in the diets of low-income groups in both rural and urban areas. 1/ As Jones noted nearly twenty years ago: "The same facts that caused the spread of manioc [cassava] over a great part of West Africa in the past 150 years may be expected to continue to operate in the future. Increased urban population, more intensive cultivation, recognition of manioc's higher productivity and of its unusual resistance to pests, disease and drought will continue to favor its production. Manioc is probably not yet grown in all areas where economic production is possible and it may be expected to gradually enter such areas. It is likely to be increasingly important in native diets, especially in cities and towns. The ease with which manioc meal can be transported, as compared with yams and other roots, gives manioc a clear economic advantage in the provision of cities. And its low production cost will make the market price of manioc meal less than the price of cereals." Jones, Manioc in7Africa, op. cit., pp. 183-4. - 36 IV. THE TROPICAL ROOT CROPS AS LIVESTOCK FEED 4.01 It has been noted that the use of the TRCs animal feeds, although a relatively small portion of total utilization, is expanding rapidly. The most notable gains have been made by cassava and the most visible increase has been in the European Common Market countrieso The use of both cassava roots and foliage as animal feeds in other parts of the world, particularly in the producing countries, also is increasing. This section examines the factors which determine the use of the TRCs as animal feed and assesses this iotential market in the developed countries. Most of the analysis pertains Lo cassava, the TRC for which both production and export prospects in the developing countries seems to be the most promising. Nutritional Considerations 4.02 The great attraction of the TRCs as animal feedstuffs lies in their ability to accumulate energy at a high rate, with cassava the most notable in this regard. It is bulky, high in easily-assimilable energy and produces high yields per unit area of land and is increasingly recognized as an attrac- tive substitute in livestock feeding to traditional energy sources like maize. But the roots of cassava, the principal source of feed nutrients, are as poor in protein as they are rich in energy. The crude protein content of whole cassava root averages no more than 4% of dry weight. Moreover, 4-60% of the nitrogen content is non-protein nitrogen. Thus, when cassava root products are used to feed animals, they must be combined with protein-rich foods to balance the ration. Table 7 compares the nutritive value of different cassava products with sorghum and maize. The low protein and fiber and high content of soluble carbohydrates are notable features of the cassava root. 4.03 But cassava root products are only part of the plant utilizeable as animal feed. The principal parts of the mature cassava plant exploitable for feed, expressed as a percentage of total plant weight, are: leaves 6%; stems 44%; and roots 50% (composed of 11% water, 8% peelings, and 31% starch). The crude protein content of the leaves is 20-35%, while that of the stems is perhaps half this amount. The quality of the leaf protein is generally good, although deficient in the amino acid methionine. Total protein yield from the leaves is much higher than that of tropical legumes. 1/ Cassava leaves are low in crude fiber and relatively high in calcium and phosphorouso Sun-dryir2 of leaves is easy and dry leaves may be preserved as leaf meal or pellets. The aerial parts of the cassava plant also offer potential for recovery of leaf protein, either for human consumption or for mono-gastric animals (poultry, swine). Research on this technology is being carried out at IITA and CIAT and in other countries. The TRCs as Energy Sources in Compound Feedstuffs 4.04 Imports. TRC products, dominated by cassava, have become important raw materials in the manufacture of compound feedstuffs in the EEC countries 1/ C. Devendra, "Cassava as a Feed Source for Ruminants" in Cassava as Animal Feed, IDRC, 095e, p. 124. - 37 - !anTLI 3AT3$ f _n S _ In 4 0% CO 0D 0 _ - 0% n Cr N . _4 -4 .4 N ZU.TeATnba 4piX1s 4 c -4 4 0% 0o u0 4 qlr 1 0 X< ao . _ I0 90 0 0 0 rD- G }rFaatzv o o1 1 -4-4 !: X t oqlaJ * .OG Z IXi " > la oe -aU a !; n Z o .^ _ 02Z ; o g ,@,, ~~~~jaqiw 19° I 8 Q a e ^ F~~~~~~~~~~ a S . *>xa sa pa X o o o o " 4 0 ^ o~~~~~~~~~c -4 -4 ow . L;. z *D Z I0 .4 _ o.4 0~~~~~~~~~~ co r4 co c Or r :1 bC~~~~~~~~~4 0 ~~~~~0 0 10 0 0 %0 a ~ zuzaII~mZ 0~ 4. .0 -0 ~~ - - ~~~~~~0 0 0 0 e4n0 A c.o -4 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~4 ~4. 0% 0 0 U ~~~~~~~~~~~0 o.anls.o - 38 - and, more recently, Japan. Imports of cassava products into the EEC coun- tries grew by 15% per annum in 1971-75, with the Netherlands, West Germany, Belgium/Luxembourg and France the dominant markets. Import growth has been rapid because of the sharp increase in EEC animal feeding since 1960 and the Community's Common Agricultural Policy which maintains high prices for alternative energy sources (cereals)qX Cassava is now imported largely in the more concentrated pelleted form to reduce shipping costs and facilitate handling. Thialand and Indonesia are the overwhelmingly important export sources, accounting for 85 and 14%, respectively, of total EEC imports in 1975. Virtually all trade in cassava pellets is handled by six European- based shipping companies, with one handling nearly half of the total. 1/ The major traders operate pelletizing factories in Thailand and Indonesia0 4.05 Because freight charges account for about 20% of the COIOFO value of cassava products, competitiveness in compound feed mixes depends to a significant degree on shipping economies0 Economies of scale are important0 In April 1976 freight costs, ex-warehouse, Bangkok to Rotterdam, were about DM55 per long ton for cargoes of 80,000 tons, but DM 67 per ton (22% greater) for smaller loads of 15,000 tons0 Some 80% of exports from Thailand are now shipped in consignments of up to 90,000 tons0 Shipments from Indonesia are mainly small consignments of about 12,000 tons, a factor which places exports from that country at some disadvantage vis-a-vis Thai exports0 4.06 Qualit;y Compound feed mills of Western Europe prefer a quality cassava product graded to specification and are becoming increasingly de- manding in this regard0 Quality standards exist for pellets, based on starch, moisture, fibre and sand content, although these are not always followed by processors and exporters0 Levels of bacterial and fungal content in cassava products are not governed by official standards, however, these matters also are of growing concern to importers0 Analyses of pellet quality are done at point of export and in Europe. 2/ 4.07 European specifications differ slightly, requiring 65% starch for Grade A pellets0 Imports into Belgium must meet more rigorous standards, with a maximum mineral and raw fiber content of 4.5% and 6%, dry matter basis0 Maximum permissible hydrocyanic acid content in cassava products used in EEC compound feed manufacturing is 100 mg/kg (ppm)0 Limits to pesticide content also are specified, although this adulterant is not usually prob- lematic because little is used in most cassava production technology0 1/ Krohn and Company, Hamburg0 2/ The Thai specification for Grade A l"native" (ioeo, produced with locally= made machinery) pellets are: Starch (minimum) 62% Raw fibre (maximum) 5% Sand (maximum) 3% Moisture (maximum) 14%-14.3% (depending on month) - 39 - 4.08 The role of price. Cassava products are competitive in the EEC compound feed industry only because of the Community's Common Agricultural Policy (CAP) and its regulations governing cassava imports. The CAP for cereals is designed essentially to safeguard EEC income objectives for the Community's farmers. Under this policy EEC cereal prices are set at levels usually resulting in threshold prices which exceed world prices. 1/ A variable levy, equal to the difference between the C.I.F. Rotterdam price and the threshold price, is charged on cereal imports. 4.09 Under EEC and GATT regulations, imports of cassava pellets are sub- ject to a levy equivalent to 18% of the levy on barley or 6% of the value of pellets, whichever is lower. This keeps the price of pellets low in relation to cereals. Levies on cassava meal are higher, a factor which in addition to higher shipping and handling costs, has placed cassava meal at a disadvantage vis-a-vis pellets in the EEC market. Cassava pellets containing more than 3% by weight of binding agents are considered a compound product and thereby subject to higher levies. EEC imports of cassava from ACP countries are small, although imports from these countries enjoy some preferences in the Community. 4.10 The extent of cassava use in compound feed manufacture depends in part on the type of ration being formulated (with higher proportions of cassava in swine rations than, e.g., poultry feed), but also on the price relationships between cereals, other cereal substitutes and the major pro- tein source (usually soybean meal). Among the cereal substitutes of some importance in the EEC are sweet potato slices from China and locally-produced beet and citrus pulp. 2/ As a general rule, if cereal and soybean meal 1/ The CAP for cereals relates to target, intervention and threshold prices. The target price is the price EEC wishes to see averaged for the most important grains on the Duisberg market in Germany (the largest deficit area for cereals in the EEC). The intervention price, some 5-10% below the target price, is the price at which EEC agencies buy grain. It therefore constitutes the floor price for the market. The threshold price, established between the intervention and target prices, ensures that important cereals do not enter the market below the Duisberg target price. In January 1978 the CAP was extended to all nine EEC member countries. 2/ In 1975 sweet potato imports into the four major EEC feed-importing countries (Belgium, West Germany, France, Netherlands) totalled 114,000 tons. While of good quality, sweet potato chips are more difficult to handle than cassava pellets and supplies are irregular. A more impor- tant competitor is beet and citrus pulp. Citrus pulp has 6-7% protein, but starch content of only 38%, with relatively high fiber content. It is substituted for cassava in cattle feeds if its price is at least 10% lower than that of cassava. 40 prices to the compounder are about equal, the price of cassava must be some $25/ton less before maximum use of it can be made as a cereal substitute. 1/ 4.11 Cassava prices follow a regular seasonal pattern, rising in August and September with the onset of the heavy rains in Thailand and the subse- quent decline in shipments. Over time, cassava prices in the EEC have remained about 20% below cereal prices, with a secular increase in the 1971- 1976 period. Annex 4, Table 6, shows average monthly prices for Thai native cassava pellets in 1971-76. 4.12 Future growth of demand for cassava in the EECo Imports of cassava products into the EEC may approach 5 million tons in 1978D a figure which im- plies average annual growth in excess of 20% in 1975-78 and total market expansion by more than 107 million tons during the period0 2/ Prospects for imports after 1978 are uncertain and probably imply growth rates considerably lower than those of the recent past0 Major determinants of future grovth include the rate of expansion of the compound feed industry in the EEC and price relationships between cassava and other energy sources0 The price factors are determined by both political and economic considerations0 An analysis by Phillips, building on previous work, emphasizes the sensitivity of future cassava demand to price relationships with alternative nutrient sources, but confirms the generally attractive prospects for continued expan- sion of cassava use in the EEC0 3/ In this analysis the market for cassava products in the major EEC countries is projected to expand from 3.2 million tons in 1976 to about 7.6 million in 1990, or at an annual rate of approxi- mately 6.4%. 1/ An example may be useful0 In early 1976 cassava pellets were being quoted CoIoFo compounder, West Germany, in the range of DM 330-340/ton0 Comparable prices for barley and soybean meal were DM 450 and DM 420D respectively, or some DM 110 (about US$45) higher than the price of cassava pellets0 With the above rule of thumb, cassava pellets were very attractively priced0 Not surprisingly, record shipments of pellets were made to Europe in early 19760 2/ UNCTAD/GATT, Cassava: Export Potential and Market Requirement (Geneva1, 1977), po 40o 3/ Truman P0 Phillips, Tropical Roots and Tubers: Future Markets, Supplies and Uses (draft), University of Guelph, Ontario, February 1978o - 41 - Table 8 PROJECTED MARKETS FOR CASSAVA PRODUCTS IN SELECTED EEC COUNTRIES 1976 1980 1985 1990 (actual) Federal Republic of Germany 0.6 0.7 0.8 France 1.0 1.3 1.5 Italy 0.5 0.6 0.7 Belgium 0.7 0.8 0.9 Netherlands 2.4 2.8 3.2 United Kingdom _ 0.5 0.5 0.5 Total 3.2 5.7 6.7 7.6 SOURCE: Phillips, Tropical Roots and Tubers: Future Markets, Supplies and Uses (draft), February 1978, p. 48. 4.13 Several basic assumptions underlie the analysis. The 1971-75 growth rate in production of compound feeds is projected to continue through 1990 and the future production "mix" of compound feed (proportions of total production for poultry, swine and cattle) is assumed unchanged from 1973-75. (Annex 5, Table 7. Projected Production of Compound Feed in Selected EEC Countries.) Other basic assumptions are: (i) the proportion of cassava is determined by a linear programming exercise which assumes a cassava price at the feed mill of US$160/ton in the Netherlands and $165 in other West European markets (reflecting differences in transport costs and cassava quality); and (ii) only 60% of compound feed production in Germany is a potential user of cassava pellets as feed compounders in the south, account- ing for 40% of the country's output, have cheaper energy alternatives. 4.14 Political factors add uncertainty to these projections. One consideration relates to potential cassava supplies. It has been reported that the Thai Ministry of Agriculture is attempting to discourage the expan- sion of land devoted to cassava in response to the view that cassava produc- tion degrades the soil. 1/ Encroachment by cassava cultivators on the country's forest reserves also may be a factor. The impact of these factors on future cassava supplies from Thailand will be determined largely by the effectiveness of any efforts to control plantings. Additional land suitable for cassava is available. Moreover, it seems likely that Thai farmers will increasingly utilize new high-yielding cassava varieties and improved cul- tivation practices which permit continued growth in output from already- cropped lands. Annex 4 suggests that intercropping of cassava with legumes and foodcrops in Thailand is feasible and may be the basis for sound long-term cropping systems involving this crop. 1/ Average national yields of cassava in Thailand declined slightly since 1960. The annual production growth rate of more than 12% has been the result of steady increases in planted area. - 42 - 4.15 A second political consideration is the request of the French government that the EEC apply import levies to all feedgrain substitutes, of which cassava constitutes some 45%. Whether other EEC governments would support such a request is open to question because of their greater reliance on imported energy sources for feed manufacture, Moreover, because roughly half of the feedgrain substitutes come from developing countries, introduc- tion of additional trade barriers might be opposed by the larger international community. And, while market prospects in the EEC are somewhat uncertain, growing livestock industries in other developed countries such as the USSR and Japan may provide additional markets for low-cost energy in the manufacture of compound feeds. The TRCs as Feedstuffs in Developing Countries 4.16 Less is known about the use of TRCs as feedstuffs in developing countries, although this is substantial in some countries and appears to have great potential as poultry and livestock industries are expanded. The reasons for optimism are both technical and economic. The importance of, e.g., cassava in tropical livestock nutrition arises from the fact that the usual deficiency of energy in typical livestock rations reflects the shortage of soluble carbohydrates when compared with rations in more temperate envi- ronments. 1/ T:opicaL forages typically are more fibrous, bulky and less palatable than c'ieir temperate counterparts. 2/ In these circumstances, animal nutritionists in the tropics face the difficult task of formulating livestock rations with adequate levels of available energy0 4.17 Usually quoted stdtistics on the feed energy yields of tropical feedstuffs frequently are misleading because they tend to classify photosyn- thetic output in terms of gross energy, giving fibers, starches and sugars equal value0 In practical terms, actual nutritive values of these substances differ greatly0 In comparison cassava products such as root meal are distin- guished by their high digestibility and low crude fiber content0 (Annex 5, Table 8. Percentage of Structural Carbohydrates on a Dry Matter Basis of Some Tropical Feeds tuffs and Forages)0 This high degree of digestibility is the 1/ As noted by M4iller, the tropical biosphere favors photosynthetic out- puts in the form of structural carbohydrates (cellulose, lignin, etc.), rather than soluble carbohydrates (starches and sugars)0 The economic feasibility of converting, e.g., cellulose, to more nutritionally-usable forms, e.g., glucose, still presents a challenge to scientific research0 Cf. "Improving the Quality of Cassava Root and Leaf Product Technology" in Cassava as Animal Feed, opo cit., pp. 120-126. 2/ One study showed that at maturity 96% of all temperate grasses (760 samples) exhibited a total digestible nutrient (TDN) value over 55% whereas only 48% of the tropical grasses (from 312 samples) exhibited a TDN value of over 55%. Butterworth, WO H., "The Digestibility of Tropical Grasses0" Nutr0 Abst. Rev0, 37 (2), 349-369 (1967). - 43 - main reason why the supplementation of tropical forages by cassava and other sources of soluble carbohydrates have such a marked positive impact on livestock performance compared with other tropical feedstuffs such as rice bran, copra cake or palm kernel meal. 4.18 Much of the recent research work on the technology and economics of whole plant utilization of cassava in livestock feeding has been done in Venezuela. 1/ In 1977, costs of producing a ton of dried cassava foliage, containing 182 kg of crude protein and 435 kg of carbohydrates, were less than US$50 in experimental conditions. On a protein equivalent basis, this was about one-fifth the cost of dried alfalfa leaf meal in that country. With this large cost differential, supplementation of thecassava protein with methionine-rich proteins is economically possible. 4.19 Jones and Akinrele have outlined some of the potential of feeding the TRCs to livestock in Nigeria. 2/ The Nigerian cattle herd exceeds 10 mil- lion head, with annual off-take of 7-10%. Some 300,000 head of live animals were imported annually from neighboring countries during the early 1970s. The Federal Ministry of Agriculture estimates that effective demand for meat during the 1970s will grow by at least 3.4% annually, equivalent to 12-15,000 tons liveweight. If 3 kg of cassava starch, appropriately supplemented, are assumed to produce 1 kg liveweight, some 150,000 tons of fresh cassava roots would be required to meet the annual increase in meat consumption. If cassava were used to produce the last 50 kg of weight gain of the approximately three-quarters of a million animals slaughtered annually, some 500,000 tons of roots would be required. If this were fed to produce an additional 50 kg per animal slaughtered, the number of animal imports could be reduced by up to half. Supplementary feeding is increasing in the country, but almost all of this is based on cereal-oriented compound feeds and by-products. The local availability of cassava and groundnut cake would seem to offer attractive possibilities as substitutes for maize in livestock and poultry feeding. Some Conclusions 4.20 Much of what is known regarding the use of cassava in animal feed- ing has been summarized by a recent workshop on this topic: 3/ - Cassava products can, to a large degree, be used suc- cessfully as substitutes for certain cereals in nutri- tionally-balanced rations for different species of 1/ A. Montaldo, "Whole Plant Utilization of Cassava for Animal Feed," Cassava as Animal Feed, op. cit., pp. 95-106. 2/ W.O. Jones and A.I. Akinrele. "Improvement of Cassava Processing and Marketing Recommendations and Report" National Accelerated Food Production Program, prepared for the International Institute of Tropical Agriculture, March 1976. 3/ Ibid., pp. 127-130. 44 ° livestock in both tropical and temperate climates0 It is, however, difficult to define precisely the optimum levels of cassava use under particular circumstances. In poultry feeding, incorporation of high levels of cassava into rations appears to be more acceptable for broilers than for layers where egg production and quality may be affected by nutritional imbalances asso- ciated with rations high in cassava. In the case of rations for swine, it appears possible to practice life- cycle feeding using high levels of cassava, provided the ration is supplemented with appropriate nutrients. The amino acid methionine appears to be particularly important as a supplement because of its role as an essential amino acid and its function in the body processes of detoxifica- tion of the hydrocyanic acid componento (See Annex 1)o The physical properties of a cassava-based ration are important because they affect both palatability and feed consumption0 With both poultry and swine, feed intake and growth rate are increased through pelletization, al- though these benefits can also be achieved by adding molasses, fat or water to eliminate the powdery charac- teristics of the cassava-based rationo Recent agronomic work suggests distinct possibilities in treating cassava as two different crops, with the roots rich in energy and the foliage rich in protein and pig- ments. It also appears that the same plant can be used to produce several cuttings of leaves prior to harvesting of the roots, but at the cost of some reduction in root yield0 The prospects for using both cassava leaves and roots in the ruminant rations (cattle, sheep, goats, horses) appear to be promising, but more information is needed on the feeding value of both products0 Little is known about the digestion and utilization of cassava in the rumen, although cassava starch is known to be a good vehicle for promoting microbial synthesis of non-protein nitrogen compounds such as urea0 This combination of cassava and urea appears promising as a low-cost source of essential amino acids0 The potential for using cassava roots and foliage in ruminant feeding in the tropics is largely unexploredo Preliminary work at CIAT has shown that by harvesting cassava first at 90 days, and 70-day intervals thereafter, it is possible to produce a product that has an overall foliage crude protein level of more than 20% on a dry matter basis0 In a one-year trial, total production per ha was 20 tons of dry matter and 4 tons of crude protein0 - 45 - In a ration in which 28% of the intake was cassava foliage (which provided 80% of the protein requirement), yearling steers gained 621 grams/day. In considering the use of cassava roots or foliage for animal feed, potential is mre clearly seen if output\is expressed in terms of tons of meat/milk per ha under cassava than in terms of crop yield. The use of cassava, or other TRCs, as a substrate for the production of single-cell protein for use as an animal feed has considerable potential, although the economics of the process remain questionable. (See Chapter VI below.) 46 V. TROPICAL ROOT CROPS AND THE STARCH KA=(ET 5001 Most of the world's starch supplies are derived from either grains (corn, sorghum, wheat, rice), the major root crops (potato, sweet potato, cassava, arrowroot) or the pith of the eago palm. While stercheSo fom these various plant sources vary slightly in their physical and chemic&l proper- ties, they can be substituted for each other across a wide spectrum of end uses. Cassava starch must compete with other sterches and relative prices, quality and dependability of supplies are basic considerations in the deter- mination of market shares. Starch usage is conveniently divided into two basic markets: food products and industrial demand. In the fooe market, starches are used in the preparation of sauces, baby food, confectionery and bakery products, while in industry, starches are used in paper and textile manufacture, production of adhesives and in other products such zs pharma ceuticals, explosives and building materialso 5.02 The United States and Japan are the world's largest markets for starch, but in both countries mere than 90% of supplies are produced locally from corn, potatoes and sweet potatoes0 Starch imports in each of the two countries in recent years were in the range of 100,000 tons annually, with cassava starch accounting for more than three-quarters of the total0 Total starch imports into the United States and Japan of some US$30 million in 1975 can be compared with the value of cassava chip exports to the major EEC importers of about US$275 that year0 5.03 In the United States, importS of cassava starch declined almost steadily in recent years, reflecting in part the sharp increase in ocean freight costs and the reduced price and quality competitivenecs with local supplies0 Also contributing to the decline has been the erratic nature of cassava starch supplies and the increased availability in the United States of specialized corn starch modified to meet particular end uses0 However, in some uses such as the food industry, cassava starch is preferred and seems likely to be used0 In the United States starch imports of root crops origin are likely to continue to decline through 1990o Prospects for imports of TRC starch into Japan may be somewhat brighter because of proximity to major sources of supply (Thailand), although the Japanese goverEiment policy of protecting the local starch industry suggests that future growth of imports will be limited. - 47 - Table 9 IMPORTS OF STARCH INTO THE UNITED STATES AND JAPAN, 1970-75 (thousand tons) 1970 1971 1972 1973 1974 1975 United States Cassava 207 182 142 108 164 85 Potato 3 5 13 37 27 14 Arrowroot and Sago 3 3 1 1 5 3 Starches, n.e.s. 4 3 10 4 3 1 Subtotal 217 193 166 150 199 103 Japan Cassava 50 47 51 72 140 71 Sago 21 17 13 15 16 17 Starches - 1/ 1 4 11 8 10 Subtotal 71 65 68 98 164 98 Grand Total 288 258 234 248 363 201 1/ Less than 1,000 tons. SOURCE: UNCTAD/GATT, Cassava: Export Potential and Market Requirement (Geneva, 1977), pp. 44-45. 5.04 There appears to be an unexploited potential for TRC starch imports among developing countries, particularly those with large textile industries. For example, in 1974 Brazil imported US$1.2 million of starch products from the United States even though Brazil produces nearly one-third of the world's cassava supplies and manufactures cassava starch. Indonesia is another example. In the first nine months of 1976, Thailand exported 59,000 tons of cassava starch to that country. 5.05 The major cassava and sago starch exporters are Thailand, Brazil, Malaysia, and Togo, with Thailand accounting for about half of the total. The Thai starch industry consists of about 60 small mills with unit capaci- ties of 2-3 tons of starch per day and a similar number of modern mills each capable of producing 30-60 tons daily. A few larger plants have a daily capacity of 100-150 tons. Total annual capacity of the Thai starch industry is in the neighborhood of 800,000 tons, of which 700,000 is accounted for by modern mills. Exports of cassava starch from Thailand average about 150,000 tons annually while domestic consumption totals 120,000 tons. This excess capacity was built up largely in 1974 in response to increased demand from Japan and, to a much smaller extent, Indonesia. ° 48 - 5.06 Given the limited prospects for increased exports to major markets in Japan and the United States, the cassava starch industry in Thailand is likely to continue to operate with substantial excess capacity. Since the price of cassava roots is normally governed by the pellet industry, only limited possibilities exist to reduce starch prices in order to make exports more competitive. At the same time, however, the excess capacity in the Thai starch industry places that country in a good position to meet the require- ments of any future new markets elsewhere. In sum, the possible export of TRC starches does not seem attractive to other developing countries which do not have established positions in the market. 5.07 In several countries the traditional starch industry is an impor- tant source of starch for local users and provides a readily-accessible market for TRC production from small-scale farmers. A significant amount of rural non-farm employment also is generated by the industry0 Factories typically are small scale (one ton of raw root throughput per hour), are equipped with locally-fabricated machinery and use crude sedimentation processes which result in a product of variable qualityo This local industry frequently finds it difficult to compete with large-scale, semi-automated factories (throughput of up to 20 tons per hour) or with factories using grain, sometimes imported at low prices, as the raw materialo Producers of such equipment consider it possible to design reasonably efficient smaller- scale facilities (3-5 tons per hour) which utilize the centrifugal processes generally considered necessary to produce a high-quality product0 Where small-farm development programs are underway and non-farm rural employment is accorded some priority as a development objective, measures to improve the efficiency, and contribute to the survival, of the local industry should be considered0 These would include support of efforts to develop medium-scale technology as well as technical and financial assistance to the local industry. Table 9, Annex 5, indicates the capital structure of a small-scale starch factory in Colombia0 - 49 - VI. OTHER INDUSTRIAL PROCESSING OF TROPICAL ROOTS CROPS 1/ 6.01 It has been noted that additional large yield increases seem possible from most TRCs because of the present early stage of plant breeding and selection work on these crops. Cassava is perhaps the most attractive in this regard since it is already among the highest known yielders of carbohydrates and crop biomass. And unlike other high yielding crops with high fiber contents, the major component of the cassava biomass is starch, a compound which is much easier to process than cellulose. The aerial parts of the cassava plant also have considerable unrealized economic value. This section examines some of the work on technical processes to enhance the economic value of the cassava plant. All of these processes (see Figure 1) are technically possible, but the economic viability of some remain to be demongtrated. Success in this regard will depend heavily on the extent to which cassava yields can be increased and unit costs of energy from this crop lowered relative to costs of alternative energy sources. Two of the most interesting of these technologies relate to the production of single-cell protein and alcohol, processes for which economic potential has been enhanced within the last five years by the sharp relative rise in fossil fuel prices. Similar processes could be employed with other TRCs (and similar products emerge), although the higher energy-producing capability of cassava suggests the economics of this crop as a raw material would be more attractive. Production of Leaf Protein 6.02 It has been noted that cassava leaves are rich in protein, the B vitamins and vitamin C. Only limited efforts have been made to utilize these nutrients, but possibilities appear to be good. A leaf protein con- centrate can be extracted from cassava tops using relatively simple equipment such as a screw press or a belt press-. A belt press system has been developed for use at the village level. 2/ A hammer mill reduces the tops to a fibrous pulp which is squeezed over a perforated or grooved cylinder. The joice is then coagulated by live steam and the coagulum is filtered, washed and dried to a leaf protein concentrate. Further treatment can improve palatability. The potential of leave protein concentrate from cassava is sufficiently great to warrant an expanded research and development effort. Starch 6.03 The previous section concluded that export starch markets were not very promising for TRC-producing countries without established market posi- tions. Vigorous research and development work by TRC starch producers might 1/ This section is based on D.J. McCann, "Cassava Utilization in Agro- Industrial Systems," in Proceedings of the Fourth Symposium . . . op. cit., pp. 215-221. 2/ N.W. Pirie, "Leaf Protein: Its Agronomy, Preparation, Quality and Use," Blackwell Scientific Publications, Oxford, 1971. FUlunE ' AN AGRO-fl*DUTRlAL SYSTEM -OR CASAVA _AMM~ =i~~ ProySWd LEAt' PROTCIN CATRACTION ~~~~~~~~~~~~~~~~~~~~~~~~~I uO ra.duo STEAM CO;Y.USTIO N ANO TOP1 u wuo X PCWERPRNCNT'AIO a r° Ctn ° O wcso2=O IUC7 CASSAVA=° _TOSI TO SOIL D C,MVOPATIOW4| A4SLPE _~~~~~~~~ULTN ]== =a tvi00QEDIEw .~~~~~~~~APNN = _ STARCHt fraro rc3'uo tcnciact ocD) ; 4VDi ILVt318 1 fic'ntoco cmr!s GL m- - [ - - rrwcaas#rota AP4D/tn ot;~vf c!ic1G F:Ct EQ1CP4TATION P crcotono biumDn ciri =1 ote AND/OR AND/OR EEQci4TATION > TO SIPGL> SCP CELL PROTEIN Sourco DJ. i.:Crnn. "Ccc U 3Ls)tieoa ,n A.nAo.InAbtr.oI SVotc7ar b n Proococrnes of tho iourth Sv.t'ro.um of tho Imornotencl S2dcizv Cr Tropicof Pait Cropa, CIAT. Col, Colombo, Autjn 1-7, 107G: p. 21Gi. -ctlGn -ioc94 - 51 - improve these prospects, building on the unique properties of the T1.C starches. Cassava starch, for example, has particular properties (low amylose content) which makes it uniquely suited for some food preparations. Simple modifica- tion of cassava starch by cross bonding or use of corn starch/cassa va blends give properties ideally suited for use in a wide variety of conveni ence foods. This suggests that some inroads could be made into, e.g., 1the corn starch market if modern cassava starch processing plants are established and high-yielding cassava varieties are grown to ensure low-cost raw materials. But first priority probably should be accorded to efforts to exparid the local TRC starch and flour industries and increase domestic utilization of this output. 1/ The manufacture of composite and fortified flours, us Ing TRC components, is attractive in this regard. Fermentation Processes 6.04 Fermentation processes based on agricultural raw materials have been used for many years to produce such derivatives as butanol, alcohol and acetone. But with the post-World War II development of a petro-chemical industry using low-priced fossil fuels, many of these organical:Ly-based processes become economically non-viable. Under the stimulus oE sharply higher relative prices for petroleum in recent years, fermentation technology has again been emphasized. Substantial improvements have occurred, largely as a result of the new continuous single-cell protein processes; based on petroleum fractions. These technologies, which could use high,-yielding cassava as the basic raw material, may provide the means for the production of several valuable products. 6.05 Sugars. At present, much of the glucose sugar (dextrose) is made through hydrolysis 2/ of corn or wheat starch. From a technical viewpoint, 1/ Cassava starch mills which are largely self-sufficient -Ln energy are possible. As McCann points out, a starch factory requires about 75 kW of peak electric power for each ton/hour of tuber throtighput. Assuming an 80% load factor for motors and a 25% electricity gerierating effi- ciency, this implies that some 865 MJ of primary fuel aire required per ton of tubers processed. To this must be added a steaun requirement for drying of another 800 MJ per ton of tubers. Each ton of tubers gen- erates 40 kg of dry solid as peels and 50 kg of pulp waste. If these two waste streams are used to feed an anaerobic digester, it is possible to produce 1,100 MJ of methane gas, roughly equivalent, to two-thirds of the energy requirements for the whole process. The rvamaining one-third of energy needs can be provided from pulped cassava tops or from the fibrous residue left after extraction of leaf proteirn. Residue from the digester can be utilized in an aquaculture system invrolving algae, fish and ducks. Research in this area is underway in the Chemical Engineer- ing Department of Sydney University. Cf. D. J. McCaican, op. cit., p. 219. 2/ Chemical decomposition involving splitting of a chemical bond and addi- tion of the elements of water. - 52 - cassava stasrtch is equally feasible as a raw^J material, and may ba preferable because cat3sava's low protein content permits easier hydrolysis of the starcho The glucose, output is used primarily by the food industry as a metabolic energy source and sweetener, although being only 70% as sweet as sucrose (cane and beet sugar), it is not always an ideal "sugar" substitute. This limitation lies prompted the development of glucose/fructose mixtures which compete moret directly with sucrose. Use of the underlying process, based on corn starch, has expanded rapidly in the United States. 6.06 Bu;tanol and acetone0 1/ The fermentation of starch to produce butanol and iacetone predates World War I Introduction of the organism Clostridium aicetobutylicum is capable of producing a 30% yield (of starch dry weight) of mixed solvent containing butanol, acetone and ethyl alcohol in the ratio t5:3:1 Distillation then yields the pure solvents0 At world prices for thetse products prevailing in recent years, a process utilizing high-yielding cassava as a starch source appears to be financially attractive0 Additional basic experimental work is required to refine the technology in butanol fermenial:ion of cassava. 6.07 Single-cell Protein0 An important questlion in assessing the future role of the TRCs in agriculture and rural development is the possible contri- bution this group of crops could make to world protein requi.zemezts0 Per capita and futui-e global needs remain topics of debate among nutritionists and others, and it is clearly necessary to distinguish market deimand for protein from any additional needs to meet nutritional requiremer-ts0 This paper has argued (Chapter III) that in much of the world, the most critical nutritional requirement among low-income groups is additional calories0 But it must be recognized that sustained improvements in nutrition across a wide spectrum of the world's population will require protein supplies to grow more rapidly than population growth0 2/ There is doubt as to whether this increase can be realized fxom conventional protein sources and made available to those most in need at piices they can afford to pay0 6008 The likeY.-y deficit of protein supplies from conventional sources has encouraged resevarcI and development work to develop new sources of this l/ Butanol is a buttyl alcohol derived from butane0 Acetone is a volatile flamable liquid used chiefly as a solvent and in organic synthesis0 2/ To achieve a minimum average daily per capita intake of 70 grams of 100% protein, the FA.0 has projected the following protein needs: Avo Annual Growth (X) Popula- Pro- Popula- Pro- Popula- Pro- 2opula- Pro- tion tein tion te4.n tion tel-n tion tein Developed countries 100 38 102 45 1.3 47 103 lol Developing countries 2.5 53 3.6 84 4.1 98 2.5 3.1 Total 3 5 91 468 129) 5T4 145 202 2.4 *Population in billios; protein needs in million metric tons of 100% protein0 Source: Documents of the World Food Conference, 1974. - 53 - nutrient. 1/ In contrast to using large, slow-growing macro-organ:Lsms such as plants and animals for protein production, much of this work in receint years has focused on micro-organisms such as yeasts, bacteria, fungi, and algae as fermentation agents. The major advantages include rapid growtl and reproduction (perhaps 30 minutes as compared to a few months for crops and several years for large livestock), 'and the availability of a wide variety of potential feed-stocks ranging from petroleum fractions to agricul- tural wastes. These processes are technically proven, some of them haviilg been used in simpler form for many years. 2/ It is understood that prodtic- tion of single-cell protein for animal feed now amounts to more than 1 mi llion tons annually in the U.S.S.R. and several plants with annual capacities o f 100,000 tons are under construction elsewhere. The most popular feedsto6c1s in the large plants are certain petroleum derivatives or methanol. Substrate s used, or proposed for use, on a smaller scale include pulp-mill effluents, other agricultural wastes and cattle feed-lot manure. Staff of the World Bank have considered a pilot project using these processes to produce single--ce-El protein from waste bananas. 3/ 6.09 In the petroleum-based technology noted above, plant costs peir un:Lt of output fall sharply as plant size increases. A basic question from i:he viewpoint of this paper is whether smaller-scale plants, utilizing agriL:ul- tural materials, i.e., the TRCs, are technically--and economically--fea.sible. Promising research is underway at CIAT on this matter, although a defin3Lte answer may not be available until sometime in the future. A pilot plant: to produce SCP from a cassava substrate was built at CIAT during 1976 and I)egan operation in 1977. 4/ Some of the equipment was built in Colombia. Th(! 1/ This material is based upon the work of Mr. Christopher Pratt of the World Bank Staff. 2/ Traditional fermentation of cassava in Indonesia produces "tapeh" through a process which raises protein content from 1-2% to 8-10%. This could go beyond 15%, though with sacrifice of 30-40% of calorie!s. Effect on protein quality is apparently positive. 3/ Report by C. J. Pratt, "Proposed Pilot Project for Animal Feed Bio- protein by ESPOL-Ecuador Based on Waste Banana Substrate", June 6, 1977. This proposal indicates the possibility of installing five simple bioprotein plants in Ecuador with a total cost of about US$15 million. Using microfungi as the conversion agents, this project could produce as much as 32,000 tons of SCP in Ecuador per year, at a cost believed competitive with alternative protein sources, i.e., soybean meal at a delivered price of US$300/ton. One metric ton of bioprotein, contain- ing some 50% of crude 100% protein, requires 11 tons of waste bananas. The report estimates that as much as 30% of total banana production is damaged in handling and is therefore potentially available as a substrate for this process. 4/ J. Santos N. and G. Gomez, "Pilot Plant for Single-Cell Protein Prodluc-- tion", in Cassava As Animal Feed, op. cit., pp. 91-94. 54 - plant is dE!singed to operate with a minimum of instrumentation and includes two fermentors (200- and 3,000-liter capacity)0 6010 The conversion process begins with either fresh cassava roots or cassava meal/flour0 Fresh whole roots are washed and then rasped to open the ce!ll walls and facilitate the suspension of\the starch granules in the fermeiitation medium. The rasped cassavan is then transferred to the ferE mentor, which is half filled with hot water (700 C)o More water and other neces sary ingredients (urea, monopotassium phosphate, sulfuric acid) are addecl to provide adequate nutrient supplies for optimal growth of the micro- orgaiiisms (Aspergillus fumigatus I-21A)o Fermentation is usually completed with:In 20 hours under constant temperatureo At the end of the fermentation peri cid, the biomass is harvested and can be fed fresh or sun/air dried to be Lncorporated into feed rations for animals. A proven product would seem equ.ally suitable for humanso 601 1 In early experimentation under these pilot plant conditions, the cru-de protein content of the final dried product was about 28%o This biomass was thien fed to growing rats to ascertain nutritive quality of the protein cc,nterit0 The effect of supplementation with the amino acid methionine was also t:ested0 Total weight gains over a 28 day experimental period were very rpoor for the diets based on the unsupplemented material0 Methonine supple- mentat:ion significantly improved protein quality of the fungal protein and result:ed in body weight gains similar to those obtained with casein and superior to those from soybean meal proteino 6.12 An important factor requiring further study is the possible health risks to individuals from continued exposure to spores of the micro-organism employed in the fermentation process0 Equally important is the need to examine the effects on animals to which the SCP is fed0 Extensive tissue anayE;is at the University of Guelph has not revealed any pathological changes in raLts fed the SCP produced by any of the three most promising cultureso UntiJL complete safety for personnel at the pilot plant can be assured, CIAT inte1idS to restrict fermentation experiments to the 200-liter fermentor. Regular medical examinations of research personnel have not revealed any health problem to dateo Despite these uncertainties, the process appears to have. considerable potential as a means to produce protein 'from a cassava sub- straLte, As MacLennan has shown in other experimental work using continuous fernmentation processes, when the yield of SCP is expressed in terms of land area, cassava, through this transformation, becomes a high protein cropo Because raw material costs make up at least 60% of total production costs, the:y are critical to determining the economic viability of these processo - 55 - Table 10 PROTEIN PRODUCTION DIRECTLY FROM CROPS AND IN SINGLE CELL PROTEIN PROCESSES Crop/System Crop Yield Protein Content Protein Yield (tons/ha) (%) (kg/ha) Rainfed grain 1.2 14 168 Irrigated grain 6.1 10 610 Irrigated soybeans 2.4 38 912 Irrigated grain/ single-cell protein 6.1 65-73 b/ 2,430 Rainfed cassava/ single-cell protein a/ 11.4 55 b/ 3,662 SOURCE: D.G. MacLennan, "Single-cell Protein", p. 27 (4), 13-17 (1974). a/ The cassava yield figure used by MacLennan of 10 tons/acre dry weight (24.2 tons/ha dry chips or 63.7 tons/ha fresh roots) has been adjusted to 30 tons/ha fresh roots. b/ This high protein content reported for a continuous fermentation process may be compared with the 28% realized in batch fermentation at CIAT. 6.13 Ethyl alcohol. Several of the TRCs and cassava in particular have been used at various times and places over more than 50 years as the basic raw material in the production of ethyl alcohol. 1/ Cassava starch can be 1/ Considerations of ecology and national security have long been recognized in the use of ethyl alcohol. H. Ricardo, writing in 1935, remarked: "The mobile internal combustion engine is now no longer a luxury and assurance of its fuel supply should be considered a matter of national importance. It is perfectly well-known that alcohol is an excellent fuel and there is little doubt that sufficient supplies could be produced within tropical regions. By the use of a fuel derived from vegetation, mankind is adapting the sun's heat to the development of motive power, as it becomes available from-day to day. By using mineral fuels, he is consuming a legacy--and a limited legacy at that-- of heat stored away many thousands of years ago. In the one case he is, as it were, living within his income. In the other he is squandering his capital." (Preface, The High-Speed Internal Combustion Engine; Blackie and Sons, Ltd.). - 56 - readily converted to alcohol in a simple chemical processo The starch is hydrolized to glucose by an enzyme or acid process , then concentrated and converted to alcohol by the action of a yeast. Purification of the product is done by distillation to yield either industrial alcohol (85%) or absolute alcohol which may be used as a solvent or fuel subsituteo It may be blended with gasoline up to 20% by volume without altering the carburetion systems of most gasoline engines. In the process, one ton of cassava roots produces in excess of 150 liters of ethyl alcohol. The product has a higher octane rating than regular unleaded gasolene (120 vs 90) and burns with fewer pollutants. 6.14 The potential of ethyl alcohol as an extender of petroleum-based fuels has created considerable interest in these processes based on agricul- tural raw materials. Much of this research and development work has been done in Brazil with active support by the Brazilian Government through the National Alcohol Program initiated in 1975. Installed capacity to produce ethyl alcohal in that country exceeds 9.5 million liters/day, much of it based on sugar cane as the raw material0 The use of cassava is growing because of the possibility of growing this crop over a much wider range of soil and rainfall conditions, although debate continues over the relative merits of this crop, sugarcane and other crops as raw materials in the production process0 1/ 1/ A comparison of the net energy ratios (output to input) of cassava and sugar cane as the raw material for alcohol production suggests a slightly greater ratio for cassava: Energy (10 k cal) a/ Raw Input Ratio (output Material Output Agric. Distillery Transp0 Total to input) Sugar Cane 5.59 0.42 0017 0.26 0.70 8.0 Cassava 5.59 0Q30 0.045 0.27 0o62 9O0 a/ Basis: 1 m of anhydrous ethanol. Total on-sight generation of electric power0 CTP Newsletter, Vol. 3, No. 1, February 1978 (Centro de Tecnologia Promon, Rio De Janeiro). A more recent study concluded that, from a net energy viewpoint, sugarcane is the most efficient crop for ethyl alcohol production among three crops examined (sugarcane, sweet sorghum, cassava), while cassava was least efficient0 However, all three crops produced positive net energy balances0 It was also noted that cassava has some advantages in that it grows over a wider range of agro-climatic conditions and may be the most susceptible of the three to further yield increases through research0 J de Silva, G. Serra, JO Moreira, J. Concalves, J. Goldberg, "Energy Balance for Ethyl Alcohol Production from Crops," Science, Vol0 201, 8 September 1978 (ppo 903-906). - 57 - 6.15 In December 1977, PETROBRAS, the Brazilian government oil monopoly, started up the world's first large-scale plant to produce ethyl alcohol from cassava. The plant, with full capacity of 60,000 liters/day, produced some 30,000 liters of alcohol to specification during its first test runs. The cassava is produced in plantation systems. Early runs have been hampered by the inadequacy of raw root supplies. Also problematic has been the relatively high price of raw roots compared with the government-established price of alcohol. The major output for the product is used as automotive fuel, while the by-product carbon dioxide is recovered and sold commercially. Stillage will be used to generate methane and possibly converted to SCP. The research organization which developed the technology in the PETROBRAS plant has subse- quently designed a pilot plant to permit alcohol production from cassava through a continuous fermentation process which is expected to show further improvement in processing efficiency and economy. 6.16 In late 1977, technical studies in Brasil estimated the costs of producing ethyl alcohol from cassava at US$338/m , assuming known agricultural and industrial technologies and the Sfficial US dollar-cruzeiro exchange rate. This cost compares with the US$290/m fixed price for alchohol established by the Brazilian government. With the cost of the agricultural feedstock (cassava roots) accounting for more than 50% of total production costs of alcohol, the economic viability of the process is sensitive to the level of cassava yields. The Brazilians estimate that a doubling of cassava yields from the assumed level of 14.5 tons/ha would result in a 25% reduction in costs pe unit of alcohol produced, or to something in the neighborhood of US$250/m . 1/ VII. THE TRCs AND PROJECT DESIGN 7.01 This paper has examined the TRCs as sources of human food, livestock feed and industrial raw materials and concluded that each use offers opportuni- ties around which development projects can be designed. It was noted, however, that some of the industrial end uses, while holding great potential, are not yet fully proven in economic terms and are therefore likely to be of only limited importance in the near term. Nevertheless, they are of sufficient promise to warrant priority for the research on both TRC production and utilization necessary to realize that potential. Because the most promising immediate opportunities for expanded use of the TRCs are as human food and livestock feed, they are of particular importance in poverty-oriented rural development activities. 7.02 It is now necessary to look more specifically at the roles which the TRCs might play in the Bank's future lending for agriculture, agro-industry and rural development and what policy actions are required to realize more 1/ V. Yang, W. Wilfont, A. Scigliano, C. Massa, S. Sresnewsky, S. Trindade (Centro de Tecnologia Promon, Rio de Janeiro), "Cassava Fuel Alcohol in Brazil." A paper presented at the International Symposium on Alcohol Fuel Technology, Wolfsburg, Germany, November 21-23, 1977. 58 - fully the potential of this group of crops0 1/ Rural development, as defined by the Bank, is the process of extending the benefits of development to the poorest--small-scale farmers, tenants, the landless-among those who seek livelihoods in rural areas. An appropriate strategy for rural development must reflect several factors: that numbers of the poor in rural areas will remain large, requiring low unit costs of development programs; that labor is their most abundant resource, although some may have land; that cash reserves are generally non-existent; and that improvement of their welfare requires the introduction of technology which permits greater returns from the resources under their control0 An appropriate strategy should also reflect the persist- ent calorie shortage among the rural poor and the fact that many of the poor of the future must seek livelihoods in difficult agro-climatic zones with lands of limited natural fertility. 7.03 The importance of the TRCs in the production systems of smallholders has a number of implications for rural development programs which include these crops0 2/ Given the small size of their operations, smallholders tend to be risk averse, face severe constraints on capital availability, usually depend heavily on family labor, generally consume a substantial part of production and have access only to poorly developed input and output markets0 These factors make development programs involving the TRCs costly for national governments as compared to efforts which focus on crops produced by larger farmers0 In such circumstances the cost or -xtension becomes a major compornent in raising farmer productiviLy, as docs an effort to imrrove market .nf_J!-s.:ucture for production lnpizs and outpu:. These considerations may encourage governments to design TRC industrialization projects (for starch, alcohol, livestock feed) around large-scale production schemes, an approach which reduces the risk of unde .tilizing pi-ocessing facilities, but which thereby foregoes the opportunity :. utilize a group of crops which, by virtue of their importance to smallholders nroughout the world, are uniquely suited to provide benefits related to both agricultural growth and increased income opportunities for the poor0 1/ As of mid-1978 the TRCs and potatoes had been included in the cropping systems of only a small number of Bank-supported projects, in most cases to provide staple calories for the farm family, or more rarely, as a raw material (cassava chips) for export to the livestock feed industry0 These include Bolivia/Ongavi Rural Development (potatoes, papalizas, ocas); Brazil/Ipiababa (cassava largely as a food item; some for sale); Nigeria/Lafia and Ayangba Agricultural Development (yams and cassava for human consumption)0 The root crops are included in several other projects: Brazil/Paraguacu Rural Development (cassava as cash crop for sale as pellets or for- alcohol manufacture); Tanzania/Mwanza-Shinyanga Rural Development (cassava as staple food); Indonesia/Transmigration II (cassava as staple Nood plus some production of chips for export); Indonesia/Jogjakarta Rural Development and Brazil/Tabuleiros do Sol (cassava as food)0 2/ See John K. Lynam, "Options for Latin American Countries in the Development of Integrated Cassava Production Programs," CIAT, February 1978, po 12f. - 59 - The TRCs as Food Crops in Small-farm Development 7.04 In poverty-oriented, small-farm development projects, a basic question is whac cropping systems are likely to both provide adequate income and establish the production mix for sound nutrition of the farm family. The solution would be easier if maximizing net income to the family would also ensure an adequate diet. Unfortunately, such is not always the case and a number of considerations must be balanced in reaching a decision. 7.05 When an improved cropping system is proposed, it is likely to compare favorably in terms of gross production with existing traditional systems. Thus, a proposed package yielding, say, 20 tons/ha/yr of a TRC and some legumes would probably dominate available alternatives in terms of gross value of production. An important consideration is that the TRCs, particularly cassava, will often grow in soil and climatic conditions where maize, sorghum, millet and other grains do poorly. But several problems emerge. With even 2-3 ha, the proposed system would result in perhaps 30 tons of fresh roots whereas maximum family consumption is almost certain to be less than 2 tons. 1/ While some of the output could be utilized by livestock, much of it would be sold if market outlets exist. However, many small-scale farmers are reluctant to use the small amounts of cash which may be generated to buy food, preferring instead to use it for the purchase of non-food essentials. In some situations, men control cash crop production (and income) while women take care of subsistence output. Production and spending priorities within the family in these cases may be very different. The local market for food items frequently is not well-developed in near- subsistence economies. And newly-commercialized farm families often have inadequate knowledge of how to purchase foods which are both economic and nutritious once traditional diets are left behind. 7.06 As experience with some rural development projects indicates, many small-scale farmers, having experienced the vagaries of markets for their cash crops, frequently prefer the security of producing their own food even if it means some sacrifice of income. In such cases it seems useful to establish a cropping system which provides a varied and popular assortment of crops for home consumption and only then allocating the balance of the holding to commercial crops. In many situations, an evolutionary process of cash crop development, based on proven technological packages, is preferable to a sharp shift to commercialized cropping which requires the purchase of basic foods. 7.07 Such an approach poses the question of what portion of the typical smallholding needs to be diverted to basic food production, and how much income will be foregone to accommodate this security-oriented cropping system. Fortunately, this proportion is likely to be small if root crops are included in the cropping program. If cassava yields of 10-20 tons/ha/yr are assumed 1/ A family of five adult equivalents consuming cassava at high levels of up to 1,000 calories daily would need only 1.8 tons of fresh roots per year. - 60 - along with, e.g., maize yields of 1-2 tons/ha, then aproximately one half ha in these two crops, inter-cropped with beans, would provide the family's calorie and protein needs. Diversion of one-half ha, or perhaps 10-30% of the cropland in a typical small-farm rural development project, seems reason- able as the basis for reduced risks and adequate calorie and protein supplies for the family., 7.08 Central America. Recent work on farming systems among small-scale farmers indicates some of these possibilities. In Central America CATIE 1/ has examined various small farm cropping systems utilizing, singly or in combination, the common bean, maize, cassava and sweet potato. To simulate small farm conditions, initial soil preparation was done with a pedestrian tractor, but all subsequent cultural operations were carried out manually. Basic cropping systems wre monoculture, intercropping, rotation and relay cropping. Results were examined in terms of calories produced and net returns per ha. As expected, systems which produced the most energy/ha/year included cassava and/or sweet potatoes. (See Annex 3 for details.) Of particular interest in the context of small-farm development was the relationship between the energy content of a particular system and its economic returns. 2/ Screen- ing of cropping systems was done on the basis of two nutritional considerations: (i) calorie production/ha should exceed 25 mega cal/year; and (ii) at least 5% of the total energy produced should originate from protein sources. 3/ The two systems which rank at the top in terms of net income/ha (1974/75 prices) also meet these nutritional requirements (beans and cassava intercropped; beans and cassava intercropped with maize in rotation). Moreover, these two systems also had among the lowest labor requirements per ha and the highest ratios of gross value to labor cost. The system of intercropped beans and cassava resulted in the highest net income of any of the combinations examineA. It also generated the third highest level of calories/ha/year and produced about 50 kg of crude protein/ha/year. 7.09 Nigeria. In Nigeria IITA has examined small-farm cropping systems which include the root crops and has developed a cassava production package for southeast Nigeria as part of the government's National Accelerated Food Production Project. While cassava is the doiminant crop in the region, the recommended practice was to intercrop with maize using improved planting 1/ Tropical Agricultural Research and Training Center, Turrialba, Costa Rica. 2/ In the Cost Rican context, perhaps typical of the humid tropics with relatively fertile upland soils, beans and sweet potato are the main sources of income, although production and price of the latter vary a great deal from year to year. Maize and cassava are the crops with the lowest production risk. 3/ A mega calorie is one million calories. It may be thought of as one individual's annual calorie requirement at 2,740 calories/day. The above criteria employed in screening are of course quite arbitrary. In practice, considerations of economics and risk are likely to be paramount in the farmer's choice of a cropping system. - 61 - materials, and some fertilizer and insecticides. This has proved to be more profitable than sole crop cassava and reduces the spread of bacterial blight. The yield of improved cassava intercropped with maize was three times that of local varieties, while maize yields were at least doubled. After allowing for increased costs, net returns/ha\were doubled compared to local cultivars and practices. 7.10 Indonesia. In Indonesia's Bank-supported Transmigration II project, a food cropping system is being considered for settlers in the new lands to provide subsistence food needs and a marketable surplus during the early years of settlement. Five-hectare plots are allocated to each family, of which 2 ha would be clean-cleared for intensive food cropping. In each of three basic models (food crops, coconuts or rubber as the main crops) cassava is included largely to provide a relatively secure low-cost source of basic calories from lands of limited suitability for traditional annual food cropping. In the food crop model the settler would receive 2 ha of clean-cleared land of which 1.3 ha would be planted to food crops, 0.45 ha for the family garden and 0.25 ha for the houselot. An additional 3 ha would be planted to rubber or other perennial crops. With 0.5 ha each of rice, corn and cassava, the fmaily should have some surplus available for sale. As an intercrop or rotation crop with beans, grain legumes and garden crops, cassava provides the basis upon which to build a nutritionally-balanced food productin system for the settler family. This initial emphasis on food cropping in settlement where perennial crops are expected to be the main cash crops offers the advantages of providing some food security to the settler family during early years while permitting the development of the perennial crops as family labor becomes available and technical support services are provided. It is assumed that chemical fertili- zers would be used in food cropping. 7.11 The TRCs and protein supplies. An earlier section of the paper noted the concern of nutritionists that increased dependence on TRCs may exacerbate protein deficiencies, particularly among the young. Several approaches to this problem are possible. The use of beans, chickpeas or other protein-rich vegetables intercropped with TRCs is possible, although with some effect on yields. In a CIAT trial, beans planted 6 weeks after cassava did not reduce cassava yields but cut bean yields from over 3 tons/ ha to just over 1.5 tons. The planting of beans six weeks before cassava does not appear to affect bean yields but reduces cassava production to about two-thirds of normal. 7.12 Sole crop cassava yields in this case were in the order of 30 tons/ha. A plausible yield trade-off with intercropping might be 20 tons/ha of cassava and one ton of beans or, say, 14 tons of cassava with two tons of beans. Since beans contain 20% protein of rather good quality versus 1-2% poor quality protein in cassava, such a 7:1 trade-off would favor beans if protein is desired. To provide each of five adult equivalents with 100 grams of protein per day (over 20 grams of protein) would require only 180 kg of beans per annum, or, at 1 ton per ha, 0.18 ha without any reduction in cassava yield. In other words, if just a few tenths of a ha are intercropped to favor cassava, all protein needs could be met, assuming grain provides most - 62 of the other calorie requirements. This appears to be aEong the most promio- ing means to ensure that both protein and calorie needs are met in awsll-fac situations involving the TRCso 7o13 If this general approach is to be widely used, there is need to inform project staff more fully of intercxlopping, mixed cropping or relay cropping possibilities as suggested by adaptive research efforto in local situations. It is not apparent that Bank staff are always familiar with this type of information and therefore may find it difficult to incorporate the most appropriate cropping systems in rural development programco This sugesto that improved communication among Bark staff on ouch tecinical w:ttersa as well as between the Bank staff and the research coEmunity, way be called forE 7.14 In favorable soil and cliwmtic conditionsc, the con umption of root crops other than cassava can supply much of the required proteinso Altheugh potatoes are deficient in some essential amino acids (those containing sulphur), these can be readily supplied by beans or groundnuts0 Consumptien of 1 kg of potatoes per day provides 820 calories and 20 graes cf protein0 If combined with 100 grams of beans or groundnuta, 40 grTams of W2ll0balned protein would be provided. This is adequate for Eost adults evcm if cmasoav supplies the balance of the dieto Dehydration of potatoes, nou bening Ee- searched by CIAT/IDRC with the use of solar-powered driers but practiced with cruder technology by the Andean population for many yearsD would reduce bulk and improve palatability of the potato-legume=criented diet0 7.15 Simple fermentation processes also can contribute to protein sup- plies0 It was noted that cassava starch has been used as a substrate for production of single-cell protein in traditional fermentation procedures in Indonesia. Variations of this process might be used more widely, and while amounts ingested daily are likely to be modest and not all groups would ptefer this products, it is an important possibility which seeme to deserve additional consideration0 The use of small-scale ferm2ntation procedurss in the production of SCP1, along lines being developed at CIATD also seems likely to represent important potential sources of good quality protein for both man and animal0 The young leaves of many of the ThCs may be consumed by humans and are a fair source of protein and other nutrients0 Commercial exports of cassava leaves have been made from Thailand, although these were apparently used in formulating livestock feeds0 Leaf harvesting machines are available0 7.16 Small-scale fish culture appears to be another way of utilizing the TRCs for protein production0 Several species of fish can be produced using various parts of the TRC plant, including leaves, stems or skinc. of roots or tubers0 This may be an efficient way to utilize protein and carbohydrates in these materials which are often discarded0 Further research and develop- ment is warranted0 TRC Production for Off-farm Sale 7.17 The technology factor0 One valuable use of the TRCs in the Bank0s rural lending is likely to be as directly-consumed, low-cost foo&s for hos - 63 - consumption in small-farm development projects. But demand for the TRCs in this form is limited by the perishability of the raw roots, as well as the capacity of the human stomach and the palatability of this food group. Alternative outlets must be found if TRC-based development programs are not to founder as a result of price declines from increased production in the presence of limited on-farm demand. This requires a broadly-oriented effort which fosters the development and adoption of improved technologies in both the production and utilization of the TRCs. Much of this technology is available or under development. There is opportunity and need to disseminate it more widely both within and among countries. 7.18 Technology to increase the consumption of TRCs by humans should aim to produce a low-cost, palatable product which is less perishable than the fresh root. Urban consumers would constitute the major market. Because traditional processing techniques in the making of, e.g., gari in West Africa or farinha de mandioca in Brazil are laborious and time-consuming, techniques which reduce labor requirements are likely to be needed if a convenient, low-cost product is to result. 7.19 Very useful work has been done in processing/utilization research in, among other countries, Nigeria and Brazil where much of the cassava is consumed as gari-type products. In Nigeria one thrust of the research effort has been to reduce the labor requirement in traditional gari-making, with the most successful innovation being the introduction of a powered root grater. 1/ Subsequent work is aimed at developing improved techniques for the cooking and drying processes. In several important TRC-producing countries improved technologies to make gari-type products are available at the village, intermediate and industrial level, but not all have been used sufficiently to provide reliable estimates of their economic worth. 2/ Further research, coupled with government-assisted develoment projects which build on research results and experience, are needed to encourage wider use of proven technolo- gies. Moreover, because these products are bulky and perishable (although less than in raw root form), efficient marketing systems are required. This in many cases requires improved roads, transportation and storage facilities. 7.20 Opportunity also exists to produce TRCs for other off-farm uses--as a carbohydrate source in domestic livestock production/fish culture programs, as an export product to the livestock industry in developed countries or for the local manufacture of starch or alcohol. Again, questions of choice of technology exist--both with respect to the technology in produc- ing raw materials (fresh roots) and the scale of technology for industrial 1/ This village-oriented technology reduces the time spent in grating cassava for a kg of gari from 20-30 minutes to 1-2 mintues. 2/ Jones and Akinrele, op.cit., conclude (p. 43) that the most economical mechanized gari plant would have a daily output of 10 tons and utilize 35-40 tons of cassava root. Plants of this size are considered suitable to meet gari demand in larger consuming centers, but pose problems in the organization of raw material supply. A few plants of this design have been constructed in Nigeria and Ghana and are generally linked with cassava plantations of at least 1,600 ha. 64 processing of the TRCso The options are numerous and governments and donor groups can be instrumental in technology choice. At one extreme are small' holder production systems supplying fresh roots to small-scale chipping and drying facilities costing a few hundred dollars or to local starch produc- tion plants with capital investment of a few thousand dollarso At the other end of the technology spectrum are plantation-based systems to supply fresh roots to large, capital-intensive starch factories based on imported plant and equipment or to alcohol plants with capacities of several thousand liters per day. Technology choices which contribute to broadly-based growth and income distribution objectives are possible for many of the industrial processes which utilize the TRCs, but this requires a clear delineation and understand- ing of technological alternatives and correct specification of the real economic costs of the resources involved. 7.21 The choice of the technology in processing the roots is perhaps the aost critical in determining the eventual distribution of development benefits since this technology then is an important determinant of production systems for the raw materialo A decision to invest in large-scale processing equip- ment, be it for the manufacture of pellets, starch or alcohol, is likely to lead to pressure to invest/develop large-scale systems for the production of raw material in order to increase control over the supply and price of the raw material0 For the TRCs, for which transport costs are high and timely processing is critical, the economic viability of the industrial operation depends heavily on the reliability and cost of raw material supplies0 In general, the larger the capital costs, or scale of technology, the greater the sensitivity of profitability to continuity of supply of raw material because fixed costs constitute a larger proportion of unit production costs0 Small-scale production and processing technologies in TRC-oriented develop- ment programs are possible, as evidenced by the cassava chip/pellet industry in Thailand and the cassava starch/flour industry in Northeast Brazil0 1/ However, because of the industrial requirements for continuous raw material supplies, close integration of root production and processing activities is required for economic viability0 7.22 Conscious efforts by the research community, by governments and aid donors are required to ensure that production systems and processing technolo- gies will optimize benefits from TRC-based rural development programs0 Where development of smallholder agriculture is national policy, an integrated program of TRC production and industrialization is likely to require a higher degree of public investment in extension and marketing infrastructure than would an approach based on large farms0 An emphasis on smallholder production would also require parallel research and development work on small-scale chipping and drying technology and small-scale pelletizing and starch plants0 1/ In Thailand average size of cassava farms is less than 9 ha of which about half is planted to this crop0 Average fresh root capacity (tons/day) of various processing facilities is. chip plants 16 tons; pellet plants 21 tons; starch plants 32 tons0 ToPo Phillips, Cassava Utilization and Potential Markets, IDRC, Ottawa, 1973. - 65 - Little research work appears to be underway in the development of smaller- scale but efficient, starch plants capable of producing the quality product increasingly demanded by the market. Large-scale processes in alcohol pro- duction from TRCs may be required for technical reasons, although with good organization, this could well include a role for small-scale producers of the raw material. 7.23 An attractive TRC-oriented development project for a country where e.g. cassava is important in local production and consumption would have as basic objectives increased supplies of various cassava products for local consumption by humans and livestock. The project could usefully encompass research of an applied nature (related to production and utilization) as well as support of directly-productive activities using technologies which are already available. Components might include agricultural production and research (cassava and complementary crops), livestock production (cassava-based feeding programs for dairy, beef, swine or poultry), physical infrastructure (construction of rural roads, marketing and stroage facilities), agro-industry (processing facilities for gari-type products and cassava chip/pellets) and nutrition (production and utilization of fortified or com- posite flours which include cassava). Possible variants of these components are numerous and provide great apportunities for the imaginative project designer. Replication of projects along these lines would provide a wide range of benefits to local producers and consumers of this crop. 7.24 The export market. While the export market for TRC products such as cassava pellets is not unlimited, it is likely to offer good possibilities for a well-organized cassava development program. A key determinant of success in this area is the competitiveness of this source of feed energy in terms of price and quality with alternative energy sources. Feed compounders require regular, dependable supplies, attractively-priced and in adequate quantities, if they are to base their industry on the use of cassava products. Consistency of quality is no less important. Quality standards have been discussed in Section IV. The quantity of raw root production needed to justify the establishment of a cassava pelletizing industry is governed in part by ship size and the need for regular supplies by the feed compounder. To benefit from economies in transport costs, cargoes should be no smaller than 10,000 tons. Assuming shipments are made regularly at four- to six-week intervals, this implies an annual export requirement of 100-200,000 tons, equivalent to at least 250,000 tons of wet roots. At a yield of 10 tons/ha, 25,000 ha of crop would be needed in each major project. This requirement could be reduced if other animal feed ingredients acceptable to the shipper and buyer were available, e.g., copra expeller or -rice bran pellets. 7.25 Most international trade in cassava products for livestock feed is in the form of pellets. The move away from cassava chips is likely to continue in order to benefit from the reduced shipping costs and greater ease of handling which pellets offer. Pelletizing factories range in capacity from 20-90,000 tons of pellets per annum, with factories producing brand name pellets for the European market having a capacity of about 60,000 tons. Thus, to achieve exports of the size envisaged (100-120,000 tons), two factories would be required. The total cost (1977) of establishing a pelletizing 66 factory of 60,000-ton capacity is in the range of US$500,000 to $750,000 if European or American machinery is used. Machinery made in Thailand also is available and widely used in that country. Costs of Thai machinery are consierably lower than the above figures. 7.26 Because neither shipping nor compounders are anxious to handle cassava chips in view of the extra cost and inconvenience, one approach, now at the experimental stage, may be the manufacture of cassava cubes. Produc- tion costs of cubes per ton are about half that of pellets and the stowage factor is somewhere between chips and pellets. Cubes can be handled as easily as pellets and machinery for cubing is much less costly than that required for pelleting. Cubing machinery is operating in Lampung, Indonesia, and plans are underway to introduce it in South Sulewesi. 7.27 To ensure competitive CIF prices, efficient transportation systems must exist to move low value-to-weight cassava from field to feed mill. Producing countries must have available satisfactory roads in producing areas and to ports as well as adequate harbor facilities0 Much of the success of the Thai cassava industry results from the excellent network of roads in producing areas which allows exports to continue through much of the rainy season0 Distance from producing areas to the port is also an important consideration, Port facilities must be adequate to accommodate ships of at least 10,000 tons capacity0 In Thailand cargoes of 50-80,000 tons are in- creasingly common0 Any Bank-supported project aimed at the export market for TRC products must consider the road and port requirements for efficient handling of these bulky raw materials. 7.28 This analysis suggested that CIF prices of cassava pellets, Northern European ports, no higher than about 80% of.maize prices would be required to ensure financial attractiveness to European feed compounders. Assuming competing cereal prices of about US$150/ ton (CIF), as they were in 1976, cassava prices at Northern European ports could be no higher than US$120/tono Assuming freight charges of US$30/ton, as charged on relatively small Indo- nesian cargoes of 10,000 tons in 1976, an FOB price for cassava pellets of something less thanoUS$100 would be required0 With approximate stability of coarse grain prices forecast for the next several years, feasibility planning for cassava exports could prudently use an FOB price for pellets in the range of $90-100/ton. In a number of important producing countries, eogo, in Latin America, where cassava is a staple food, higher production and lower prices would be required before exports could compete with low-cost supplies from traditional sources in Thailand and Indonesia. 7.29 Inherent in the above statement regarding the need for increased productivity and what has been said earlier about the potential of the TRCs as growing sources of food nutrients and industrial raw materials, is that production research at the national and international centers, together with effective extension, must continue to be given high priority0 Existing production technology, effectively applied, already provides a sound basis on which to build rural development projects which incorporate the TRCs as basic components0 But the full potential of these crops can only be realized when farm yields more nearly approach those being realized in experimental work0 - 67 - ANNEX i Page 1 of 2 CASSAVA TOXICITY I/ A main concern of those involved with the utilization of cassava is the problem of toxicity from the cyanide compounds found in the fresh root. While all cassava cultivars (varieties) possess some amount of these compounds, concentrations are higher in some than in others. Moreover, a particular variety will produce varying amounts of these compounds, depending on soil conditions, humidity, temperature, age of plant, etc. The cyanide is concent- rated in or near the skin and is liberated into its active form when the skin is broken. In this manner, the cyanide compound contributes to the pest resistance of the plant. But under various conditions, a cassava-based diet can lead to chronic cyanide poisoning and, ultimately, to severe mental disorders, goiter and the birth defects which accompany goiter during pregnancy. Chronic cyanide poisoning is notable in some areas of Africa, with the most serious problem in Zaire. The typical syndrome is a poor population which depends heavily on cassava for its staple food, has few sources of high quality protein and, in cases where goiter develops, has an inadequate or marginal supply of iodine. Proper processing is the most effective solution to the problem. Cassava flour, or most kinds of washed or dried cassava, is known to be very low in cyanide. Indeed, areas in Nigeria, ecologically-similar to those with cyanide poisoning, show none of these symptoms, presumably because the cassava is processed in running water. Another approach, basically complementary, is to supplement diets with sulphur-containing amino acids (high quality protein), Vitamin B and iodine. The process of cyanide detoxification in the body utilizes the first two substances, while iodine seems to be used in disposing of the resulting product. An adequate iodine supply facilitiates detoxification and eliminates dietary goiter and the risk of associated birth defects. However, this gen- eral approach is not fully satisfactory because of the difficulties of pro- viding high-quality protein to poor people and because physiological limits exist to detoxification by the human body. Providing adequate supplies of iodine is not difficult if salt is traded and an iodine supplementation program exists. Recently, studies of the human population in the Lake Kivu region in Zaire suggested that a steady diet of cassava inhibits iodine uptake by the thyroid gland and may contribute to endemic goiter with associated cretinism and mental retardation. 2/ The precise relationship of cassava consumption 1/ Much of this material is from the IDRC publication, Chronic Cassava Toxicity (1973), published as the proceedings of a workshop on this topic. 2/ Reported in "Cassava May Lead to Mental Retardation," Nature, Vol. 272, March 9, 1978, p. 121. E- Go A - AIMEX 1 PaS2 2 of 2 to goiter and the related health probleE io yet to be determined. Findings to date point to several arzea of needed research: the prevention of cassava- mediated goiter; improved B@thods of detowifying the fiber before it is con- sumed; and the possible developesnt of neu ¢casava cultiveas ir, which this toxic agent is absent or found at lower concentrations. - 69 - ANNEX 2 Page 1 of 2 DOCUMENTATION ON CASSAVA In 1972 CIAT established a documentation center for cassava, the first objective of which was to acquire, abstract and publish a bibliography of the world's literature on this crop. This is believed to total as much as 4,000 documents. By mid-1975 3,300 references had been identified and over 3,000 acquired. A first bibliography including abstracts and indicator words covering some 2,000 references was published in 1975. Copies of reference in the master collection can be provided on request. Contact: Cassava Informa- tion Center, Apartado Aereo CIAT, Cali, Colombia). CIAT/IDRC Publications Much of the recent work on the tropical root crops has been pub- lished and distributed with financial assistance from the International Development Research Center at Canada. A partial listing of CIAT/IDRC publications includes: Araullo, E.V., Nestel, B., and Campbell, M., eds. Cassava processing and storage; proceedings of an interdisciplinary workshop, Pattaya, Thailand, 17-19 April 1974, IDRC 031e, 1974. 125p. Booth, R.H. Cassava storage. CIAT EE-16, 1975. 20p. Cock, J.H., Macintyre, R., and Graham, M., eds. Proceedings of the Fourth Symposium of the International Society for Tropical Root Crops, CIAT, Cali, Colombia, 1-7 August 1976. IDRC-080e, 1977. 277p. Wholey, D., and Lozano, J.C. A rapid propagation system for cassava. CIAT EE20, 1976, lOp. Diaz, R.O.; Pinstrup-Andersen, P., and Estrada, R.D. Costs and use of in- puts in cassava production in Colombia: a brief description, CIAT EE-5, 1975, 40p. Doll, J.D. and Piedrahita, W. Methods of weed control in cassava. CIAT EE-21, 1976, 12p. Lozano, J.C. et al. Field problems of cassava. CIAT GE-16, 1976, 127 p. et al. Production of cassava planting material CIAT GE-17, 1977, 28p. and Booth, R.H. Diseases of cassava (Manihot esculenta Crantz). CIAT DE-5, 1976, 45p. Nestel, B. Current trends in cassava research. IDRC-036e, 1974, 32p. 70 - aNNE 2 PLge 2 of 2 ed0 African cassave mosaiic, report of an intediOciplinary 'ork- shop, Nuguga, Ke(nyaD, 1922 February R9760 IDRC07le, 1976, 48po and Cock, J0Ro Cassavag the development of an interanational re- search netuorko IDRCO059ea, L976, 6)po and Graham, NMo eds. Oassava as eniiEi feed D pocedingeS of£ work- shop0 University of Guelph, 1-820 April 1977, ODRC°0OSa )47pO _ _ and Nacintyre, R., eds0 O hronic cassava tonicity; picoceedings of an interdisciplinary workEhop. London, D ngland, 29-30 January 1973, IDRC01Oe, 1973, L63po and Nacintyre, Ro, eds0 The interzntional e hc2ange nnd testing of cassava germplasm, proceedings of an in tadiSocplinszy uorkshop. CIAT, Palmira, Colombia, 4-6 Februasy 1975, IDRC=049e, 1975, 74po Persley, G., Terry, E.R., and Macintyre, R, eds. & aGGsva bacterial blight; report of an interdisciplinary torkshop. IXTA, lbsdan, Nigeria, 1-4 November 1976, IDRC-096e (in preos)0 Phillips, ToPo Cassava utilization and potential mriketso IDRC-=020, 1974, Terry, E.Ro and Macintyre, R., eds0 The international egchange and testing of cassava germolasm in Africa; proceedings of an interdisciplinary workshop. IXTA, lbadan, Nigeria, 17°2L November, D1975, IDRC 063e, 1976, 59P° - 71 - ANNEX 3 Page 1 of 2 A NOTE ON SMALL-FARM CROPPING SYSTEMS IN CENTRAL AMERICA 1. The Tropical Crops and Soils Department of CATIE 1/ initiated a research program in 1973 at Turrialba, Costa Rica, to examine various agricul- tural and economic aspects of 24 cropping systems (each with two variants) utilizing the common bean (Phaseolus vulgaris), maize (Zea mays), cassava (Manihot esculenta) and sweet potato (Ipomea batatas). Two production techno- logies were employed (high and low), differing principally in the quantity of fertilizer applied. Small quantities of insecticides were used to control insects in beans and maize. Initial soil preparation was done with a pede- strian tractor. All subsequent cultural operations were done by manual labor. The field trials were carried out in the Turrialba area in the November 1974 - October 1975 period. 2. The basic cropping systems were monoculture, intercropping, rotation cropping and relay cropping. 2/ In all systems with beans, only one crop of beans was grown per year. Two crops per year of maize and sweet potatoes were grown in all systems which included those crops. Results are stated in terms of yields, calories produced and various economic measures related to family returns per hectare (see attached table). Not surprisingly, those systems which produced the most energy included cassava and/or sweet potatoes. Systems with a high proportion of energy from protein sources generally included beans or maize. 3. Of interest in the context of small-farm development and the role of the TRCs is the relationship between the energy content of a particular cropping system and economic returns from that system. Several economic measures are of interest, including: (i) net income/ha (gross returns less total costs, with all labor priced at the market wage); (ii) gross margin (gross returns less labor and material costs) and (iii) family income (gross returns less material costs). These measures are included in the table. 4. The table includes a selected list of those cropping systems which meet both of two nutritional considerations: (i) calorie production per ha 1/ Tropical Agricultural Research and Training Center, Turrialba, Costa Rica. 2/ The term cropping systems refers to a set of crops making up the cropping activities of a farm. Monoculture is a system involving only one crop grown on a farm during the agricultural year. Intercropping is a particular type of multiple cropping in which one crop is part of the environment of another crop system. Rotation cropping implies a time sequence of crop systems, either sole or overlapped in phase, on the same area. Relay cropping is an intercropping system in which a crop is planted prior to harvest of a standing crop. - 72 - ANNE X 3 !?age 2 of 2 exceeds 25,000 M cal/year and (ii) at least 5% of total energy produced originates in protein sources0 Ten of the 48 "cyotems" (24 systems, each with two subtreatments) meet these requirewants0 Beans, despite the high yield risks in Costa Rica, are a major source of income in these cropping programs, in part because of the attractive governwent guaranteed price0 Sweet potato is the other important contributor to incozE, although both preduction and price of this crop vary widely in that countryo Naize and cassava are the two crops with the lowest production risks in thece systems. 50 The two systems which rank at the top among all 48 ln tems of net income/ha (Numbers 10 and 32) also meet the nutritional requireaents stated above. Within the 10 systems exmmined in the table, these two also had among the lowest labor requirements per ha and the hMghest ratios of gSzos production value to labor cost. Energy yields from particulae systems would, of, course, vary from location to location and economic results are determined in part by commodity prices0 But these recults, which are probably typical over a considerable range of small-farm situations in Latin America, clearly suggest that the inclusion of root crops aS an energy source is not inconsistent with attractive economic returns as wello 73 ANNEX 3 Table 1 - B - - - - - - - 4 riq ____ i -0 ._0 co t "I _ Il I CO 0.o 0~~~~~~~~~~~~~4 - 3 10 " Oo00^b"sS 0 I,o .4 0 O09sOC .4 0. 'C° w 4404 O l- o c * 00 ___ . U 0 U 0 ND cM 4 e 0 _ w aD N r 00 X _ ~~~~~~~~.. O 0 C- 0$ CM CO 0 o - - CA 0 a o o . s .. . .. . W . C °A C *~~ ~~~ 6-.-t4 v. s S ff 0 a ~ ~ ~ 0 C " o I" V I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I _ _A 03 0 e am to co cr to lr s !o, C. . 0 S 0 a C4.4 ...OSO00S0OO .4 .. .. - 4 -4 b 4. @ 0.0' _.A e. 0. °' e - 0 > O 0 0 w b4@ LI lu 4 C4 Z -~ ~ ~ .0 I U t 0 v W P *~~~~~~~~~~~~~~~~~~~~~~ e0 0 0 o ; c Q .4 .. .0 00 J 0 flNNNeeJZ NN, - 0' t4 U) t4 O O4.4 00e e ~ ~ ~ 0 OS O : 0S o.U UN - C ^)gC 'a.0c .z tS .4_ SeC°R r 4 v <0 .4 54 f , ,5 - 74 - ANEX 4 Page 1 of 2 A NOTE ON THE INTERCROFPIWG OF CASSAVA IN NORTHEAST THAILAND 10 The remarkable expansion in the area planted to cassava in Thailand has led to concern by Thai officials over the possible deleterious effect which this development may have on soil fertility and food production in cassava-producing areas. In the mid-1970s work was begun at Khon Kaen University to explore the agronomic and economic feasibility of intercropping cassava with legumes and ttaditional food crops, thereby laying the basis for a small-farm cropping program which is less damaging to soil fertility and which incorporates both food and cash crops. Beginning in 1975, intercropping trials with cassava were carried out at the University Farm as well as in farmers' fields by university staff. Results from the 1976 work are noted below. 20 At the Unversity Farm cassava monocropping was compared with inter- cropping of cassava with locally-adapted legumes (peanuts, soybeans, mungbeans, cowpeas) and food crops (corn, sorghum, upland rice). Each of these plots of 80 sq m in area were fertilized with 3-12-12 fertilizer at a rate of 31205 kg/ha prior to planting. Insecticides were used on all crops and weeding was done three times during the growing season0 Cassava was planted in a 1 m x 1 m spacing with interplanted crops grown in rows between the cassava plants0 3. Yields, calorie and crude protein production and returns per ha for the various cropping combinations are shown in the attached table0 Results suggest that some of these intercrops have little effect on overall cassava yields and could provide valuable protein supplies to farm families while fixing nitrogen in the soil0 Cassava yields in this year (1976) were higher than in similar trials in 1975, although yields of the intercropped crops were low due to an usually severe shortage of moisture during the early part of the rainy season0 Although the figures in the table suggest that yields of cassava from monocropping were slightly higher than those obtained when intercropped with peanuts, soybeans, mungbeans, upland rice and corn, these differences were not satistically significant0 The severe drought in 1976 in the Notheast reduced yields of several of the intercrops, although peanut yields appeared to be least affected0 The resistance of qassava to moisture shortages and its potential value as food crop in times of drought is also demonstrated by this work0 4. Several of these cropping c6mbinations produced large amounts of calories per ha because of the presence of cassava. The intercrops added relatively few calories and surprisingly little protein, largely because of the disappointingly-low yields0 5. Of interest are the net returns from the various cropping combina- tions. In this analysis labor costs were excluded on the assumption that these would be provided by the farm families0 Expenditure on planting material and fertilizers was common to all cropso Additional expenditure for insecti- cides was required for all crops except cassava. As seen in the table, net returns per ha were highest for monocropped cassava although the return from - 75 - ANNEX 4 Page 2 of 2 the cassava/peanut combination was only about 5% less. If it is assumed that the peanut crop fixes 30 kg/ha of nitrogen (N2) valued at, say, US$0.40 per kg 1/, the cassava/peanut combination may be nearly as attractive in economic terms and, equally important, establishes the basis for maintaining soil productivity under continuous cropping of cassava. 6. In the trials by university staff in farmers' fields no ferti- lizer was applied, but insecticides were used. Gross and net returns per ha were substantially lower when compared with the trials on the University Farm, reflecting primarily the lower yields. But among trials on farmers' fields near Ban Muong village (see table) net returns from the cassava/ mungbean intercrop were higher than returns from monocropped cassava while net returns from combinations involving peanuts, mungbeans or sorghum were only slightly lower. 7. Generalization from this limited research, based on one year's cropping experience, is of course not advisable. But results obtained to date strongly suggest that in Northeast Thailand intercropping combinations with cassava are likely to be found which have little, if any, adverse effect on cassava yields, but which may be valuable in fixing soil nitrogen and providing a high-protein food crop to farm families. Such an achieve- ment in Northeast Thailand would be an important step forward in deriving a cropping system in a difficult, high-rish, agroclimate zone which pro- vides both a valuable cash crop (itself available as a food crop in times of scarcity of other crops) and food crops which meet basic protein require- ments. But, if such cassava-oriented intercropping programs are to be viable, fertilizer and other purchased inputs will be needed, as well as, the neces- sary agricultural support services, including credit for small farmers. 1/ Assuming a farm-gate price for urea of US$185/ton. -76 - ~~~~~n G 4v ;5! a:7 H o 6 -o > 9F 3 4> 2 H F v (:Ts %D f" U ) N N uo mU Zrs , . X4 -} 4 , - t e@ nH rf m Ch0 0 3 (n m CY§ > O 0f 0: 30, C m 9b a C) C) 4 en Mar OK 0 0 IN ia -r. A8a 4o ., ........CO ID VI'' ,% Go oi o o, . d O D C] cH oooo ooo R g 2~~~~~~~~~~~~~~~~~U 12 ba °1" U > C° a t 1: 29 je . . . . . . . ................. . . . . 4 W~~~~~1 e0000 ls O OOO 0OO 00< 0 S | @ k c9 2 a a~~~~~~~~~~4 n |~~~~~~~S C3 C) ri cuoNGs^N¢ N WBWt 94 t!~~~~~~~ C) a..... .............9;o: g A d~~~~c A 6600 09 0 0 0 h 0 §~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ Or ^9 c i IR o O c3 ua00 O ussovO~% cu t-t 3X4 0 0Pw vs~~~~~~~C Ct C91vX -! 11 -! C! E "°~~~~~~~~~~~~W CO en v" N 14 -Q cz 'O O4 On I7 W -- -- . w (@ a 0 10 2 -0 0sH s aOF>-^ d S 0 9 b F g b < iO b S t~~~~~~~~~Cd C GX 29 # a C9 0 O 41 O > ^ GJ v ., Dii t 3~~~~~~~~~~~~~~~C 14 en cu CO 1.b 9. C) ~~~~~~~79 ANNEX 5 "I - ',able 3 'a, 0 0 0 U U U 0 04 c -4 -*0 I.- w~~~~~~~~~~~~ a, 0% en t- oi o % ,,,oLn 0- oo r, IT un 4 " cr cou- °1~~C N N 4< N ^ o- 0 t-orv a u- _Hc Ai cz 0 Koa e a HINlYI. c aa 02 co i-4 w ~~~~~~~~~~~~~4.4 co U 3 ~ ~~~~~~~~~~~~~~~~~~~~~" 41a 41 41 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~4 A A. C 4. *- coU q . 404 -4U iwc c U U N a 4 5 .4.4 Wl >t 0.0 U 41 co 41 uj 41 Ai44 U.D 4 *~~ U U 410 41 U 0 .,4 5.-4 4J~~~~..- 1e.CUU 4-400 ~~~~~4 14 Z 3 '~~~~~~~41-c 41 ~ 4 IC~lenI-*4 0 AN"_X 5 TahIlc 4 AMINO ACID CONTENT PER 100 GRAMS FOOD Total essential Total Conversion Hethi- Threo- Trypto- Amino amino Food Noloture Nitrogen factor Protein Lysine onine nine phan acids acids (sreme) (grams) (t) (gram) (me) (mg) (u8) (me) (mO) (mg) Cereal grairs Barley 12 1.88 5.83 11.0 406 196 .89 10 4,203 11,118 Maize 12 1.52 6.25 9.5 254 182 L42 67 3,820 9,262 Millet 11 1.55 6.25 9.7 332 239 3`74 189 3,979 9,505 Oats 10 2.23 5.83 13.0 517 234 462 176 5,169 12,998 Rice - browi 13 1.26 5.95 7.5 299 183 207 98 3,033 7,973 - poliched 13 1.13 5.95 6.7 255 150 234 95 2,695 6,785 Rye 12 1.89 5.83 11.0 401 172 ^95 87 3,732 10,868 Sorghum 11 1.62 6.25 10.0 204 141 306 123 3,945 9,756 Wheat 12 2.09 5.83 12.2 374 196 382 142 4,280 12,607 Roots and Tibero Potato 70 0.32 6.25 2.0 96 26 75 33 667 1,572 Sw.eet Potataeo 70 0.21 6.25 1.3 45 22 50 22 414 994 Taro 72.5 0.29 6.25 1.8 70 24 74 26 707 1,737 Yam 72.4 0.3G 6.25 2.4 97 38 36 30 821 2,009 Cassava Mea] 13.1 0.26 6.25 1.6 67 22 S3 19 404 1,184 Legumres (pu' eee) Beans (Phaseolun) 11 3.54 6.25 22.1 1,593 234 Gi72 223 8,457 20,043 Beans, Broad (Vicia) 11 3.74 6.25 23.4 1,513 172 786 202 8,244 20,951 Chick-pea 11 3.22 6.25 20.1 1,376 209 736 174 7,802 19,290 Cowpeas (Vilna) 11 3.74 6.25 23.4 1,599 273 852 254 8,640 21,086 Peanut 5.2 4.62 5.46 25.6 1,036 338 754 305 9,502 27,610 Lentil 11 3.87 6.25 24.2 1,739 194 950 231 9,504 23,447 Peas 11 3.60 6.25 22.5 1,692 205 914 202 8,464 20,901 Pigeon pee 11 3.34 6.25 20.9 1,607 107 6)8 117 7,505 18,460 Soybean 8 6.65 5.71 38.0 2,653 525 1,6)3 532 16,339 40,945 Meat and Poiltry Beef and Ve;1 61 2.83 6.25 17.7 1,573 478 612 198 7,875 17,163 Chicken 66 3.20 6.25 20.0 1,570 502 ;94 205 8,380 18,206 Pork 42 1.90 6.25 11.9 961 321 '83 162 5,203 11,496 Egg (hen) 74 1.98 6 25 12.4 863 416 634 184 6,338 12,763 Fish, fresh all typeo 74.1 3.01 6 25 18.8 1,713 539 861 211 8,465 18,338 Fish, meal 10.1 12.00 6.25 75.0 5,808 2,052 3,180 720 30,360 70,308 Milk, cows 87.3 0.55 6.38 3.5 268 86 153 48 1,620 3,553 Source: FA), Amino Acid Content of Foods and liological Data on Proteins, Nutrit:onal Studies No. 24, Rome 1970. - 81 - ANNEX 5 Table 5 ESSENTIAL AMINO ACIDS OF SOME ROOT CROP PROTEINS Amino Reference Sweet Xanthosama Acids Protein 1/ Cassava Potato Yam Taro Bp. Composition 2/ ---------------- Percent of ----------- (%) Reference Protein Isoleucine 4.2 43 95 91 91 79 Leucine 4.8 60 117 133 117 L42 Lusine 4.2 93 100 105 95 90 Phenylalonine 2.8 75 175 168 179 186 Tyrosine 2.8 56 111 107 150 107 Methionine 2.2 45 77 82 67 59 Castine 2.0 na 55 55 140 140 Threonine 2.8 100 193 122 168 129 Tryptophan 1.4 na na 71 na na Valine 4.2 62 124 98 138 158 1/ Signifies a protein of ideal composition for human nutrition. It approx- - imates whole egg protein. 2/ Only essential amino acids are included. Source: D.G. Coursey and P.R. Haynes, "Root Crops and Their Potential as Food in the Tropics," World Crops, July/August, 1970, p. 264. AURAGE makay IMCIN Yon (DaufLehQ E /g)A Tnw k 971 1972 L9173 974 R975) 197C. Januez- 2,36.0 199.5 233.0 270.0 255o0 290.0 rebT1uagy 237.05 173.o0 225. c¢) 2)32 o 0 230.3 o 2 90.0I HmTch 23500 20§o0 209o0 270.5 MM)o 2(0.0 ApriL 242.5 227.5 223.0 230.0 2G.So0 293.0 Hm9 240.0 230.0 233.0 255.0 275.0C 300.0 Juna 245.O0 232.00 250 o 273.0 2 t5.O0 310 0.0 July 255io0 207.0 1>7Si.0 M.0o 290.0ol 3R2o.i AugRust 24500 220.0 22500 205.0 33000 320oO Setpem ber 237.5 265o0 237-5 3R10o 335.0 330Do0 O¢tobeg 22000 27705 2400C 3L1000 33(o5 3Q0-90 Novembiabt 200.o0 295.0 26g o 0 3 R5 o0 332.o5) 290.0 Dacebe 1EA(0 .0o 2783.5 2!73.0 SO0.C 300.0C 300ot0 YeaE-ly Aveirege 231Lo3 23404 23207 27/Co& 2g)20. MMlo /Afloat, c9r CIS om-'Jgio SoureC: Alfzed G. Toepf(mT HmbuTg, %plorTcGAE in /Sg Cassava: Expozt Poteastimi and Hmrhcu RWzadsat Sssa (1977) 9 p. 14. - 83 - ANNEX 5 Table 7 PROJECTED PRODUCTION OF COMPOUND FEED IN SELECTED EEC COUNTRIES (million tons) 1980 1985 1990 Germany Poultry 4.4 5.2 6.0 Pigs 5.2 6.2 7.1 Cattle 4.4 5.2 6.0 France Poultry 5.0 6.2 7.4 Pigs 5.6 6.9 8.2 Cattle 3.0 3.7 4.4 Italy Poultry 3.5 4.3 5.3 Pigs 2.6 3.3 4.0 Cattle 1.6 2.0 2.4 Belitium Poultry 1.5 1.8 2.0 Pigs 3.6 4.3 5.1 Cattle 1.3 1.5 1.8 Netherlands Poultry 2.8 3.3 3.8 Pigs 5.6 6.6 7.6 Cattle 4.5 5.3 6.1 United Kinadom Poultry 3.8 4.0 4.2 Pigs 2.7 2.8 3.0 Cattle 4.6 4.9 5.1 Total Poultry 21.0 24.8 28.7 Pigs 25.3 30.1 39.0 Cattle 19.4 22.6 25.8 Source: T.P. Phillips, Trovical Roots and Tubeisg Futur- MarkptsA Suppliers and Uses: University of Gue lph, Ontario, Canada (February, 1978) p. 45. A 9) SE BAX$ IZ $0 O EXCXL BE3YB AiX 70O6;G$ _IRCWTA Y . _ @MuTa -MRan C i sn $a:0 En 00 ° l Rffico ble m 1 $ o 3 ?CID zEXi m SS toDl a 17o Catcon oood Romi 5.X 2.03 0 CoP-00 E2QI 23 10.7 0.7 ° OGRIGT SEOOD (G'Ufto9 20 G.2 0.7 0 Pnscia iEcOo (6 wft) 23 5.o3 0.O5 0.O9 $tylooenf-PhG (blooED S-R 5.4 1. 7 0.0 s6uTc@. Z. g guliG in fXi?zZgz hp OuQif-2 09 ° O&VCa ost md lanm pEzadeg TechaolaV" (197G), :Zzo Cg§s@&ve EL Aim LCO49 RDUC9 l77. - 85 - ANNEX 5 Table 9 CAPITAL STRUCTURE OF SMALL-SCALE CASSAVA STARCH PLANTS IN COLOMBIA (1977 US$; '000s) Sweet Starch Sour Starch 1/ Fixed Capital 12.5 17.2 Equipment 2/ 3.1 7.8 Buildings and Cement Work 3/ 9.4 9.4 Working Capital 6.1 10.6 Raw Material Inventory 0.6 0.6 Product Inventory 2.7 5.1 Accounts Payable, Receivable and Misc. 2.8 4.9 Contingencies 1.9 2.7 Total Investment 20.5 30.5 Annual Capacity: 1,500 tons of fresh roots; 300 tons of starch Employment: 5-8 workers 1/ Sour starch is a fermented product unique to Colombia and used in the making of a local bread, pan de bono or pan de yuca. 2/ Consists essentially of locally-fabricated washing, grating and screen- ing equipment, sometimes supplemented with mechanized pumps. 3/ Major items are settling and fermentation vats. Source: T. Salazar, E. Zapata, S. Cardenas, E. Cabra, "Small-Scale Produc- tion of Sweet and Sour Cassava Starches in Colombia," (Institute of Technological Research, Bogota), paper presented to the Cassava Harvesting and Processing Workshop, CIAT, Palmira, Colombia, April 25-27, 1978. RECENT PAPERS IN THIS SERIES No. TITLE OF PAPER AUTHOR 290 Major Refdrms of the Swedish Education System A. Heidenheimer (consultant) 291 Industrialization, Technology and Employment - T. Rawski (consultant) China 292 Development and Income Distribution - Zambia C. Blitzer 293 World Potash Survey W. Sheldrick, H. Stier 294 The Economic Dimensions of Malnutrition in M. Selowsky Young Children 295 The Technology of Rural Development J.P. McInerney (consultant) 296 The Financial Cost of Agricultural Credit: C.D. Datey (consultant) A Case Study of Indian Experience 297 Agricultural Sector Planning Models: A.C. Egbert A Selected Summary and Critique 298 Textbooks and Achievement: What We Know S.P. Heyneman. J. Farrell, A. Sepulveda-Stuardo .consultarts) 299 An Economic and Social Analysis of the C. Bruce Chao Phya Irrigation Improvement Project II Y. Kimaro 300 Two-Studies of Development in Sub-Saharan Africa S. Acharya B. Johnston (consultant) 301 The Intermediate Sector, Unemployment, and The W.F. Steel Employment-Output Conflict: A Multi-Sector Model Y. Takagi (consultant) 302 The Economic Theory of the Household and Impact D. Chernichovsky Measurement of Nutrition and Related Health Programs 303 Trade Restrictions and International Price M. Bale Instability E. Lutz 304 Intergovernmental Fiscal Relations in Developing R. Bird (consultant) Countries 305 A Programming Approach to Fertilizer Sector A. Choksi Planning A. Meeraus 306 The Foreign Exchange Gap, Growth and Industrial K. Dervis Strategy in Turkey: 1973-1983 S. Robinson 307 The Importance of Risk in Agricultural Planning P. Hazell, R. Norton Models M. Parthasarathy. C. Pomareda (consultant) LCEIT PAPERS IN THIS SERIES No. TITLE OF PAPER 308 Guidelines for School Location Planning YY. Gould (consultEnt) 309 Gzroth and Poverty in Developing CountKiGo N. -uinD Zg CUGT 310 Teacher Training and Student Achievement in T. Huoena L. Saha Leas Developed Countries R0 Zoonast (consultants) 311 Optimum Economic Power Supply Reliability - Rknsainghe N. GocllsrJ)n (ConsultEan) 312 IntraEIndustry Trade and the Integration of 'L BeSa1s: Developing Countries in the World Economy 313 Export Promotion Policies de vri@e 314 The Changing Composition of Developing N. 7enery Country Exports ID 9e8sing 315 Urban Growth and Economic Development in the M'. Chen Schelg Prospects and Priorities 316 World Trade and Output of Nznufacturea: i Neeoing Structural Trends and Developing "ountries' EIports 317 Cuba: Economic Change and Education Reform M. Carnoy, J. Wertheim 1955-1974 (conoultents) 313 Sources of Fertility Decline: F,ctor tna'ysis R. Faruqee of Inter-Country Data 319 Educational and Economic Effects ef Promotion . R Haddad and Repetition Practices 320 Small Farmers and the Landless in South Asia I.J. Singh 321 Fruit and Vegetable Exports from che R.D. Hunt Hediterranean Area to the EEC 322 Ability in Pre-Schoolers, Earrj,g ,D and R. CKrawe Rome-Environment 323 Priorities in Education: Pre-School; Evidence M. Smilanasy (conssultant) and Conclusions HG3881.5 .W57 W67 no.324 c.3 Goering, T. James. Tropical root crops and rural development / DATE NAME AND EXTENSION ROOM l_____ _ lNUMBER