Upper Volta - Livestock subsector review

Report No. 3306-UV FILE COPY Upper Volta Livestock Subsector Review November 30, 1982 Western Africa Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FISCAL YEAR January 1 - December 31 EXCHANGE RATES-/ 1970 US$1.00 = CFAF 277.71 1971 US$1.00 = CFAF 277.03 1972 US$1.00 = CFAF 252.21 1973 US$1.00 = CFAF 222.70 1974 US$1.00 = CFAF 240.50 1975 US$1.00 = CFAF 214.32 1976 US$1.00 = CFAF 238.98 1977 US$1.00 = CFAF 245.67 1978 US$1.00 = CFAF 225.64 1979 US$1.00 = CFAF 212.72 1980 US$1.00 = CFAF 210.00 WEIGHTS AND MEASURES: METRIC SYSTEM Metric British/US equivalent 1 Kilometre (Km) 6.2 miles 1 hectare (ha) = 2.47 acres 1 liter (1) = 1.057 quarts 1 metric ton (MT) = 1.1 tons 1 Kilogram (kg) = 2.2046 pounds 1/ Period average  FOR OFFICIAL USE ONLY UPPER VOLTA LIVESTOCK SUBSECTOR REVIEW Table of Contents Page I. SUMMARY AND RECOMMENDATIONS .............................. 1 II. COUNTRY BACKGROUND ....................................... 8 III. LIVESTOCK RESOURCES .11 Livestock Population .11 Animal Health .16 Feed Resources .18 Agricultural Byproducts .20 Water .23 IV. LIVESTOCK PRODUCTION AND USE ............................. 25 Production ............................................... 25 Exports ........... ....................................... 28 Transit ........... ....................................... 30 Imports ........... ....................................... 31 Consumption .............................................. 32 Processing ............................................... 35 V. ECONOMICS OF LIVESTOCK SUBSECTOR ......................... 38 Production ............................................... 38 Livestock Marketing ...................................... 41 Credit ............ ....................................... 46 Taxes and Subsidies ...................................... 47 Prices ............ ....................................... 54 VI. INTEGRATION OF LIVESTOCK AND AGRICULTURE .... ............. 59 VII. SUPPORT SERVICES .65 Government Institutions .65 Parastatals .69 Other Agencies .71 Training .73 Research .75 VIII. PAST LIVESTOCK PROJECTS .77 Bank Projects .77 Other Aid Agencies .79 | This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table of Contents (Cont'd) Page IX. GOVERNMENT LIVESTOCK DEVELOPMENT STRATEGY .... ............ 81 Description of Current Strategy ........... .. ............. 81 Evaluation of Government Strategy ..... ................... 84 X. DISCUSSION ........................ ....................... 90 TABLES: 1. Geographic Distribution of Livestock Population and Agricultural Products 2. Evolution of Estimated Livestock Population 3. Geographic Distribution of Livestock Production 4. Evolution of Livestock Production 5. Imports of Livestock Products 6. Destination of Livestock Exports 7. Animal Feedstuffs 8. Estimated Market Prices for Livestock and Livestock Products 9. Prices of Veterinary Products 10. Price Trends 11. Value of Livestock Labor 12. Estimated Government Revenues from the Livestock Subsector 13. Government Investment 14. Export Tax Rates for Livestock 15. Import Duties on Livestock Products 16. Operating Costs, Staffing, and Salaries of Government 17. Infrastructure and Personnel of Livestock Services 18. Geographic Distribution of Vaccination and Treatment of Cattle in 1978 19. Evolution of Incidence, Vaccination and Treatment of Cattle Diseases  Table of Contents (Cont'd) CHARTS: 1. Herdgrowth for Cattle, Sheep, and Goats 2. Cattle Exports 3. Livestock Price Trends 4. Livestock Vaccinations 5. Densities 6. Organigram of Livestock Service MAP: Livestock Subsector Resources, IBRD 14895  - iv - ACRONYMS AIDR Association Internatinale de Developpement Rural ARCOMA Atelier Regional de la Construction des Materiaux Agricoles AVV Autorites des Amenagements des Vallees des Volta BND Banque Nationale de Developpement BOAD Banque Ouest Africaine de Developpement CBPP Contagious Bovine Pleuropneumonia CEBV Communaute Economique du Betail et de la Viande CENATRIN Centre National de Traitement Informatique CERCI Centre d'Experimentation du Riz et des Cultures Irriguees CFDT Compagnie Francaise de Developpement des Fibres Textiles CILSS Comite Permanent Inter-Etat de Lutte contre la Secheresse dans le Sahel CITEC Societe des Huiles et Savons de Haute Volta CNCA Caisse Nationale de Credit Agricole COVOC Conseil Voltaique de Chargeurs CRED Center for Research on Economic Development CRTA Centre de Recherches sur les Trypanosomiases Animales ELAT Ecole de Lutte Anti Tse-tse ENESA Ecole Nationale d'Elevage et de Sante Animale FAC Fonds d'Aide et de Cooperation (France) FED Fond Europeen de Developpement FU Feed Unit IEMVT Institut d'Elevage et Medecine Veterinaire des Pays Tropicaux INSD Institut National de la Statistique et de la Demographie IRAT Institut de Recherche Agronomique Tropicale et des Cultures Vivrieres MDR Ministere de Developpement Rural  -v - ACRONYMS (Continued) MPC Ministere du Plan et de la Cooperation ONERA Office National de l-Exploitation des Resources Animales ORD Organisme Regional de Developpement PEOV Projet de Developpement de l'Elevage Ouest-Volta PHANAVET Pharmacie Nationale Veterinaire RAN Regie Abidjan-Niger (railroad) RDF Rural Development Fund RHV Republique de Haute Volta SCET Societe Centrale pour l'Equipement du Territoire International SCP Societe de Commercialisation des Peaux SODEPRA Societe de Developpement de la Production Animale SODEXPAD Societe d'Experimentation, d'Exploitation et de Promotion des Produits Agricoles et Derives SOFITEX Societe Voltaique des Fibres Textiles SOLVOCOM Societe Voltaique de Commercialisation SOSUHV Societe Sucriere de Haute Volta SOVICA Societe Voltaique de Montage de Materiel Agricole SVCP Societe Voltaique des Cuirs et Peaux USAID United States Agency for International Development   I. SUMMARY AND RECOMMENDATIONS 1.01 This study was prompted by efforts to identify a second livestock project for Upper Volta.l/ Because of the complexities of the subsector, difficulties with the first project, and the.dearth of past studies,2/ it was deemed necessary to evaluate the subsector as a whole. 1.02 Major points from the study include: (a) Livestock is one of Upper Volta's important national resources, contributing importantly to GDP and foreign exchange earnings. However, the subsector has not lived up to the label given to it by the Bank in 1964 as "probably one of the most promising economic activities." 3/ (b) Evaluation of the Voltaic livestock sector is difficult because there are so few firm data on herd size, offtake rates, carcass weights, livestock exports, and levels of domestic consumption. (c) The subsector faces severe physical constraints on expansion of herd size and output, which may explain why herd growth and offtake are lower now than in the early 1960s, despite allegedly better animal health. Lack of water and pasture inhibits the further expansion of extensive livestock producton in areas free of tse-tse flies, and further herd growth there may reduce future productivity if the carrying capacity of rangeland is injured and livestock cannot be adequately fed. Trypanosomiasis suppresses production in the wetter zones with more pasture, both by keeping livestock out and by lowering the performance of those that are grown there. Although most agro-industrial byproducts are exported, traditional agricul- tural byproducts are already being used in traditional extensive and intensive livestock production. 1/ This report has been written largely by Svend Steengaard (WAPA 4) and Charles Humphreys (West Africa Programs), based on several missions to Upper Volta in 1979 and early 1980. Eugene Sinodinos (RMWA) partici- pated in one of the missions and provided valuable commentary. The report was submitted to Government in mid-1981. The Government appointed an ad hoc commission which reviewed the report and conveyed written comments to the Bank in January, 1982. Representatives of the Government and the Bank met in Ouagadougou to discuss these comments in May, 1982. The present version of the report reflects the main points of this discussion. 2/ The last full study was Republique de Haute Volta (RHV), Ministere de l'Agriculture et de l'Elevage, La Production Animale Voltaique-- Perspectives de Developpement, by SCET International under FAC financing, Ouagadougou, 1972. 3/ World Bank, The Economy of Upper Volta, Report No. AF-22A, Washington, 1964.  - 2 - (d) If herd size under existing herding methods cannot be significantly increased, higher output in the future must come from more pasture or better production parameters. More pasture is already available in the east and the south, but access to it requires additional water and control of trypanosomiasis. Otherwise, the carrying capacity of pastures already in use must be improved. Improved production parameter includes lower mortality, higher calving rates, larger carcasses, and slaughter at younger ages. (e) Although herd size and offtake numbers appear to have generally increased--albeit at low rates, there is little evidence to suggest that productivity (in terms of kg of meat produced annually per animal or per hectare) has increased. There are even signs that more slaughter stock is being sold younger (and smaller), reducing current productivity and pointing to a possible future decline in herd growth. Exports declined following the drought as the herd was reconstituted, and it is not yet certain they will recover, in part because of growing domestic consumption. (f) Major economic constraints include stable or bearish world export prices in the face of rising costs of extensive livestock herding and high or rising prices of feedstuffs and young stock, which will tend to reduce the profitability of more intensive production methods unless coastal markets restrict their imports of beef from outside West Africa. (g) The government's strategy to develop the subsector, while coherently formulated and basically oriented in the right direction, has remained largely unimplemented because of insufficient public funds, difficulties in coordinating various administrative services, and some unsound policies (like feedlots and efforts to export meat). Donors, however, must share part of the blame for poor implementation. Finally, there may be insufficient knowledge to carry out the strategy. (h) Major epidemic livestock diseases have been successfully combatted-- reflecting concentration on prophylactic animal veterinary care. But the present situation is precarious because of a decline in the ability of the livestock service to provide vaccinations and an increasing risk of contamination from abroad. (i) The livestock extension service is not well equipped to provide advice to improve both animal health and husbandry and to furnish curative animal health treatment. These would be critical to help intensify livestock production. (j) Land tenure appears to be a growing problem for livestock produc- tion, especially as expanding crop cultivation takes over former grazing lands. The government has not yet defined or enforced a land tenure policy that would both protect the rights of herders and encourage farmers and herders to conserve and improve pasture resources.  - 3 - (k) In most of the country, the association of agriculture and livestock is already a fact, reflecting the traditional relationships between herders and cultivators. The large transfer of livestock ownership to farmers brought about by the 1968-73 drought and the increasing cost of extensive herding have strengthened this association. As population density increases, the development of a fully integrated mixed farming system is a prerequisite to increasing the produc- tivity of Voltaic livestock. On this point, most Voltaic officials concur. 1.03 It is important to realize that extensive livestock herding in much of Upper Volta in the long term will be a declining subsector -- owing to increasing constraints on resources caused by growth of population and expan- sion of agriculture. If a fall in livestock production and exports is to be avoided, government interventions must be designed that can give herders and other livestock owners new opportunities to raise livestock. 1.04 While it is obvious that there are a number of small, separate actions that might aid the livestock subsector, it appears that these would not necessarily constitute a major donor involvement in the subsector. Moreover, it appears that there may be insufficient knowledge at the moment to devise major projects focused on expanding production. In light of this, there are perhaps five options for donors to consider. Options 1.05 Indirect Support to the Subsector. Donors could choose to support the livestock subsector indirectly, effectively side-stepping the thorny issue of how to design and implement successful livestock projects. Any assistance for livestock would thereby be contained in components in agricultural develo- pment projects (e.g., mixed farming), RDF, and small and medium scale enter- prises (e.g., poultry processing). Such piecemeal intervention is incon- sistent with the importance of the subsector in the economy and would fail to address the major constraints faced by the subsector. This option will not stop a mediocre situation from further deteriorating. 1.06 Animal Health. Health care is a prerequisite for protecting exist- ing livestock production, and protection against epidemic diseases is a long-term, inherently deficit operation for which foreign funds are perhaps necessary. Donors could limit their direct support to strengthening pro- phylactic veterinary care against major epidemic diseases, which might be provided free to herders. This would come at a time when the danger of epidemic outbreaks is considered by some to be imminent and could help forestall a future decline in output. The support could perhaps be expanded to improve the livestock service's capability to provide routine diagnostic and curative treatment for individual animals, which herders would pay for. To provide a viable health service, given the Government's severe financial constraints, two approaches need to be pursued more vigorously: (a) the establishment of an effective revolving fund at the national level to finance costs of veterinary products and their delivery, and (b) greater reliance on  - 4 - collaboration between groups of livestock owners and private health agents to arrange for veterinary services on a cost covering basis. While health protection is vital insurance, it will not raise productivity unless it improves fertility, reduces calf mortality, and combats parisites, which may be more difficult and costly than vaccinating against epidemic diseases even if herders pay for the drugs. Gains from improved health could even be offset by a decline in productivity caused by lack of feedstuffs in the face of more animals. 1.07 Grazing/Herding. Donors could try to expand and to ameliorate extensive livestock production, which is traditionally and still the most important source of Voltaic livestock output. The option would involve two approaches: (i) moving livestock from overstocked areas to pastures that are not now fully grazed (east and south); and (ii) improving or increasing pastures where stocking rates are already high (center and north). In some ways, it is an extremely attractive alternative because it makes greater use of unemployed resources. It is also possible to envisage better livestock management and better use of pastures than practiced by traditional herders. Such improvements would enhance the gains made possible by better health and expansion of grazing into new areas. But there are several impediments to a program with this focus. Areas that are not grazed now exist because they lack water or are infested with tse-tse flies. Given the cost of controlling trypanosomiasis at the present time, however, use of pastures in tse-tse areas may not be economical for extensive livestock herding where returns per ha are low. The best approach to confront the tse-tse challenge at the present time is a continuation of drug treatment of susceptible animals, coupled with a gradual multiplication of the trypano-tolerant herd. Provision of more water in the east will be technically difficult and expensive because of scarce groundwater and the importance of minimizing disease that might be caused by surface water. There will also be the institutional problem of controlling the use of this water and surrounding pastures. There is little evidence that donors know how to deal with the social problems associated with livestock keeping. Pasture improvement through better range management and forage cultivation is only in initial stages, and few, if any viable tech- niques are available even for pilot schemes. Finally as crop cultivation advances, extensive livestock production must be linked with measures to improve the land tenure for herders; but this problem is hardly being addressed. 1.08 Perhaps most importantly, donors lack the knowledge to aid extensive livestock production successfully in the sudano-guinean zone, beyond basic veterinary protection. The highly structured group ranch component of the Bank's first livestock project was intended to develop a method of assisting herders, and after initial delays and difficulties there are now signs that the component has been successful in delivering some benefits, like veterinary care and facilities and some security of land tenure, to herders. There is now even a waiting list of herders who wish to participate and have expressed a willingness to finance future ranches themselves by using credit from the BND. Experience so far suggests the following lessons:  -5- (a) Ranches take time to create and become workable--perhaps 10 years or more; thus donors should proceed slowly and be prepared for a protracted involvement; (b) Herders must be involved from the beginning in the design of both physical infrastructure and social organization, so that investments correspond to their needs and they--not a government agency--will take responsibility for managing the ranches; (c) Initial investments should be simple--like watering places, dips, and squeeze chutes, which would avoid the heavy investment in buildings and other structures under the first Bank project that now seems excessive after the number of ranches has been reduced from 9 to 3; (d) Rights of land tenure have to be clarified and enforced by the government if conflicts between herders and farmers are to be dealt with in a coherent fashion; (e) Potential disease problems--like trypanosomiasis--should be planned for (e.g., by surveying the distribution of tse-tse flies) before replacing traditional, migratory herding with ranch schemes; and (f) There is unlikely to be empty, unclaimed land that can simply be taken for pastoral schemes, and thus relations between farmers and herders will usually be an important issue to resolve. Numerous problems remain to be solved, such as improving management and quality of pastures, controlling herd numbers on the ranches, and developing land tenure codes. The Government recognizes these problems and has asked for financing of master plans of land use potential, hydrology, and socio-economic studies, particularly in the East. Such studies should assess the actual patterns of land usage as well as their theoretical potential and should engage the cooperation of other concerned ministries in the design of any land development schemes. 1.09 A less ambitious alternative to group ranches is small pastoral associations, similar to those the Bank supports in eastern Senegal. Such associations lack the heavy investment and overly supervised management of the ranches. They also focus on the use of collective pasture and water resources and the management of them by stock producers. These are considered appropriate for eastern Upper Volta where lack of water appears to cause underutilization of pastures. It is unclear, however, whether herders want these and whether they could successfully provide the range management called for, especially in light of existing conflicts over land use with farmers--who have primary claim on land; possible tse-tse problems; and the donor's lack of experience in implementing them in ethnically heterogeneous areas. 1.10 Intensification. Intensive livestock production--based on a closer integration of crop cultivation with raising animals--seems to be consistent with efforts to augment livestock productivity in the face of physical (pastures and water) and institutional (land tenure) constraints to extensive production. This option would probably concentrate on livestock production by farmers and by herders that are more or less sedentary. Greater attention  - 6 - would be given to animal husbandry and feeding. Additional feed-stuffs would come from agricultural by-products, cultivated fodder, and improved pastures. Of all proposals, this production system would have the greatest impact on productivity since it includes the gains from better health, careful manage- ment of herds, and supplemental feeds. However, financially viable technical themes for improving animal husbandry and forage cultivation must still be researched, tested with producers, and elaborated into packages for extension. The economics do not yet appear terribly encouraging, and government services are not yet adequate to promote this type of livestock production. 1.11 It is here that donors may be able to provide a substantial support in the form of technical assistance to help both with necessary institutional changes in the government and with applied research and pilot schemes to develop models for eventual production projects. Although many facets could, in fact, be carried out piece-meal through other credits, the need to make changes throughout the sector probably requires a separate program. In any case, progress is likely to be achieved only over the long-term. 1.12 Marketing/Processing. Instead of focusing on livestock production itself, the fifth option would be to intervene in marketing and processing. While improvements in transport infrastructure are clearly possible, there is little evidence that marketing problems constitute one of the major constraints in the subsector. The available evidence suggests that further development of facilities for processing of livestock--in particular to export meat as opposed to live animals--would not be economical because of inefficiencies in the utilization of processing infrastructure, higher transport costs, and the loss of valuable byproducts (such as the fifth quarter of cattle) which are likely to occur. Investments in either marketing or processing are therefore not recommended at this stage. Next Steps 1.13 Although there may not be an overall approach that is ready to be funded, there are a number of individual actions to be considered--some of which may be included in an eventual technical assistance, or pilot project. There are also some actions that might best be avoided for the present, including additional group ranches, large commercial feedlots, eradication of tse-tse flies, most marketing and processing facilities, and--in most cases--support to parastatals. On the positive side, actions to be considered include financing for: (a) Pilot schemes, perhaps in the context of agricultural development projects, to explore models for closer integration of crop cultiva- tion and livestock production. The most obvious place for these is on the crowded Central Plateau or in the agriculturally rich south- west, although trypanosomiasis may be a problem in the latter. They could also possibly include agro-pastoral schemes in the east, provided tse-tse is not a problem there and progress can be made on land tenure conflicts. But elaboration of such components may have to await results of ongoing projects;  -7- (b) Specific, limited investments--for example, possible stock routes and poultry processing facilities; (c) Vehicles, veterinary equipment, and operating expenses for the Livestock Service to make its agents more mobile and effective; (d) National vaccination campaigns against major epidemic diseases; (e) Research in trypanosomiasis drugs, applied and adaptive research in animal husbandry and feeding, breeding selection criteria, fodder production and conservation, and control of parasites, and field research on socio-economic issues; (f) Training of agricultural extension agents (in animal husbandry) and livestock agents (in individual animal health); (g) Budget support to the government if proposals to raise productivity require that taxes on livestock be reduced or that compensation be paid to farmers for land allocated to herding groups; (h) Research--and possibly for production facilities--to improve the quality and availability of feedstuffs for pigs and poultry, if the feed can be sold at full cost; (i) Promotion of animal traction, probably through agricultural development projects; (j) Increased production and marketing of milk by mixed farming operations; and (k) Improved and increased processing of hides and skins. External assistance could also be directed to: (a) Maintaining a dialogue with the government to help it elaborate a strategy for livestock development that is consistent with available means, including policies to favor land reform and removal of subsidies in slaughter fees and feed prices; (b) Encouraging better coordination between the Livestock Service and the Agricultural Service and ORDs; (c) Encouraging a better coordination of donor programs and helping to assure longer-term financing of projects started; and (d) Encouraging the government to rein in ONERA from being a commer- cial entity itself to a supporting role for private producers and merchants and to refrain from further public investment in ventures like slaughtering, meat exports and intensive feeding operations.  - 8 - II. COUNTRY BACKGROUND 2.01 Upper Volta covers some 275,000 km2 in the center of West Africa, with borders that are more than 600 km from ports on the Gulf of Guinea (see Map IBRD 14895). There are three major drainage systems in the country: the Volta basin, the Comoe basin, and the Niger basin. The major rivers--the Volta Noire and the Comoe--rise together in the high rainfall area near Banfora but flow in opposite directions. The Comoe flows directly toward the Ivory Coast, and the Volta Noire reaches north only as far as Dedougou before descending toward Ghana. The other important, permanent watercourses are all near the fringes of the country. Other rivers rise in lower rainfall areas and usually have only small, even intermittent flows. The topography is a rather monotonous plain, lying between 250 and 350 m above sea level and inclined slightly toward the south. Most of the country is covered by shallow, poorly- structured tropical ferruginous soils that are generally acidic and deficient in phosphorous, nitrogen, and organic matter. The richer vertisols and hydromorphic soils are largely limited largely to the valleys of the main rivers--mostly in the western and southern portions of the country. There is one rainy season, lasting about 6 months in the center of the country but concentrated during the period from June to September. Rainfall coincides with the distribution of soils and creates 3 major zones of vegetation, ranging from sudano-guinean in the south and southwest (over 1,000 mm per year) to Sahelian in the extreme northeast (less than 600 mm). Most of the vegetative cover outside of cultivated areas consists of open forests, savannahs, and bush land. Mineral resources which are known to exist include large deposits of manganese, rock phosphate, and massive sulfide (zinc-lead- silver), and small concentrations of gold. Only the gold and phosphates have been exploited to date, as the economic and financial costs of mineral devel- opment are very high. 2.02 For the purpose of the analysis in this review, the country can be divided into four livestock zones, based on ecological conditions and densities. These are the Sahel, with low rainfall, low population and high livestock densities; the Central Plateau, with a sudanian climate, moderate rainfall and high densities of human and livestock populations; the Eastern Savannah zone, with a climate similar to that on the central plateau but with low densities of both human and livestock populations; and the Southwest, with high rainfall, perennial grasses, tse-tse flies, fairly moderate population densities, and low livestock densities. The Sahelian zone coincides with the Sahel ORD; the Central Plateau coincides largely with the ORDs of Ouahigouya, Kaya, Koudougou, Ouagadougou, and Koupela, except that the southern portions of the latter three are more like the southwest. The Eastern Savannah coincides largely with the Fada ORD. The Southwest is represented by the Bougouriba, Banfora, Bobo-Dioulasso, and Dedougou ORDs, although the latter is similar to the Central plateau in its northern part. Densities shown in Chart 5 are based only on data from the ORDs that fall fully within the zones.  - 9 - 2.03 The resident population of Upper Volta is not accurately known but has been estimated at 6.1 million for 1980. 1/ As many as 0.7 million persons born in Upper Volta--generally young single men better educated than their peers--live and work in other countries, mainly Ivory Coast and Ghana, giving the nation the highest rate of emigration in West Africa. The annual growth rate--estimated by the Bank as about 1.7% during 1960-80 --reflects this high rate of emigration. Over 90% of the population live in rural areas, but its geographic distribution is uneven and poorly correlated with geographic and climatic resources. Almost three-fifths of the people inhabit the central plateau--with its relatively poor, eroded soils and only moderate rainfall, giving a density of over 30 persons/km2, compared to only 12 for the rest of the country. Because of the unfavorable balance of population and resources on the Central Plateau, most of the immigrants come from this area and settle in the relatively underpopulated south and southwest, in foreign countries, or in the main urban centers of Ouagadougou and Bobo-Dioulasso, which contain four-fifths of the urban population. The increase in the urban population, which averaged about 6% per year during 1960-75 2/, would reduce recent rural po.pulation growth to only about 1.3% per year if the overall growth rate is accurate. Although there are some 60 ethnic groups in the country, the Mossi alone comprise almost half the population. The Peuhl--representing most of the herding population--are the second largest group, with only 10%. 2.04 Economic data for the country are poor and much foreign trade goes unrecorded. 3/ However, certain observations are possible. During 1970-79, the growth rate of GDP as estimated by the Bank was 3.9% in real terms, despite fairly high rates of inflation and a decline in the terms of trade. By 1980 GDP at market prices was estimated at about CFAF 266 billion, and GNP per capita is estimated at about US$210. Trade represents a sizeable share of the country's economy; in recent years, imports--mostly equipment and consumer goods--have amounted to over one-third of GDP. Exports--almost entirely agricultural commodities--are inadequate to pay the import bill; the trade deficit averages more than one-fourth of GDP and has grown by more than 50% since the mid-1970s. This deficit is largely but not completely covered by earnings remitted by the substantial number of Voltaic emigrants and by 1/ This section is based on recent Government data compiled and analyzed by the Population, Health, and Nutrition Department, Operations Policy Staff of the World Bank (1982). In addition, information has been taken from Julien Conde, "Migration in Upper Volta" World Bank, Washington, June 1978, (draft) as well as from the Atlas cited above. 2/ See Julien Conde, "Migration in Upper Volta", World Bank, June 1978, draft. 3/ The national accounts for 1970-81 have been revised and updated by the Bank in the forthcoming Upper Volta Country Economic Memorandum (Report No. 4040-UV).  - 10 - inflows of foreign aid, furnished primarily by European agencies on con- cessional terms. Government spending amounts to more than 15% of GDP at factor cost, and revenues--provided mostly by indirect taxes of which import duties are most important--have in the past been largely sufficient to balance the budget. But since 1978, budgetary deficits have occurred and show a worsening trend as the result of significant increases in government spending on personnel, defense, and general administration. 2.05 Upper Volta is primarily an agricultural country. The primary sector contributes nearly 40% to GDP of which about 25 percentage points are supplied by crop cultivation and about 10 percentage points by livestock. Although the entire sector appears to be declining as a proportion of GDP, livestock seems to be declining more rapidly than crop cultivation. Agricul- ture is even more important in international trade, providing over 90% of recorded foreign exchange earnings. Livestock alone furnishes from one third to over two fifths of exports; most of the remainder comes from sales of cotton fiber and sheanut butter. The country is largely self-sufficient in agricultural products, although cereal and milk commodities account for about 10% of imports and much of these is food aid. Over 80% of cultivated land is used for cereals, almost entirely rainfed millet, sorghum, and maize. Rice is a very minor foodcrop, grown only in lowlands and in irrigation projects. 2.06 Cultivation is distributed spatially along with population; and even in regions devoted mainly to livestock, cereal land per capita is roughly the same as elsewhere. Owing to river blindness in many of the more fertile, wetter areas, most of cereal production takes place on the poorer soils where rainfall is uncertain and only marginally sufficient. Groundnuts are also grown in most regions, although they are clearly most important in the east and west. The dominant cash crop is cotton, which climate and development efforts have restricted mostly to the southwest. Tree crops are important, but only mangoes, citrus, and cashews are planted in orchards. Sheanuts are a major source of edible oils, and trees such as acacia albida and caya senegalensis provide fodder for livestock. Vegetables, often a dry season crop, claim little land yet are a non-neglible source of foreign exchange. Cultivation methods are almost entirely traditional and rely largely on the manual resources of small family farms. Chemical fertilizers are used on both cotton and cereals, but only a very small share of the national production of cereals can be attributed to fertilizers. Animal traction is used on less than 5% of the holdings--usually those growing cotton. There is, however, already a significant degree of integration between crop cultivation and livestock production. Most farmers raise some livestock, and most herders cultivate their own fields. The value of manure in maintaining soil fertility is generally appreciated, and crop residues--especially legume tops--are widely used for livestock feed.  - 11 - III. LIVESTOCK RESOURCES 3.01 This chapter presents technical information about physical aspects of the livestock subsector, including a description of the livestock herds and how they are managed, a discussion of animal health, and an assessment of feed resources available for livestock. Livestock Population Cattle 3.02 Data on livestock numbers are poor, and any numbers cited are at best indicative. The Livestock Department estimated in 1977 the cattle population at 2.5 million of which the northern region (the Sahelian zone), accounted for about 20-25% of total cattle numbers. The central region (the northern and eastern sudanian zones), accounted for about 60% and the southern region (the southwest sudanian zone) about 15%. The average cattle density in the country is about 9 head/km2 (see Table 1). The north has about 15 head/km2, the center about 10 head/km2, and the south and east about 5 head/km2 (see Chart 5). 1/ The proportion of cattle in the central and southern regions is considered to have increased somewhat since the drought. 3.03 Zebu. The Voltaic cattle herd consists of two main types, Zebu and Taurin breeds, although crosses of the two breeds are common. Zebu cattle, which account for about two-thirds of total cattle numbers, are found in the northern part of the country. There are crossings between the main Sahelian breeds with variation from one region to another and from one ethnic group to another. The size of the cattle diminishes from north to south with an average mature liveweight of 300 kg. The Zebu is especially suited for draft purposes but their lack of trypanotolerance limits their use to the northern two-thirds of the country. Productivity is low under existing management conditions with calving rates about 50-60%, calf losses 30-35% (calves from 0-1 year old), resulting in an annual offtake estimated at 11% and a herd growth estimated at 2.3% per year. 3.04 Azaouak. These cattle, which are a type of Zebu originally from Niger, are found only in a few herds, numbering about 1,000 head. The breed is well adapted to a dry climate. The animals are larger than normal Zebu-- having a liveweight of 350-400 kg, and are known for their fairly high milk yields. A small breeding herd is kept at the Markoye breeding station in the Sahel, but efforts to introduce them into existing herds have not yet been very successful. 1/ Although not clear, these densities probably reflect herdsizes during those times when livestock are kept closest to the owners' homes. Except for the Sahel, movements probably change overall regional densities very little because such movements are short distance.  - 12 - 3.05 Taurin. These cattle are small with an average liveweight of 200-250 kg. They are trypanotolerent and have an appreciable degree of resistence to streptothrichosis. Like Zebus they respond to higher levels of feeding and management, but under poor conditions--caused largely by tse-tse infestations in southern regions where they are found--productivity is low with calving rates about 55% and calf losses 30%, giving an annual offtake of only 9-10%. They are found in the southern parts of the country mainly below the 12th parallel (below a line through Koudougou and Fada) and in an area almost identical with the area infested with tse-tse flies. 3.06 Cattle raising varies with regions and ethnic groups. In the Sahelian and Sudano-Sahelian zones to the north, cattle are herded mainly by Peuhl families who practice a relatively short seasonal movement with pre- dominantly Zebu herds. These cattle are owned by both Peuhl and farming groups. Cattle in the south of the country are mainly owned by sedentary crop farmers who keep predominantly Taurin cattle or crosses near the villages. Some cattle are also kept near towns for supply of milk or beef. 3.07 Cattle move primarily in search of surface water and fodder as the seasons change. This pastoral system allows the best utilization of pas- tures at the right moment. In the Sahelian zone, in the beginning of the rainy season, animals graze throughout the area. Cattle then make their transhumance to rainy season pastures. In the dry season, they gather around permanent watering points and graze dry season pastures. 3.08 In the Sudanian zone, in contrast, cattle usually stay near the villages during most of the crop season and begin their transhumance near the start of the dry season (September to December). Each area has its specific transhumance pattern according to water and pasture resources. Trajectories here tend to be smaller (30-100 km). 1/ During transhumance, cattle graze on pastures in river valleys or on regrowth in the Savannah. Following the harvest, they also eat stovers on crop fields. 3.09 Cattle raising that is more sedentary takes place from the Central Plateau to more southern areas, where perhaps two-thirds of the cattle are owned by farming groups. The distinction between farmers and herders is becoming much less pronounced. For example, in a study near Kaya one-quarter of Mossi farmers own cattle, and a wider study of the central plateau and eastern savannah zones showed that virtually all the Peuhl have their own fields. The practice of other groups entrusting their cattle to Peuhl herders also appears to be declining, ostensibly as the result of increasing distrust by farmers and growing difficulties faced by Peuhl. The same wider study reports that two-thirds of the Mossi livestock owners herd their own cattle, even while on transhumance. Those that do entrust often consign their cattle to other Mossi, rather than to Peuhl. 2/ These cattle may be kept as much as 1/ See Richard Vengraff, Upper Volta: Environmental Uncertainty and Live- stock Production, International Center for Arid and Semi-Arid Land Studies, Lubbock, 1980, pp. 64ff. 2/ See Vengroff, cited above, pp. 21 and 60-62.  - 13 - a symbol and store of wealth as for production and draft purposes. For cattle herded by the Peuhl, methods are similar to those practiced for Peuhl cattle. For animals kept by farmers themselves, most cattle divagate or are sent on transhumance during the dry season but are herded by children during the crop growing season. At night they are driven into small enclosures without access to feed. The base for cattle feed is natural pastures, fallow land, bottom- land and crop residues of millet and sorghum. Both Taurin and Zebu cattle clean crop fields after harvest, but the shortage of feed during the long dry season is one of the major constraints for further expansion of cattle produc- tion as the use of improved pastures is inexistant and little use is made of agro-industrial byproducts. Sheep and Goats 3.10 There are at present about 1.7 million sheep according to the Live- stock Department. There are several breeds such as the Peuhl sheep in the north, the Mossi sheep in the center and the Djalonke sheep in the south. The Peuhl sheep is the biggest with a liveweight of about 25 kg, while the Mossi and Djalonke sheep have liveweights of about 17-20 kg. Offtake is around 25% or perhaps higher. Goat numbers are estimated at 2.5 million. There are three goat breeds and numerous crossings; the biggest breed is the Sahelian goat to the north with a liveweight of about 25 kg, the Mossi goat on the Central Plateau with a liveweight of about 19 kg and the West African Dwarf goat in the south with a liveweight of about 15 kg. 3.11 Almost all rural families keep sheep and goats, and goats are a much more important economic activity than cattle for women. They are kept as scavengers without necessitating any particular care, except herding during the crop sesson and perhaps some feed supplements such as legume tops during the dry season or when being fattened for religious holidays, like Tabaski. They graze everywhere, including on leaves of trees. Goats, and to a lesser degree, sheep are hardy and adapted to the environment and resist drought better than cattle. The main constraint facing small ruminants is health hazards such as pasteurellosis and gastro-intestinal parasites. Due to their small size and restricted mobility small ruminants are often consumed locally, but their production and export for religious feasts is probably an important economic activity. Poultry 3.12 Poultry production exists in two separate production and marketing channels, a dominant one within the village economy, and a small one within the peri-urban and commercial economy. Data on poultry numbers are poor because the livestock service up until the last few years has neglected all work or development of the traditional poultry sector. In 1969 a SEDES study estimated the national poultry flock at 10 million with an offtake of between 100% and 120%. According to a recent FAO study, the actual poultry numbers are estimated at 19-20 million. This study claims a clear relationship between the poultry flock and the human population, estimated to be of a ratio between 3 and 4 birds per rural inhabitant. However, figures provided by the Livestock Service show no growth in the national flock over the last decade.  - 14 - The size of the poultry flock fluctuates a great deal (up to 100%) according to the season. The numbers are highest from September to November following the hatching period during the rainy season and prior to the heavy mortality period, which starts in the dry season. Crosses between local and imported varieties of chickens make up about 60-70% of the national flock, and the balance is made up of Guinea fowl. The greatest concentration of poultry is in the departments of Kaya, Ouagadougou, Dedougou and the south of Ouahigouya. 3.13 The modern poultry sector is very small and mainly found in the urban areas of Ouagadougou and Bobo-Dioulasso. In 1978 the number of day-old chicks distributed to poultry farmers was only about 60,000 of which about half were produced at the poultry center and the other half imported. Most chicks were layers for egg production. 3.14 Traditional poultry raising is an important activity as most families in the rural areas keep poultry. The village or traditional poultry production is characterized by a system where families on small-holdings keep indigenous chickens and Guinea fowls together in a semi-domesticated state with little care, mainly as scavengers. Poultry raising (but not Guinea fowl) is the most common livestock activity for women. 1/ Local chickens are small, between 1.0-1.5 kg liveweight, very hardy, and lay 40-50 eggs per year. Most eggs are hatched, a few sold at the local market and almost none are consumed by the family. Traditional poultry suffer high mortality, caused mainly by Newcastle disease, although Guinea fowl are more resistant than indigenous chickens. The mortality is seasonal and is alleged to reach 80% in the cold dry season (December through March). Hatching takes place during the rainy season (June through September). The traditional system could be considerably improved through better housing, feeding, and vaccination against the most common diseases. Few improvements have been made, except for a FAC-financed vaccination campaign of village birds in the 3 departments of Koudougou, Ouagadougou and Kaya, started at the end of 1978. Where vaccinations have been accepted, the project has successfully combatted the Newcastle disease; and the mortality among birds vaccinated has been considerably reduced. Similar intervention in other parts of the country should be considered. 3.15 Modern poultry production is characterized by the use of foreign breeds, proper housing, well-balanced feed rations, and adequate sanitary conditions. Under good husbandry the performance can be much higher than the traditional poultry production. However, relatively high prices for feed and cereals limit--but may not preclude--the prospects for commercial poultry production in Upper Volta. 2/ 1/ See Helen Henderson, "The Role of Women in Livestock Production: Some Preliminary Findings," in R. Vengroff, cited above. Based on a study near Kaya. 2/ At a 2:1 feed conversion ratio, and if feed accounts for 40% of costs, chickens would have to sell, ex-producer, for CFAF 400 if feed costs CFAF 75/kg.  - 15 - Pigs 3.16 Data on pig numbers are unreliable, but the national pig population is estimated by the Livestock Service at about 160,000. Compared to depart- mental estimates, this number seems far too small. In Koudougou ORD the number was estimated at about 100,000 in 1977, and Fada ORD recorded pig numbers for 1977 at about 300,000. Most pigs are found in a belt across the southern part of the country. Pigs in rural areas are of the indigenous type, long-bodied, and seldom reach more than 40-50 kg liveweight. There are a few piggeries around the cities with foreign breeds, mainly Large Whites with an average liveweight of 90-100 kg. Offtake is estimated at 60% for indigenous pigs and 80% for improved breeds. The average carcass weight of indigenous pigs is estimated at 25 kg and from improved breeds at 65-75 kg. 3.17 There are, as it is the case in most African countries, two types of pig production: a traditional village type and modern type pig production. The majority of pigs are found in the traditional sector and only a few modern piggeries are found around the few cities. Pigs are raised in the villages in the traditional, rudimentary way where pigs are scavengers mainly eating household wastes and some byproducts. Most pigs raised under these poor sanitary and nutritional conditions are worm and measle infested and the pork is of low quality with a too heavy layer of fat. These pigs are slaughtered and consumed by the family or sold at the local market. Pig production could be increased if farmers were provided with better breeding stock, some con- centrated feed, veterinary drugs, technical advice and assistance in marketing. A small, commercial-type pig production is located near the few cities. This production is characterized by having a larger herd size, using proper housing with cemented floors, keeping improved breeds, and purchasing most of the feed ingredients, supplemented with brewers grain. The feed quality is often poor and most farmers mix their own feed as available commercial feed is expensive and without any guarantee for quality. There is a reasonable scope for improvement of this type of pig production through better management, higher quality feed, and better breeding stock. The Banankeledaga pig breeding station might be reopened and stocked with breeding pigs from the Ivorian pig breeding station in Korhogo. Donkeys, Horses and Camels 3.18 The donkey population is estimated at 200,000. Donkeys, which are sensitive to trypanosomiasis, are widespread throughout the country except for the furthermost southern areas. Donkeys are useful pack and draft animals and are especially prized for small transport at the village and farm level. In fact, most of the carts sold for animal transport are designed for donkeys. Donkeys are not kept for meat production, but when culled after having worked as draft animals for four or five seasons their meat may be consumed. Horse numbers are about 70,000 and, like donkeys, they are kept for transport, although primarily for persons rather than commodities. They may also provide some meat. Both donkeys and horses are major consumers of agricultural byproducts --especially legume tops and cereal stovers. The camel population is low, estimated at 6,000 and mainly found in the Sahelian zone. Like donkeys and horses, they serve as a means of transport. Except for donkeys, none of these appear to be important to either the livestock subsector or to agricultural production.  - 16 - Animal Health 3.19 With respect to epidemic diseases, the health of the Voltaic cattle herd might be considered precarious, even though government records (Tables 18 and 19) show very few outbreaks of major cattle diseases. Even if there is no underestimation of current mortality and morbidity--which is highly unlikely given the weakness of the livestock field service--there is a higher risk that Voltaic cattle will be contaminated with contagious diseases like rinderpest and CBPP by transhuming and commercial herds. Border control posts are inadequate, and there is little control of livestock movements. Nor are Voltaic cattle adequately protected, because vaccination coverage is both very low (10-20% of the herd), even assuming the vaccines are always potent, and probably skewed toward urban centers. As shown in Chart 4, vaccinations against epidemic diseases have fallen steadily since 1974, and half or more of those given are financed under the Bank-s first project in the southwest. Even there vaccinations are declining, however. 3.20 Rinderpest. During the inter-state JP-15 campaign in 1962-76 rinderpest was brought under control. But outbreaks reappeared in the follow- ing years because follow-up measures were inadequate and recently the danger of a new epidemic appears to be increasing. 1/ However, this threat could be totally eradicated by a well coordinated vaccination campaign in all West African countries. Meanwhile, annual vaccination campaigns need to be carried out in Upper Volta. 3.21 CBPP was once widespread but the incidence decreased in the early 1970s. The disease, which can cause heavy mortality in cattle, can be effec- tively controlled by annual vaccinations of all animals. 2/ 3.22 Anthrax and Blackleg are two other infectious diseases that occur in cattle, but their occurence is more sporadic than rinderpest and CBPP. Both can be controlled by vaccination confined to areas where outbreaks occur. 3.23 Pasteurellosis does occur. It can be controlled by blanket vaccina- tion of weanlings or by strategic vaccination in localized areas where out- breaks occur. 3.24 Tse-tse flies, which carry trypanosomiasis that is a major threat to Zebu cattle, occur widely in the Sudanian and Guinean Zones. There are three species of tse-tse flies: Glossina palpalis, Glossina tachinoides and Glossina morsitans. G. palpalis is abundant in forest galleries of the country. G. tachinoides is a riverine type found throughout the Sudanian zone with its northern limit corresponding to the 800 mm isohyet. G. morsitans, a savannah 1/ See Communique of 27-28 October 1980 from the Office International des Epizooties, Reunion d'Urgence sur la Peste Bovine an Afrique de l'Ouest. 2/ At present the Ti freeze-dried vaccine from the Dakar laboratory is used.  - 17 - fly, is found in the western part of Upper Volta, including the area of the first Bank project. The only available map of tse-tse distribution does not cover the eastern part of the country. In the southwest, recent work shows that the northern tse-tse limit has moved southwards, which most likely is caused by the recent drought and perhaps by increased migration to and settle- ment in the south. 1/ Chemotherapy is more and more used to protect Zebu cattle going on transhumance or Zebu cattle raised in infested areas. Treat- ments are paid for by livestock owners at the cost of vaccines (about 90 CFAF per dose). Research on biological tse-tse control is carried out in Bobo- Doulasso. 3.25 The costs and benefits of trypanosomiasis control depend on the method used. Chemical control of tse-tse flies in the Cameroon is estimated to cost initially about CFAF 4,000/ha, with annual maintenance costs of CFAF 1,000/ha. In southern Upper Volta, under extensive herding (4 ha/animal), output would be worth about CFAF 1000/ha per year, in gross terms. 2/ Since most of the value of production appears to accrue to herding labor, net returns would be insufficient even to cover recurrent costs. Furthermore, experience in Africa with these programs has not been encouraging because of the stringent recurrent administrative demands. Biological control of tse-tse flies is not yet cost effective because of the high cost of producing sterile males. Use of drugs appears to be more cost effective (CFAF 300-800 per animal per year compared to a gross value of output of perhaps CFAF 4-5,000 per animal), but there is the danger that the only available drugs (try- pamidium and Berenil) will eventually lose their effectiveness against mutant trypanosomes, especially if treatments are only partially adequate. 3.26 Internal parasites are common in livestock and cause poor produc- tivity of Voltaic herds. An Entente Fund project financed by USAID for deworming young stock has been carried out recently in some areas. 3.27 Poultry diseases are numerous and a main handicap for poultry development, both in the large traditional poultry sector and the small commercial sector. The most common poultry diseases are Newcastle disease (a virus disease), Fowl cholera (Cholera aviaire) and Gumboro disease. All poultry diseases can be controlled but the low management level, especially in commercial flocks, results in high losses. At present, yearly vaccination campaigns of village birds against Newcastle Disease are being carried out in three departments, under a 4-year poultry project financed by France. Reports on the success of the campaign are mixed, but the project has scarcely begun. 1/ These limits are shown in the accompanying map. The first map was prepared by Challier of ORSTOM, with 1977 updates. A new map is being charted for the southwest under German financing. 2/ Based on 11 kg of meat per animal per year, valued at CFAF 400/kg.  - 18 - Feed Resources 3.28 Unimproved natural pastures constitute by far the most important feed resource for the country's 2.5 million cattle and 4 million small ruminants. Also straw from fallow land and leaves from trees are important feeds for ruminants. The use of cereals and byproducts as feedstuffs is small, except by cattle owners who are also farmers. Small stock probably depend more on them for feed than do cattle. 3.29 Pastures. It is roughly estimated that half of the country-s land area is covered with natural pastures of a quality which varies with the country's ecological zones. The other 40% not in current cultivation is in fallow, which can also be grazed. A classification of the various pasture areas and a map have been established in a study by ORSTOM. 1/ In the classi- fication, the country has been divided into six types of rangelands by taking into consideration the composition and quality of grasses, soil types, length of grazing period and the distinction between permanent and seasonal pastures. The different rangelands are shown on the attached map. Category 1 indicates the seasonal flooded grasslands (bourgoutieres), which only cover a small land area, mainly the Sorou Valley in the northwestern part of the country. It provides the key survival link in the seasonally transhumant grazing cycle because of permanent watering sites and better dry season vegetation quality. Main grasses are Echinochloa species. Most likely, these important grazing areas will be lost when future irrigation schemes will be implemented. In the Sahelian zone are found numbers of water pools where flooded grasslands mainly consist of Cyperacees and Vetiveria species. Category 2 indicates areas where crop farming by small-holders is important and more than 25% of the land area is cultivated. Large areas in the central part of the country, like the Central plateau, are under this category. Pastures in this area are poor and mainly consist of grasses on fallow land; stovers from fields is an important cattle feed. It lasts only a few months but creates an important link between crop farmers and livestock herders; it also requires better herding of cattle. The stocking rate of these areas is estimated at about 5 to 10 head of cattle per km2. Depending on the length of the fallow period the main grasses are a combination of Loudetia, Andropogon and Cymbopogon species. In category 3, most of the pastures are of good quality, used all year round and also serve as fodder (standing hay) during the long dry season. The area lies in the Sahelian zone with a low rainfall between 400mm and 600mm. Livestock popula- tion density is high, but human population density is low; cash crop opportuni- ties are limited and livestock exports are the principal source of revenue. Many range sites are severely degraded and severe desertification occurs in village perimeters and near watering points. The area is tse-tse free. Main grasses are Aristida species, Schonefeldia gracilis, Cenchrus biflorus and Loudetia. The area in Category 4 lies in the 600mm to 800mm rainfall belt in the Sudanian zone. Between 5 and 15% of the area is cultivated and grasses are almost the same as under category 2, namely Loudetia, Aristida and Andropogon species. The best arable soils are used for near permanent cultivation with fallow land, and marginal soils are used for grazing. 1/ Introduction a la geographie des aires pastorales soudaniennes de Haute-Volta by Michel Benoit. (Office de Recherche Scientifique Technique Outre-Mer), Paris, 1977.  - 19 - Category 5 is a transitory area between category 4 and 6 with pastures of annual grasses as under category 4 and with perennial Andropogon species as under category 6. Rainfall is between 800mm and 900mm. Due to higher rain- fall, (about 900mm) and deeper soils, the area in Category 6 has a high fodder potential with good grasses which can be used year round due to regrowth during the dry season. Trypanosomiasis lowers the Zebu cattle population and overgrazing is unlikely. Main grasses are a combination of Andropogon species. 3.30 Although it is difficult to estimate carrying capacities of pastures, the densities in Table 1 and Chart 5 suggest that areas free of tse-tse flies may already be fairly fully stocked with animals. In the Sahel ORD, there are only 6 ha of all types of land--including cultivated and sterile areas--for each head of cattle and 1-2 sheep and goats and yet the carrying capacity in this area was assessed by the World Bank's Economic study in 1970 at 8-10 ha/animal. In the two ORDs of the Central Plateaus that are entirely free of tse-tse flies (Kaya and Ouahigouya), there are less than 8 ha of land for each head of cattle and 1-2 sheep and goats. Despite its high livestock density, this region is also more heavily cultivated and populated than the Sahel, which reduces potential pasture. A recent CILSS study suggests that Upper Volta has a potential carrying capacity of only 2.2 million tropical livestock units if grazing is limited to assure adequate regeneration. It notes that this level had nearly been reached in 1977. 1/ Other studies argue that there is no shortage of pasture for the country as a whole, given its present herd size. But the distribution of livestock and pasture are not in balance, resulting in overstocking in the Sahelian north and much under-exploitation in the southwest and east. 2/ If these calculations are correct, most zones where livestock can now be grown are already overstocked. 3.31 Stovers from millet and sorghum are an important fodder resource for cattle. Following the harvest in the beginning of the dry season, cattle graze on the fields, which benefit from the cattle manure. Stovers are often collected to feed livestock, especially in mixed farming systems. Grass and groundnut hay are more and more common, especially around larger towns where some steer fattening (boeuf de case) takes place. Research on fodder crop production is still in an early stage. 3.32 Fodder trees. Leaves and fruits from a number of trees are important fodder reserves for livestock during the dry season. One of the best known is acacia albida, whose leaves and pods are eaten by cattle, sheep, goats and camels. The feeding value is high, especially for the fruits. 3/ 1/ CILSS, "Strategie du Developpement de l'Elevage dans les Pays Sahelians", by IEMVT, Maisons-Alfort, January 1980, draft. The estimated number of tropical livestock units is based on 1.9 million cattle and 3.7 million small ruminants, considerably fewer than the numbers furnished by the Voltaics and used in this review. 2/ See P.N. de Leeuw, "Animal Nutrition and Livestock Development in West African Savanna," Sahel Workshop - Boven Volta, Amsterdam, 1976. 3/ Three to four kg leaves and 1.1 kg of the fruit respectively per Feed Unit (FU); one FU is equivalent to 1 kg of barley.  - 20 - Agricultural byproducts 3.33 Molasses is a good supplemental feed for cattle. It is produced at the country's only sugarcane factory SOSUHV at Banfora. In 1979 the factory produced 13,000 tons of molasses (47% of the sugarcane crop); about 10,000 tons were injected into the irrigation water and the remainder was exported to the Ivory Coast or used locally. The ONERA feedlot at Banfora used 469 tons in 1979. A feedlot created by SODEXPAD in collaboration with SOSUHV experimented a few years ago with cattle fattening by using byproducts such as molasses, wheat bran and cottonseed. Projects might encourage the use of more molasses as a feed supplement for milk cows, work oxen, and steers. The costs of transporting molasses would, however, probably limit such projects to the southwest, and even there costs may make it unattractive to producers. At present, planning is underway to produce alcohol to be mixed with gasoline to reduce petroleum imports. If feasible, the project could utilize the entire production of molasses, thereby eliminating its use for livestock. The choice between using molasses for alcohol or for livestock depends in part on relative profitabilities. Under reasonable assumptions about beef production, molasses could be profitably used up to a cost of about 7 CFAF/Kg, although the price would have to be less if other nutrients were more expensive. 1/ However, the high price of imported gasoline (107 CFAF/l, c.i.f., 1980) probably means that alcohol production will be more attractive than cattle feeding--especially if molasses must be transported very far, thereby raising the price to producers. The question needs further study. 3.34 Draff, which is a byproduct of breweries, is a fairly good feed for pigs although best for cattle. Total annual industrial production is about 9,000 tons of wet grain (80% water) from the two breweries in Ouagadougou and Bobo-Dioulasso. Only a small portion of the draff is used for livestock (5% in Ouagadougou and perhaps 20% in Bobo-Dioulasso), but much is used for fertilizer on vegetable gardens. An animal traction project near Ouagadougou is success- fully making silage from draff. The price, nominally set at 1 CFAF per kg, barely covers local delivery costs. In addition, the draff from traditionally produced millet beer ("dolo") is probably used entirely for pig and donkey feed. 3.35 Cottonseed. Total production of seed cotton was about 75,000 tons in 1979-80; the country's five ginneries 2/ produced over 30,000 tons of first quality cottonseed that is processed for oil or used for seed and about 4,000 tons of third quality cottonseed that is used for animal feed. 3/ The cotton- seed for livestock is from low-quality cotton damaged by insects, amounting to about 10% of the total cottonseed production. ONERA has a monopoly on purchasing various ORDs and livestock projects. Actual prices (early 1980) are 6 CFAF 1/ These assumptions are: 1 kg of molasses equals 0.87 FU; export price of cattle is 200 CFAF/kg, liveweight; a gain of 1 kg liveweight requires 11 FU; and feed amounts to 40% of production costs. 2/ In Ouagadougou, Koudougou, Hounde and Bobo-Dioulasso (2 ginneries); managed by SOFITEX with its headquarters in Bobo-Dioulasso. 3/ Seeds contain about 16% digestible protein.  - 21 - per kg at the ginneries, including bags. At this price the demand by livestock owners is estimated at 10,000 tons annually. The cotton-seed currently used for oil would also make excellent cattle feed; it is currently allocated by governmental policy to the local oil processing plant (CITEC) at 15 CFAF per kg, which is alleged to be competitive with export prices. 1/ At this price, these first-quality seeds could probably be used efficiently for beef production, especially in situations where supplemental feeding gives high marginal returns. Current governmental policy prevents this, however, by giving priority to the production of edible oil. A change in this policy by allowing cottonseeds to be sold to the highest bidder would require decontrol of oil prices and could mean that production of cottonseed oil would no longer be profitable. It is likely that rationing will continue, but there appear to be economic grounds for in- creasing the allocation to livestock. The question requires further study. 3.36 Cottonseed Cake is an excellent protein rich feed supplement 2/ for cattle and the only protein rich supplement available in the country in large quantities. In 1979 the CITEC oil mill in Bobo-Dioulasso processed about 21,000 tons of cottonseed and about 7,000 tons of cotton cake were produced; almost all of the cotton cake was exported, except for 150 tons that were sold locally at about CFAF 30 per kg (including bags), which is the alleged export price. 3/ In principle, this cake could be used domestically as cattle feed if purchased at export prices, but it is unlikely to be profitable at this price unless used as a marginal, high protein supplement. 3.37 Wheat Bran has a high feeding value. 4/ It is produced at the Grand Moulin flour mill in Banfora. Both in 1978 and 1979 annual production was about 4,000 tons, which was sold to various livestock owners. The demand in 1979 was especially high in the Sahelian zone. The price of CFAF 15 per kg has increased recently from CFAF 4 per kg. In a normal year the demand is small and there are about 2,000 tons wheat bran available for livestock feed at present price level. The flour mill exports excess stocks of wheat bran as the storage period is limited to about one year by the small storage capacity (1,000 mt). Future production of wheat bran depends on Government's policy of 1/ This price is slightly lower than one calculated from 1979 European import prices, but transport costs are only estimated. 2/ With a digestible protein content between 45% and 53% according to quality. 3/ This price may be too high. Information compiled in Upper Volta on export prices of cottonseeds and cottonseed cake is not consistent with import prices in Europe as reported in the Bank's commodity forecasts. First, import prices show the opposite price structure, with seeds costing some 20% more than cake, which contrasts to prices in Upper Volta that give seeds half the value of cake. Second, 1979 import prices for cake suggest an export price (fob) of 10-15 CFAF/kg--depending on transport costs, which is less than the 30 CFAF charged in Upper Volta. Thus, world prices indicate that domestic use of cottonseed cake might be taxed by as much as 100%. 4/ 0.71 Feed Units per kg.  - 22 - importing wheat grains or wheat flours. At present, owing to European subsidy policies, imports of flour are cheaper than flour produced locally from imported wheat. 3.38 Rice Bran is only available in small quantities, mainly from three rice mills 1/ near Bobo-Dioulsso, as most paddy is produced in the nearby Kou Valley (950 ha). Rice bran amounts to about 7% of paddy weight. Most is purchased by local farmers 2/ and used as pig, poultry and cattle feeds; some rice bran in the Kou Valley is also used as fertilizer on fields. 3.39 Among other feed ingredients used in smaller quantities are ground- nut cake, fish meal and blood meal. Groundnut cake is produced at the CITEC oil mill in Bobo-Dioulasso but the production is low because of small purchases of groundnuts; in 1979 only 3,400 tons were treated at CITEC which gave about 2,000 tons groundnut cake at an oil extraction rate of 40%. The fob export price of groundnut cake is the same as the cotton cake, CFAF 28-30 per kg. Local fish meal is not produced but small quantities are imported from Mali. Blood meal is produced in small quantities at the slaughterhouse in Ouagadougou, and the facility for drying blood is being rehabilitated. 3.40 Information on agricultural products and agro-industrial products that can be used for animal feedstuffs is summarized in Table 7. Several observations can be noted from this table. First, for the products for which actual consumption by livestock is known, only 3rd quality cottonseeds are fully used. Thus there is scope for expanded use if profitable. Second, the prices (per Feed Unit) of the feedstuffs produced by the industrial sector and not exported are clearly less than either those sold for export (cottonseed and groundnut cake) or those made available by farmers and the traditional processing sector (stovers, legume hay, and sorghum and millet bran). The manufactured complete feeds are extremely expensive because of the high cost of cereals used. It is important to note that the modern feedlots have had difficulty making a profit even with the cheapest byproducts, whereas the private feeding sector has managed with the more expensive ones. Third, the real economic opportunity costs of these feed-stuffs are difficult to ascertain, except for exported oilseed cake and grains that can be imported. For these, the opportunity cost is CFAF 30-40 (per kg) and 65-85, respectively. With an export value of meat at 400 CFAF per kg, Feed Units costing more than 15-20 CFAF per kg are unlikely to be economic (assuming about 12 Feed Units per kg liveweight). 3/ The economic costs of brans and hay are more difficult to determine because they are by-products of food production, but as a minimum they must include costs of handling and transport. They also have alternate 1/ SOVOLCOM rice mill and a private rice mill are located south of Bobo-Doulasso; the third rice mill is located in the Kou Valley. 2/ At a price of CFAF 15-17 per kg. 3/ This same conclusion is reached by IEMVT in its study for the CILSS, "Strategie du Developpement de l'Elevage dans les Pays Saheliens," Maisons-Alfort, January 1980, draft, p.86.  - 23 - local uses (stovers for fuel, legume hay to feed donkeys used for transport, and bran to feed chickens which are both more efficient at converting it into meat and worth as much as or more than cattle per kg at export). It appears, then, that the value of most animal feedstuffs may exceed their worth as feed to produce beef. The two exceptions may be cottonseed (of all qualities) and molasses. 3.41 The evaluation is, of course, more complex than simply comparing costs per FU with the price of beef using the feed conversion rate based on a ration mostly composed of these foodstuffs. Under these conditions, the average economic return appears too low to be justified. But if these high- quality feedstuffs are used merely as a supplement to extensive grazing during critical and limited periods, their marginal economic return can be quite high. The evidence that the demand for 3rd quality cottonseed rises during drought periods and that dry-season backyard fattening relies heavily on these feedstuffs indicates that livestock producers are well aware of the high marginal return under such circumstances. Where grazing is not limited, as in areas like Bougouriba in the Southwest, use of these agro-industrial feedstuffs is less likely to be attractive. Unfortunately, most of the surplus feedstuffs are produced in exactly these areas. Water 3.42 Water. Natural watering sites for livestock include rivers, small streams, and ponds. Many of these dry up after the rainy season, and their geographical distribution is not always well correlated with available pastures. To compensate for deficiencies in natural sources of water, a network of man- made watering points, such as dams, ponds, wells and bore holes, have been built over many years, primarily to supply potable water to villagers, not to local livestock. Although these dams generally provide a year-round supply of water, they are rather few in number and not evenly distributed for effective use by livestock. Village wells, of which more than 3,000 wells were con- structed in the period 1964-1974, mainly supply potable water to the villagers, but are often the only source of water for livestock after the rains have stopped. Water is usually drawn from village wells by hand, which limits the number of animals that can be kept near the village during the dry season. However, village wells are important in supplying the rural population with water. 3.43 Hydrogeological Conditions. Hydrologically, Upper Volta is divided into 3 main regions: (a) the zone with a crystalline substratum which covers 80% of the country. Considerable quantities of underground water exist at varying depths (which may vary by as much as 5 m according to the season) according to the profile of the crystalline strata. The infiltration may reach several thousand cubic meters of water per sq km. In comparison to the availability of water, village consumption is negligible, and quantity of water is not a limiting factor. Factors that do limit the exploitation of the water table include the permeability of the soil, the depth of wells, the means of extraction but not the quantity of water.  - 24 - (b) the Sahel - above the 14th parallel and largely included in the zone with the crystalline substratum, where the water table is no longer regularly supplied by infiltration because of low rainfall (less than 450 mm per year). (c) the sedimentary zone - covering the rest (20%) of the country - the region to the west of the sandstone plateau between Bobo-Dioulasso and Banfora which extends northward to Nouna and the Plain of Gondo. Exploitation of this water table is more assured with less local variation, and output can achieve higher levels. Moreover, water output can be determined in advance from baseline studies.  - 25 - IV. LIVESTOCK PRODUCTION AND USE Production 4.01 Voltaic livestock production can only be roughly estimated; there has never been a census of herd size. The evolution of the national cattle herd is shown in Table 2 and Chart 1. Interpretation of the trends in cattle production requires a simultaneous look at growth rates in herd size and offtake rates. In very general terms, the annual growth rate now is con- sidered to be about 3% per year, with an offtake of 11%. There is some evidence of long-term trends in these rates, but the allocation of total production between growth and offtake may vary considerably. For example, during 1948-54, annual growth rates for cattle were estimated at 4.4%, owing mostly to improvements in veterinarian services. 1/ But the average annual growth rate between 1965-73 was only 1.6%, rising to 2.2% during the latter part of the period (1969-73). During the drought years (1972-74), numbers actual'ly declined by an estimated 300,000, mostly in the Sahelian zone where cattle probably died of hunger of rather than thirst. Since the low point in 1974, estimates show the herd growing by over 2% per year, and by 1978, the herd size had almost returned to the peak level in 1973. Recent growth rates are now estimated to be back to 3%. The evolution of the sheep and goat herd appears to follow the same pattern as that of cattle although growth rates are higher. During 1965-73, the herd appears to have grown by only 2.2% per year; but more recently the rate has been higher at 4.6%. What underlies these dynamics must mainly be speculated, since there are only scattered data on population parameters and changes in the geographic distribution. The drought surely caused a southern movement of cattle, pushed by the lack of pasture in the north and encouraged by the southern descent of the tse-tse limit. But the overall decline in numbers indicates that the southern areas were unable to absorb these cattle and may have themselves even suffered a net decline in cattle population. Since the drought, the country has restocked, which is reflected in the low level of exports during this period. To some extent, these growth patterns probably reflect assumptions developed for planning and for estimating national accounts. What is unclear is why the growth rate has not been higher, especially during the period of restocking after the drought when offtake rates were presumably also low. The most sweeping hypothesis is that the more northern, tse-tse free areas are now saturated, limiting expansion to the rate at which the tse-tse challenge is reduced in the south by clearing land for crops. For example, Herman and Makinen report stocking rates in the northern tier (ORDs of Sahel, Yatenga, and Kaya) to have been about double the estimated carrying capacity in 1969; in the south the grazing capacity was sometimes less than half utilized. 2/ 1/ World Bank, The Economy of Upper Volta, Washington, 1964, p. 7. 2/ See Larry Herman and Marty Makinen, "Livestock and Meat Production, Marketing, and Exports in Upper Volta", in Livestock and Meat Mar- keting in West Africa, Vol. 1, "Synthesis; Upper Volta", by Edgar J. Ariza-Nino, Larry Herman, Marty Makinen, and Charles Steedman, CRED, Ann Arbor, 1980, p. 52. See also para. 3.30.  - 26 - 4.02 Production parameters have never been surveyed with any certainty. The figures discussed here are based on the following parameters. Herdsize 1/ Annual offtake 2/ Carcass Fifth Quarter 4/ Type of Livestock (000 head) (% of herd) weight (kg) 3/ (% of carcass) Cattle 2,653 11 5/ 105 25 Sheep and Goats 4,380 25 12 15 Pigs 164 60 25 10 Poultry 10,000 110 0.75 - 1/ Estimates for 1978; see Table 2. 2/ Offtake parameters are taken from the section describing the livestock population. 3/ Carcass weights are subject to annual variations caused by fluctuations in food supply and depend on both the composition of the herd and the age and sex of animals slaughtered. The figures given here are based on estimates from several services, which usually draw upon data reported by slaughterhouses. The Institut National de la Statistique et de la Demographie (INSD) reports a carcass weight for cattle of 104 kg in 1978, although the figure ranges from 140 in 1969 to only 95 in 1973-75. The "Sous-Commission de la Production Animale - Programme de Production Animale pour le 3eme Plan 1977-1981," (PLAN) uses a national average of 95 kg, with a range of 75 kg for female taurin to 110 kg for male Zebu. The 1972 SCET study, "La Production Animale Voltaique - Perspectives de Developpement", (SCET), also used 95 kg as a national average. For cattle, a carcass weight of 95 kg appears low, although it appears that a large number of young cattle are being slaughtered. On the other hand, the largest cattle are usually exported live. Herman and Makinen, "Livestock and Meat Production, Marketing, and Exports in Upper Volta", in Livestock and Meat Marketing in West Africa, Vol. 1, "Synthesis-Upper Volta", CRED, Ann Arbor, 1980, use 112 kg for Zebu and 77 for Taurin, with a national average of 101 kg per animal in 1978. The carcass weight for sheep and goats is taken from the SCET study, given a herd structure that is 40% sheep and 60% goats. Weights for small ruminants raised in the north are higher by about 1 kg. 4/ Taken from the SCET study. The fifth quarter includes the head, feet, stomach, intestines, heart, lungs, spleen, and liver. 5/ When account is taken of herd composition (70% Zebu and 30% taurin), the annual offtake amounts to only about 10.5%. In the interest of simplicity, and given the uncertainty of the estimates, this precision is ignored here.  - 27 - Donkeys, horses and camels are ignored because they are quantitatively insig- nificant. Annual production of livestock products for both consumption and export in 1978 thus amounted to about: numbers meat fifth quarter milk 1/ eggs 1/ Type of livestock (000 head) (000 mt) (000 mt) (000 mt) (millions) cattle 292 30.7 7.7 47 sheep and goats 1,095 13.1 2.0 27 pigs 98 2.5 0.2 - - poultry 11,000 8.3 - - 10 TOTAL 12,485 54.6 9.9 74 10 Total output would also include herd growth--about 3% per year for cattle and over 4% for small ruminants since the drought. 4.03 The physical productivity of Voltaic livestock can be indicated by the quantity of meat--excluding the fifth quarter--produced per animal in the national herd. For cattle 11.6 kg beef were produced for cattle in 1979 and for sheep and goats about 3 kg were produced. If herd growth is also included, the figures rise to 13.7 and 3.4 for cattle and small ruminants, respectively. Although these figures are estimates, they provide no evidence that cattle productivity has increased over the last decade. 2/ The productivity is among the lowest in West Africa and extremely low when compared to developed countries. 3/ 1/ Based on estimates contained in the SCET study, adjusted for herd sizes assumed for 1978. With a herd containing 40% cows, this estimate is equivalent to about 40-50 liters per cow per year. This is a conserva- tive estimate; on average, each adult Zebu cow may be able to provide 100-120 1 per year for human consumption (assumes 300 1 for calves and takes account of miscarriages and other losses of calves). 2/ In 1979: 30,700,000 kg beef for 2,650,000 head, or 11.6 kg; in 1969, according to the Bank-s Economic Report of November 1970, the compara- tive figures were: 28,315,000 kg beef for 2,500,000 head, or 11.3 kg. In fact, the governnment's own planning document (Sous-Commission de la Production Animale, "Programme de Production Animale pour le 3eme Plan 1977-1981") indicates that productivity has fallen from the level in 1969 when offtake was estimated to be as high as 14%. 3/ The figures in 1969 were: 13 kg in Niger, 15 kg in Mali, and 44 kg in Greece. (From La Production Animale Voltaique - Perspective de Developpe- ment, Tome II, "Note de Synthese," by SCET, 1972.)  - 28 - Exports 4.04 Exports of animal products are perhaps known with slightly greater accuracy than herd size and production, but available figures reflect only authorized or declared exports and may themselves be subject to clerical errors. To illustrate, data for cattle shipments from Upper Volta to the Ivory Coast during 1970 and 1972-77 from three different sources show average annual cattle exports of 46, 53, and 62 thousand head. 1/ There is also allegedly much clandestine trade. There are few firm data that could confirm or deny the importance of such trade, but it could amount to several thousand head per year. In any case, the official estimates show that a large share of national livestock production is exported, except for pigs which appear to be nearly all consumed locally. _2/ 4.05 Cattle. As shown in Table 4, anywhere from 10-30% of the annual cattle production is exported, nine-tenths or more as live animals. If exports recover from the low levels following the drought -- and import data 1/ A comparison of 9 years of Ivorian customs data for Voltaic cattle exports to the Ivory Coast with data for Ivorian imports from Upper Volta indicates the great discrepancy in figures among sources. Ivorian Imports Declared Voltaic Authorized Voltaic Year (000 head) Exports (000 head) Exports (000 head) 1970 51.2 38.9 59.2 1971 .... 41.2 59.6 1972 73.7 72.8 58.0 1973 73.1 59.2 48.6 1974 39.1 55.9 49.5 1975 58.4 71.7 57.9 1976 35.3 20.3 23.2 1977 38.3 112.6 24.7 1978 57.0 28.1 .... Average of 1970 and 1972-77 52.7 61.6 45.9 Ivorian data are from The Ministry of Animal Production, data for declared Voltaic exports are customs reports compiled by CENATRIN, and data for authorized exports are from the Livestock Service as reported by Larry Herman, "The Livestock and Meat Marketing System in Upper Volta: Summary of an Evaluation of Economic Efficiency", in Livestock Production and Marketing in the Entente States of West Africa: Summary Project, Kenneth H. Shapiro, ed., CRED, Ann Arbor, March 1979. The symbol ..." indicates data were unavailable. 2/ There are reports of substantial exports to Niger.  - 29 - from the Ivory Coast suggest they are, then up to 30% of the annual offtake (or 80-90,000 head) may be considered a normal level of exports (assuming changes in domestic consumption are matched by changes in cattle production). 1/ Most declared exports are shipped to the Ivory Coast and virtually none are walked. About 70% are transported by rail and the rest by truck. Of the 500 tons of meat reportedly exported annually by ONERA, about 30% is shipped to Abidjan and the rest to Cotonou (customs data show annual meat exports at only about 100 mt). 4.06 The trend in cattle exports is difficult to assess. It is important to note, however, that the decline in exports began in the mid-1960s, and that the precipitous decline usually associated with the drought began for Upper Volta in 1968-69--a fact masked by imports from other countries (see Chart 2). The decline in exports--both absolutely and relative to crops--does not live up to projections made by Bank reports in 1964 and 1970 that livestock exports were expected to continue to rise and to remain more important than crops. 4.07 As a result of the drought, Voltaic exports no longer enjoy a sheltered market in the coastal cities. The rupture of Sahelian supplies to coastal countries during the drought prompted them to find new sources, and to put into place the infrastructure and market for chilled meat. The drought- induced shortage from the north also coincided with the appearance of cheap meat imports from Latin America and Europe. In addition, the economic dif- ficulties in Ghana have virtually eliminated the previously high level of exports to that country. 2/ In any event, Voltaic exports appear to have reached a nader in 1976, but more recent statistics indicate a recovery. The Ivory Coast estimates that cattle imports from Upper Volta alone amounted to 75,000 head in 1979. Even if exports to other countries were nil (in 1977, about 7,000 head were exported to Ghana, Togo, Benin, and Nigeria), this figure implies a growth in cattle exports of more than 35% over the previous year. What is clear is that future exports of Voltaic livestock will be closely affected by their competitive position relative to beef from other suppliers. 4.08 Sheep and goats. Roughly 30% of national annual sheep and goat production is also exported, or some 300,000 animals (see Table 4). Although they account for less than one-third of national production, sheep comprise almost two-thirds of the small ruminants exported. Thus, this shift in relative proportions reflects the relatively higher value of sheep and means 1/ These are gross exports; during and immediately following the drought, net exports were considerably less. In other years gross and net exports were fairly similar. 2/ In the early 1960s, well over half of Voltaic cattle exports went to Ghana.  - 30 - as well that the quantity of meat actually exported is somewhat higher than livestock numbers indicate in Table 4. As with cattle, most are exported to the Ivory Coast, with about 25-40% arriving by truck and the rest by train. 4.09 Poultry. Poultry exports are estimated at 3.5-4.0 million birds, or over a third of estimated national production, but this figure is not known with great certainty. 1/ Probably over 90% of exports are shipped to the Ivory Coast, and as many as four-fifths of total exports are transported by rail. The most striking aspect of poultry exports by rail is the high rate of loss -- estimated at 20-25% -- owing to poor facilities, over-crowding, and slow travel. Efforts either to improve transport of live animals or to export carcasses could reduce these losses and improve the profitability and competi- tiveness of Voltaic poultry exports. 4.10 Hides and Skins. Including those exported with live animals, from one-half to two-thirds of the cattle hides appear to be shipped abroad, and virtually all the sheep and goat skins are exported. 2/ Despite the govern- ment's policy to treat as many hides as possible before export, most are sold in raw form. The Centre de Tannage in Ouagadougou treats slightly more than 200,000 goat skins per year, which are then exported as "wet-blues". This quantity amounts to only 10-20% of the declared exports of all hides. Exports of hides and skins are eligible for compensation under the stabilization scheme for export earnings established by the Common Market (STABEX), but no payments have yet been made to Upper Volta. Transit 4.11 Until the mid-1970s, Upper Volta was a major transit route for cattle coming from Mali and Niger. Northern herders often brought cattle to be sold in Voltaic markets, from where they were re-exported to neighboring countries in the south. During the decade 1965-74, an average of 58,000 head of cattle passed through the country yearly, more than half as many as Voltaic exports. Since then, declared transits are almost nonexistent, reflecting the efforts of neighboring countries to rebuild their herds (including the prohibi- tion of exports), shifts in markets from Ghana to Nigeria, and border conflicts 1/ See the report of Nicolas Gergely, "Commercialisation des Volailles de Haute Volta," FAO, Rome, 1980. 2/ The export figures for sheep and goat hides in 1978 show quantities almost double those estimated for offtake. There is no good explanation for this anomaly except that skins may be coming in from Mali and Niger.  - 31 - between Mali and Upper Volta in 1974-75. 1/ Transits still occur, however, but are seldom declared. 2/ The reduction in transits presumably has little impact on Votaic production, except perhaps to reduce the spread of conta- gious diseases from neighboring countries. In 1980, Upper Volta has lowered its import duties on live animals from 70.46% to 3.68% to encourage imports of Malian cattle. However, to the extent that transiting cattle are de- clared to be Voltaic when they are re-exported, exports originating from Voltaic production will be overestimated. Transits of sheep and goats have followed the same evolution as cattle, although quantities have been smaller than cattle, both in absolute terms and relative to Voltaic exports. Imports 4.12 To this national output must be added small quantities of imports (see Table 5), which in 1978 are significant only for milk products. In that year, such imports amounted to over 80,000 mt of fresh milk equivalent, over four-fifths being food aid. This level is almost twice the estimated local production. In the previous two years 1976-77, recorded imports of milk products amounted to roughly 24,500 mt of fresh milk equivalent, or about 50% of the national production of cow's milk. The much smaller quantity reflects the smaller level of food aid in those years. In 1978, the value of these milk imports amounted to about 3.3 billion CFAF, or more than the value of all livestock exports in the same year. In the previous year, milk imports amounted to over one-third the value of livestock exports. Clearly there is scope to expand domestic milk production although the comparison of local and foreign prices suggests profitability may be low; a milk scheme would probably require trade protection. Moreover, since most food aid is made available free of charge to schools and hospitals, it aug- ments demand; domestic production could substitute for it only if subsidized. 4.13 Recorded imports of livestock are shown in Table 5. Since 1975 the numbers have been insignificant, although in previous years they amounted to 30 - 50% of cattle exports and 10 - 20% of sheep and goat exports, as illustrated in Chart 2. The decline in imports probably reflects both the efforts of exporting countries to rebuild their herds after the drought and changes in trading routes. In 1980, the Voltaic government virtually eliminated import duties on livestock in an effort to stimulate imports. 1/ There are two additional reasons: prohibition of transits through Upper Volta in 1976 and increases in Voltaic export taxes. The Customs Directorate indicated that transits have never been banned. Moreover, the change in Voltaic export taxes would affect only imports, not offi- cial transits that pay no export taxes. See Larry Herman, "The Live- stock and Meat Marketing System in Upper Volta: Summary of an Evalua- tion of Economic Efficiency", in Livestock Production and Marketing in the Entente States of West Africa: Summary Report, Kenneth H. Shapiro, ed., CRED, Ann Arbor, March 1979. 2/ On this point, see Herman.  - 32 - Consumption 4.14 Consumption of livestock products can be estimated as the resi- dual of offtake minus known exports, or on the basis of estimated intake per capita. A rough estimate of domestic consumption for 1978 as a resi- dual of production in the late 1970s and a normal level of exports is shown below: Numbers Meat, including Type of Livestock (000 head) Offal (000 mt) Per capita Cattle 200 26.3 4.7 Sheep and goats 800 11.0 1/ 2.0 Poultry 7,000 5.3 0.9 Pigs 100 2.7 0.5 Total 8,100 45.3 8.1 4.15 The livestock numbers assume that net exports are at levels pre- vailing before the drought and that herd growth continues at recent rates. Meat quantities are based on parameters given above (see paragraph 4.02). Per capita consumption assumes a population of 5.6 million, the estimated 1978 level. 2/ The resulting annual meat intake of 8.1 kg per capita compares favorably with the assumptions made by the Plan (7.0-8.5 kg per capita) 3/ and by the IEMVT/ CILSS study (8.3 kg per capita in 1977), 4/ but it is much lower 1/ Even though exports contain a relatively larger share of sheep (64%) than national production (40%), the change in average carcass weight is insignificant, given other errors. 2/ Population estimate used by the World Bank prior to 1982 revision, which would place 1978 resident population at about 5.9 million. For details on revised population estimates see IBRD, Upper Volta Country Economic Memorandum, (Report No. 4040-UV), forthcoming. 3/ See RHV, MDR, "Sous-Commission de la Production Animale - Programme de Production Animale pour le 3e Plan 1977-1981," Ouagadougou, 1976. 4/ CILSS, "Strategie du Developpement de l'Elevage dans les Pays Saheliens", by IEMVT, Maisons-Alfort, January 1980, draft. The 1970 World Bank study, The Economic Development of Upper Volta, Vol. 3, "Livestock", Washington, p. 10, uses 6.5 kg p.c. for rural areas and 25 for urban areas, giving about 8-8.5 nationwide.  - 33 - than figures reported in other studies. 1/ In comparison, exports amount to roughly 18,000 mt of meat and offals, or 3.2 kg per capita. 4.16 It is clear that relatively small increases in domestic consump- tion might possibly eliminate any surplus for export, and this argument is often used to buttress proposals to augment national production. However, under reasonable assumption about herd and population growth and changes 1/ The 1972 SCET study (La Production Animale Voltaique - Perspec- tives of Developpement, Tome II, "Note de Synthese") and the recent FAO poultry study (Nicolas Gergely, "Commercialisation des Volailles de Haute Volta," FAO, Rome 1980) give the following annual per capita figures (by): Type of meat rural urban beef 3.3 29.5 mutton and goat 2.4 2.0 poultry 1.0 6.0 Total 6.7 37.5 With 90% rural population, the average intake is 9.8 kg, excluding pork. Three recent consumption surveys give figures that suggest rural consumption may be somewhat higher and urban consumption somewhat lower. FAO studies near Kaya (1978) and Fada (1979) show annual per capita meat consumption of 1.5 and 12.0 kg per year, respectively. The Fada study covered 3 periods of the year -- although the hungry season was excluded, whereas the Kaya study covered only the end of the dry season when incomes are lower and meat prices are highest (see RHV, MDR, Direction des Services Agricoles, "Bilan Provisoire de la Campagne Cerealiere 1979/80," Ouagadougou, October 1979). Thus annual rural per capita meat consumption may be higher than 6.7 kg. The Institut National de la Statistique et de la Demographie (INSD) studied the consumption of workers in Ouagadougou during early 1980 to recalculate the official minimum wage; results give an annual per capita 'neat consumption of 21 kg. Since the respondents were minimum wage earners, average consumption levels may be higher than these results indicate. If rural consumption were 9 kgs and urban consumption were 30 kgs, the national average would rise to about 11 kgs per capita per year.  - 34 - in consumption, the surplus available for export might be actually increase, as shown by the figures below: 1/ 1978 1985 1990 Meat and offals (000 mt) 64.5 78.7 91.3 Consumption (000 mt) 45.4 54.6 64.3 Surplus for export (000 mt) 19.1 24.1 27.0 The consumption estimates for 1985 are broadly in accord with those of Herman and Makinen, but production estimates for cattle and small ruminants are higher by about 15%. 2/ But, if herd size, offtake rates and per capita consumption remain unchanged, while the resident population continues to grow at 1% annually--as projected by the Bank, it could take as long as 30 years to absorb the surplus. On the other hand, if there is no herd growth 1/ The critical parameters are: Annual growth rates (%): cattle 3.0 sheep and goats 4.5 pigs and poultry 0.0 rural population 0.5 urban population 5.7 income 0.4 Income elasticity of demand: 0.75 1978 rural per capita consumption (kg): 5.7 1978 urban per capita consumption (kg): 30.0 The income elasticity is based on that in the Ivory Coast, where incomes and meat consumption are higher. John Staatz ("Meat Supply in Ivory Coast, 1967-1985," in Livestock and Meat Marketing in West Africa, Volume 3, Center for Research on Economic Development, Univer- sity of Michigan, Ann Arbor, 1980) uses 0.6 - 0.7 as the range of income elasticities; the value for Abidjan alone was estimated at 0.55. The demand elasticity in Bamako has been estimated at 0.5 (a range of 0.12 to 0.8), and for Abidjan it is considered somewhat less than 1.0 (see William F. Beazer and J. Dirck Stryker, Financement des Depenses Gouvernmentales Recurrentes pour l'Accroissement du Betail au Mali, USAID, Washington, May 1976). Herman and Makinen, "Livestock and Meat Production, Marketing, and Exports in Upper Volta", pp. 132-133, use 1.0-1.5 for red meat and 0.5-0.75 for offals. 2/ Larry Herman and Marty Makinen, "Livestock and Meat Production, Marketing, and Export in Upper Volta", in Livestock and Meat Marketing in West Africa, Vol. 1, "Synthesis - Upper Volta", Center for Research on Economic Development, University of Michigan, Ann Arbor, 1980, pp. 74, 77, and 132-33.  - 35 - while per capita meat consumption increases the surplus would be absorbed after only about 10 years. Under somewhat different assumptions, the IEMVT study for CILSS estimates that the annual rate of growth (entirely from higher productivity, not herd expansion) in meat production of cattle and small ruminants must be at least 4.2 to 4.6%, respectively, in order to maintain exports at the 1977 levels (defined as kg per capita) while allowing consumption to grow in line with population growth and urbanization. Such output growth rates even exceed European achievements. 4.17 Relative prices are probably the most critical factor, however, and more wealthy foreign consumers may continue to bid away livestock from domestic consumers despite growth in population, cities, and incomes. As shown later, evidence suggests that local livestock prices have risen at the rate of perhaps 9-10% per year since the early 1960s, which is about twice as high as general inflation. But given the poor quality of data and the lack of knowledge about important parameters, it is impossible to predict trends in domestic consumption with much accuracy. Processing 4.18 Most slaughterings are done traditionally by individual village butchers, often on concrete slabs and in hangars especially constructed for this purpose. Approximately two-thirds to three-fourths of the cattle consumed domestically are now slaughtered in this way; and from one-half to two-thirds of the sheep and goats are slaughtered traditionally. In the cities with modern abattoirs, there are legal prohibitions against tradi- tional slaughtering, but there is little evidence that these prohibitions are very effectively enforced. Slaughter fees are charged only by the abattoirs, but butchers are required to pay slaughter taxes on each animal killed, according to the following schedule (CFAF per head) 1/: Type of livestock Ouagadougou and Other Bobo-Dioulasso localities cattle 150 75 sheep and goats 15 8 pigs 30 15 donkeys 75 38 Slaughter fees in effect for 1980 for the abattoirs can be summarized (CFAF per head): Type of livestock Ouagadougou Bobo-Dioulasso cattle 2,000 2,000 plus local delivery 300 sheep and goats 130 300 pigs 1,000 1/ Two-thirds of this tax covers the income tax (Benefices Industriels et Commerciaux, or BIC) and the other third covers the butcher-s license.  - 36 - 4.19 Ouagadougou Slaughterhouse. In 1974 a new slaughterhouse was constructed at a cost of 855 million CFAF, financed by European Development Fund. It is operated by ONERA, employs about 150 workers, and has an annual capacity of 13,000 tons of carcass. There are two slaughter lines for cattle (each with a daily capacity -- 1 shift -- of 150 animals) and one line for small ruminants (with a daily capacity of 250 animals). The present annual throughput is about 5,000 tons of carcass, equivalent to 90 head of cattle, 300 small ruminants and 10 pigs daily. While the cattle lines are underutilized (only one is in operation), the small ruminants line is overloaded and an extension of the small ruminants line is presently being carried out by Dutch aid at a cost of CFAF 200 million. The slaughter- house was originally constructed for export of meat, and as such fits into the government-s strategy to export animal products rather than live animals. The installation has ten cool rooms with a cold storage capacity of about 300 tons, grossly underutilized. Only a small amount (probably less than 10% at any one time) is used by ONERA and private local butchers. Exports of chilled meat amount to about 9-10 tons per week, and butchers use the cold storage only one day a week (for 10-15 tons) since the slaughterhouse does not operate on Sundays. Despite this unused capacity, the cold storage is being expanded. 4.20 Detailed accounts were not made available, but in 1979 the slaughter- house is reported to have had a deficit of CFAF 15-20 million on operating costs alone (CFAF 600-700 per carcass, perhaps). To reduce this deficit, the slaughterhouse tax was raised to CFAF 2,000 per head of cattle in March 1980 plus CFAF 300 for transport into Ouagadougou (about 3% of the purchase price). Fees were also raised from CFAF 115 to 130 for small stock and from CFAF 500 to 1,000 for pigs. The cold storage fee was raised from CFAF 500 to 1000 per beef carcass for 1-15 days. These higher fees, however, appa- rently cover only variable costs; capital costs have been financed by foreign aid. 4.21 Bobo-Dioulasso Slaughterhouse. The Bobo-Dioulasso slaughterhouse, which was built in 1961, has been renovated and equipped under the Bank-s First Livestock Project. The renovation was completed in September 1979 and on January 1, 1980, ONERA took over management of the abattoir. The slaughterhouse purchases no livestock itself, but merely performs the slaughter as a service. City butchers bring their animals, which are then slaughtered by the 18 butchers employed by the abattoir and delivered to the local butchers shops. At present about 60-70 head of cattle are slaughtered daily. Hides and skins are purchased by the Hide Marketing Company (SCP). All the meat is consumed locally and none is currently exported. Only cattle of an inferior quality are slaughtered here as stronger animals are walked to the Ivory Coast where prices are more attrac- tive. 4.22 Hides and skins derive mainly from an annual slaughtering of some 220,000 head of cattle, 550,000 goats and 250,000 sheep. An unknown quan- tity of skins may also cross the border from Mali. Hides and skins are collected from slaughterhouses and slaughterslabs and sold to the two major  - 37 - hides and skins companies, SCP and SVCP, both located in Ouagadougou. SVCP was set up in 1968 as a mixed corporation with 49% Government shares and 51% private shares (mostly foreign capital). The purpose of SVCP was to collect, prepare, transport and buy from producers hides and skins in all stages of processing; it also establishes and operates depots, sales outlets and all types of installations relating to the treating of hides and skins. In 1978/79 SVCP handled about 80% of the hides and skins divided as: cattle hides: 64,241 goat skins 1,139,428 sheep skins 469,758 4.23 Purchase prices for dried hides and skins are summarized below (in CFAF): Type of hide Official purchase price kg per hide Per hide Per kg cattle 640 160 4.0 sheep 385 700 0.55 goats 315 900 0.35 Prices actually paid vary depending on the quality and purchase. For example, first and second quality sheep and goat skins were allegedly purchased for CFAF 500 and 600 per skin in March 1980. Private market prices for wet cattle hides at slaughterhouses ranged from CFAF 5-600 per hide in Bobo-Dioulasso to CFAF 700 in Ouagadougou. 4.24 Tanning. A state tanning company, Centre de Tannage, is located in Ouagadougou. It dates from 1964, and most of its machinery needs to be replaced. The tannery has two lines of production; the main one is the treatment, on a custom basis, of about 18,000 goatskins per month for SVCP. These are returned to SVCP as wet blues" and exported mainly to Europe. The tannery also treats about 1,000 cattle hides per month, which are sold locally for the fabrication of leather goods. The tannery has experienced serious financial difficulties over the years. A project for construction of a new tannery has been under planning, but no investment has yet been made.  - 38 - V. ECONOMICS OF LIVESTOCK SUBSECTOR Production 5.01 The economic productivity of traditional production is difficult to calculate, in part because the costs of production in traditional systems have not been well studied. But a few observations are possible for cattle, based on the figures below: Number 1/ Unit Value 2/ Net Worth Type of cattle (000 head) (000 CFAF) (billion CFAF) Calves - female 382 30 11.5 male 297 15 4.5 Young heifers 371 65 24.1 Young bulls/steers 260 30 7.8 Adult cows 1,064 70 74.5 Adult bulls/steers 279 70 19.5 TOTAL 2,653 53 141.9 At current prices, beef produced per head is worth about 5,000 CFAF, 3/ to which must be added an additional 30% to account for herd growth, a value for milk, and a small sum for the fifth quarter. On this basis, the gross return on investment is in the neighborhood of 15%. 5.02 In the national accounts for 1972 and 1974, 4/ most of this income accrues to family labor, not to capital invested. Herding costs, 1/ Based on the herd structure reported in the SCET study "La Produc- tion Animale Voltaique - Perspectives de Developpement," 1972. 2/ Based on current market prices collected by the mission, except for calves and cows, prices for which are assumed. See Table 8. 3/ 200-220 CFAF per kg liveweight, and a dressing percentage of about 50%. To this could be added a few hundred francs to take account of the fifth quarter. Including milk would raise returns even more. 4/ RHV, Ministere du Plan, INSD, Comptes Nationaux de la Haute Volta- 1972 and 1974, April 1976, Ouagadougou.  - 39 - estimated on the basis of current commercial arrangements in many parts of the country, suggest that the value of labor can perhaps be taken at about 3,000 CFAF, or about half the value of the meat produced (see Table 11). 1/ Aside from other, minor costs, net return on capital would be 5 - 10%, including the growth in herdsize. 2/ Because data are so uncertain, these figures should only be considered illustrative. 5.03 There is also some indication that production costs are rising, although the evidence is largely circumstantial. Livestock owners complain of the difficulty of finding herding labor, while expanding crop fields increase the need for herding labor either to guard fields or to extend the search for pasture. Expansion of agriculture also tends to increase litigations, which seem usually to be resolved by fines assessed on herders. The apparently large number of immature stock sold--often for slaughter--is an indication of possible financial difficulties in the sector. 3/ 5.04 The economic viability of feedlots is difficult to ascertain because all data are not available, but some estimates can be made (in CFAF per animal): 4/ 1/ Wage costs are difficult to ascertain because of varying arrangements, and differences in herdsizes. For example, a herdsize only half that used in Table 11 would double returns to labor, which would then account for nearly all the gross return. On the other hand, all the elements shown in the table are not always paid simultaneously. Payments with only milk, grain, and cash give a labor cost of 1-2,000 CFAF. 2/ Given the high margins of error, this result compares to the 8% which the Bank currently assumes to be the opportunity cost of capital in Upper Volta. Returns in economic prices would also be higher, owing to the net taxation of the sector. Delgado estimates that gross returns from buying and selling male cattle fed on natural pasture or from producing calves from cows entrusted to herders are in the neighborhood of about 20% per year, or considerably higher than this estimate based on national figures. But Delgado's estimate does not take into account any payment for herding costs, except in manure and milk, and falls to 15% when male animals are kept for 3-4 years. See Christopher Delgado, "Livestock Versus Food Grain Production in Southeastern Upper Volta: A Resource Allocation Analysis," Ph. D. dissertation, Cornell University, 1978. 3/ For example, R. Vengroff reports that Peuhl in the center and east sell male cattle that average only 4 years old (see Upper Volta: Environmental Uncertainty and Livestock Production, 1980, p. 82). 4/ These estimates are based on information in OUEDRAOGO Seydou, "l'Embouche Bovine au Ranch de la SODEXPAD Banfora", Memoire de Fin d'Etudes, Insti- tut Superieur Polytechnique, Universite de Ouagadougou, Ouagadougou, Mai 1979; GTZ, "Feedlot Banfora - Ergebnisse des Zweiten Wirtschaft- sjahres und Empfehlungen fur 1980," Eschborn, 1980; and on discussions at the ONERA feedlot in Banfora.  - 40 - Banfora Feedlot (1979) SODEXPAD feedlot (1978) Cost Item Amount Remarks Amount Remarks Purchase of animals 41,000 217 kg at CFAF 37,500 225 kg at CFAF 189 per kg 167 per kg Depreciation of 10,000 750 animals per 1,370 2,400 animals installation year per year Interest on investment (excluded) 750 8% Interest on cattle 2,160 9 months at 8% 1,000 6 months at 8% Feed 8,640 32 CFAF/day 7,650 42.5 CFAF/day over 270 days over 180 days Veteonary services 680 1,625 Wages 6,000 Partial 10,740 Operating Costs 3,000 Partial 3,090 Other costs (unknown) (unknown) Losses 2,480 12% of gross 2,813 7.5% of purchase profits (sales cost minus purchase price, food, veterinary service) TOTAL COSTS 73,960 69,438 REVENUE 70,350 335 kg at CFAF 69,200 346 kg at CFAF 210/kg 200/kg NET PROFIT -3,610 238 (Average daily gain (0.44) (0.68) These figures can be considered preliminary and indicative, and there is considerable variation between the results for the two feedlots. The SODEXPAD experience shows almost no profit, which is consistent with the decision to close the operation; and the Banfora operation shows a loss even though some costs are excluded. It is important to note that much of the gain in revenue results simply from the increase in value per kg that comes with age, which has nothing to do with the feedlot operation. Feed costs for the SODEXPAD operation ranged from 5-7 CFAF/Feed Unit, with an average conversion rate of.  - 41 - 12.6 Feed Units per kg liveweight. Feed costs and amount to about 25% of total costs and are equivalent to about 40% of the value of the weight gain achieved. As a result of the poor performance of the feedlots, ONERA and the Livestock Service are calling for monopoly control over feedstuffs as well as subsidized prices. Such a policy would not only hide the inefficiencies of the operations, but it is inconsistent with current market prices of most agricultural by-products, as shown in Table 7. 5.05 The poor performance of the modern feedlots should be contrasted with that of the small-scale fattening efforts of smallholders. Despite the high costs of feedstuffs, these operations appear to be profitable, judged by the interest that producers show in them. Although a detailed accounting of their operation has not been undertaken, certain features appear to help make them succssful. These include: operation during the dry season only, which minimizes disease problems and allows maximum gain from the seasonal movements in prices; use of family labor, which may have little or no private opportunity cost; insignificant infrastructure because family compounds are used; and higher sale prices because of higher quality beef--although the market for such beef may be relatively thin. Livestock Marketing 1/ 5.06 The marketing system is highly decentralized and operates fairly independently of government interventions, except for minor infrastructural improvements in traditional market places and stock routes, usual commercial licenses, and export fees and levies. Many sales are to other stock owners when cattle are not yet ready for slaughter or are not to be used as draft oxen. Merchants often purchase cattle directly from herders (especially from isolated farmers who sell only a few cattle), but most sales probably occur in formal markets. 2/ Herders (or owners) present their cattle for sale through a local commission agent who is known to both buyers and sellers and who guarantees title to the stock, may guard the stock, and verifies the transac- tion. For this service, the fee is usually 250 CFAF per head, paid by the purchaser. 3/ Most sales are made for cash, except that butchers may purchase 1/ Much of this section is based on Larry Herman. "The Livestock and Meat Marketing System in Upper Volta: Summary of an Evaluation of Economic Efficiency", in Livestock Production and Marketing in the Entente States of West Africa: Summary Report, Kenneth H. Shapiro, ed., CRED, March 1979, and on Nicolas Gergely, "Commercialisation des Volailles de Haute Volta," FAO, Rome, March 1980. 2/ Vengroff's study of owners in the central and eastern zones revealed that one-third sell to other owners and one-half sell to buyers that come to the village. 3/ Herman reports slightly higher fees - 350 to 500 CFAF per head per transaction, but the mission observed the lower fee.  - 42 - on short-term credit from major merchants. For cattle marketing, the observed gross margin between the producer and the wholesale butcher is slightly less than 25% of the wholesale selling price, which probably means the actual margin on meat sales is less than the 25% between slaughter and consumer that is allowed legally. 5.07 The poultry marketing system is distinct from that for other livestock, but it too is considered to be relatively flexible, efficient, and well adapted to traditional, extensive production of poultry. The initial stages are usually carried out by small merchants who travel on bicycles or mobylettes -- providing the name commonly given to Voltaic chickens -- "poulets de bicyclette". An intermediary merchant then purchases these fowl in small rural markets and either sells them directly to urban consumers, urban merchants, and exporters, or exports them himself. For poultry, the gross margin between producer and retailer is about 20% of the wholesale selling price. The retailer receives an additional margin that is about the same absolute magnitude. 5.08 The livestock marketing system appears to function efficiently, and further control, probably to be exercised by ONERA, is more likely to raise total costs rather than bring real savings. Herman's recent study of the system shows virtually no evidence of collusion among buyers to fix prices, in part because of the relatively large number of intermediaries, buyers and sellers. Price information appears to be readily available within markets and rapidly transmitted among markets -- even from major export markets in coastal cities. In general, Herman concludes that the Voltaic livestock marketing system performs well, citing its quick adapta- tion to radical changes in supplies and markets and the fact that gross margins are fairly consistent with costs and reasonable returns to capital, risk, and management. The only evidence of monopolistic gains were for butchers of high-quality meat in Ouagadougou -- reflecting the high cost of entering this specialized trade -- and for cattle shippers regularly using the RAN -- reflecting imperfections in the way the RAN allocates rail cars. The study identifies three factors that could further lower marketing costs: reduction of export taxes, improved truck and rail transport infrastructure, and reduction of numerous unofficial charges merchants claim to have to pay. For poultry marketing, the same constraints exist as for other livestock: poor infrastructure and facilities for truck and rail transport. To the extent that monopolization occurs, it is probably due in part to governmen- tal regulations and other problems that are easier for larger merchants to overcome. For example, merchants interviewed by the mission suggest that delays (of over one year) and difficulties in remitting foreign exchange from sales in non-CFAF countries may be impeding exports. ONERA, however, feels this problem is not serious, given the special arrangements being worked out between central banks. 5.09 Marketing regulations. The section on government taxes gives more details, but regulations affecting marketing are limited largely to exports. They deal mainly with health and tax revenue. Most markets, especially those improved and operated by ONERA, charge nominal market fees for each  - 43 - animal presented for sale. This fee appears to be standardized at 150 CFAF per head for cattle paid by the seller. Local merchants are also required to purchase a license each year, which costs 24,000 CFAF in Ouagadougou and Bobo-Dioulasso, but only 18,000 CFAF elsewhere. Other than these formali- ties, there appears to be no control over local livestock marketing. 5.10 The export of livestock requires that the herd possess a valid CEBV livestock passport, which contains some vital information about the herd and furnishes proof that the various fees and taxes have been paid. In addition to payment of export duties, the export of livestock requires an export license, a health certificate (150 CFAF per head for cattle and 30 for sheep and goats), a certificate of origin and an export authorization (both levied at the rate of 100 CFAF per head). In addition, grazing or passage fees may have to be paid to local authorities but these are also fairly insignificant (e.g., 25 CFAF per head for cattle and 10 for small ruminants). The cost of an export license for livestock exporters varies depending on the number of animals bought and sold and on the location. The following table summarizes the existing regulations (in 000 CFAF): 1/ Numbers of livestock 2/ per year Ouagadougou Ouahigouya Fada Leo Other Places Less than 75 3.15 3.21 3.3 3.45 3.0 75-100 6.3 6.56 6.6 6.90 6.0 150-300 12.6 12.84 13.2 13.8 12.0 300-600 21.0 21.4 22.0 23.0 20.0 600-1,000 52.5 53.5 55.0 57.5 50.0 more than 1,000 105.0 107.5 110.5 115.0 100.0 5.11 Marketing Infrastructure. Except for road and rail transport facilities, marketing facilities for livestock are largely unimproved. Stock routes and cattle markets depend on market flows, and they may develop or fall in importance as markets change. In Upper Volta, the major stock routes traditionally drain cattle from the northeast toward railheads and consumer markets in Ouagadougou, Bobo-Dioulasso, and coastal cities. Until the mid to late 1970s, these routes also carried cattle from Niger and the interior Delta in Mali. Since then, decline in Ghana as a consumer and the growth of the market in Nigeria have caused this foreign stock to bypass Voltaic channels. The routes between Dori and both Ouagadougou and Poueytenga, and between Bobo-Dioulasso and both Tougan and the Mali border (see attached Map) have been improved by delimiting them with concrete posts. The delimi- tation apparently helps reduce land tenancy conflicts by clarifying the 1/ From the government's Ministry of Finances, Code des Impots Directs et Indirects - Monopole des Tabacs, Ouagadougou, after 1975. 2/ For the purposes of measuring livestock numbers, 1 cow or bull equals 1 horse equals 2 donkeys equals 6 small ruminants.  - 44 - herders rights of passage within the routes. Other improvements, such as regular watering points, grazing reserves, and veterinary posts, have not generally been included. The government has proposed that the existing stock route between Ouahigouya and both Koudougou and Ouagadougou be de- limited, but the project has not yet been funded. In general, the govern- ment opposes the improvement of routes from major cattle markets to the country-s southern frontiers because it wants to discourage exports of live animals. The short-term effect of this policy may be to raise the cost of or to discourage cattle exports. 5.12 There are eleven major cattle markets in Upper Volta, of which six have been equipped with various modern facilities, including holding pens, scales, and loading ramps (see attached Map). Investments in cattle markets have not been wholly satisfactory; none of the new facilities in two of the six markets is used, and the scales and ramps in the other markets are seldom if ever used. While improvements in the markets -- especially holding pens -- are probably useful, it is by no means clear that these facilities significantly improve the efficacy or reduce the costs of mar- keting. In part, the modern facilities are ill-designed, and in part they miss the major constraints. Future market improvements probably should at least assure that grazing reserves are available. Improvements in communi- cations, transport means, and banking facilities might also be considered. What is clear is that modern markets cannot simply be created for cattle merchants. 5.13 Transportation facilities. Within the country, most cattle are walked between producing regions and the major consuming and export markets. Sheep and goats are both walked and trucked. For exports of live animals, there is a small amount of trucking toward Togo, Benin, Ghana and Nigeria. Most are walked across the borders or shipped by rail to Abidjan. Meat is exported in small quantities by air and in refrigerated rail cars and trucks. 5.14 With the exception of a few refrigerated trucks owned by ONERA, all the motorized transport facilities are either privately owned (in the case of trucks and refrigerated rail cars) or belong to the RAN. There is no evidence that there is a shortage of trucks, but poor roads in Upper Volta and the availability of other, cheaper means of transport make trucking generally less competitive. The RAN has recently purchased 70 new cattle cars, each with a capacity of 40 cattle, bringing its total rolling stock for livestock transport to about 250 cars. The older cars have a capacity of 27-35 cattle. In addition, ordinary railroad cars are sometimes used. Assuming all cars are used and the turn-around time is as slow as 2 weeks, the RAN has an annual capacity of about 200,000 head of cattle, and even more for small stock. Insufficient capacity no longer appears to be a constraint. The major problems with rail transport appear to be poorly organized bulking of cattle for lot shipment and inadequate holding and loading/unloading facilities in Ouagadougou and Abidjan, creating delays and increasing shrinkage and mortality. The scheduling of rail cars may also be inefficient.  - 45 - 5.15 The costs of different types of transport between Ouagadougou and Abidjan for cattle are given below (in CFAF):l/ type of transport per head per kg per ton - km liveweight carcass 2/ rail - transport only (1980) 8,000 23 3/ 20 4/ 40 - other costs 5/ 10,250 29 25 50 trekking 6/ 4,500 13 15 30 official truck transport rate 24 48 refrigerated trucks 45-50 7/ Costs vary considerably, ranging from about 6% of the purchase price to almost 25%, depending on the estimate and mode of transport. Although trekking is clearly cheaper and in some ways not so risky, it has the disadvantage of taking much longer than rail transport. As cultivation in the coastal countries becomes more dense, trekking will also become less attractive. Comparable information for truck transport is not available, but figures presented by Staatz for 1976-77 suggest it may cost more than 1/ Based on data reported by Herman (1976-77) and collected by the mission (1980). (See Larry Herman, "The Livestock and Meat Marketing System in Upper Volta: Summary of an Evaluation of Economic Efficiency," in Livestock Production and Marketing in the Entente States of West Africa, CRED, March 1979). 2/ At a 50% dressing weight. 3/ At 350 kg per animal. 4/ Based on 1,150 km. 5/ 10% shrinkage and 3% mortality figured on 75,000 CFAF per animal, plus 500 CFAF loading, unloading, and guarding costs. These must be added to transport costs. 6/ Based on some of Herman-s costs inflated to 1980 for trekking between Poueytenga and Lome (865 km), plus 2.25% mortality and 8% annual interest over 2 months on 75,000 CFAF per animal. Data collected from merchants by the mission in 1980 for shipments to Nigeria give costs of about 4,600 CFAF per animal, including mortality and interest charges. 7/ Per kg of beef. ONERA estimates its costs at 45 CFAF per ton-km, based on a 5-year truck life, which may be too long. Staatz reports that in 1977 the truck transport between Bamako and Abidjan cost 50 CFAF per ton-km (see "Meat supply in Ivory Coast, 1967-1985, "Livestock and Meat Marketing in West Africa, Volume 3, CRED, 1980).  - 46 - rail transport costs.1/ If the figures are correct, it is as economical to use refrigerated transport for beef as to ship live animals by rail and truck. Trekking remains, however, the most efficient. Credit 5.16 Information on the amount of credit used for livestock production is not available, but the value is small. Of the over CFAF 40 billion worth of credit owed to the formal banking sector in December 1978, only slightly more than 1% was allocated to agriculture, exclusive of short-term financing for purchasing crops. Most of this credit is used for cultiva- tion, such as financing purchases of fertilizers and investments in oxen and equipment for animal traction. 5.17 Credit for animal traction--which is virtually the only formal credit made available to the livestock subsector--has been provided by several institutions under a variety of conditions. The first loans were made by the National Development Bank (Banque Nationale de Developpement or BND) only for equipment on fairly hard terms: 8% interest, repayable over only two years, with 37% downpayment. This program largely failed because of poor reimbursement rates. Most of the development project have had their own credit schemes, with various conditions. For example, the AVV provides credit for animal traction at 6.2% interest for a duration of 7 years with no downpayments or grace period. The USAID-funded project in Fada provides a similar type of credit at 5.5% over 5 years with 1 year of grace and some downpayment. This program functions well, but administrative costs to pay for the functioning of the program and cover bad debts are as high as 25%, in addition to interest. 2/ Beginning in 1980, the Caisse Nationale de Credit Agricole (CNCA, or National Agricultural Credit Bank) consolidated numerous agricultural credit schemes. Medium-term interest rates were stan- dardized at 9%, which is below the rate of inflation, repaid over a period of 3-5 years with no more than 1 year of grace for amortizing principal. For oxen, a downpayment of 20-25% plus an extra charge for insurance against mortality are usually required. As much as 25-40% of formal agricultural credit is expected to be allocated to animal traction in the near future. 5.18 No formal credit is made available to herders to facilitate or improve livestock production, and private credit is also unimportant in the sector except perhaps what is extended to butchers by cattle merchants. 1/ John Staatz, "The Economics of Cattle and Meat Marketing in Ivory Coast - A Summary," in Livestock Production and Marketing in the Entente States of West Africa: Summary Report, Kenneth H. Shapiro, ed., CRED, March 1979. 2/ For example, see also, "Demande d'un Credit Relai pour le Develop- pement de la Culture Attele'e," Ministry of Rural Development, RHV. Administrative costs amount to about 15% of credit accorded and reimbursement rates (over 3 years) are about 91%.  - 47 - The CNCA will probably not provide credit for herders to purchase reproduc- tive animals and to postpone sales of young stock, for merchants to finance cattle exports, or for entrepreneurs to finance cattle fattening schemes. This credit might make an important contribution to more efficient livestock production, but it is likely that the administrative costs of these new schemes will be high and that they will be particularly difficult to implement and assure high repayment rates. The mobility of herders, coupled with a weak extension service, makes it difficult to control the use and repayment of loans. Exportation of cattle precludes their use as collateral by merchants. Thus, the lack of livestock credit may reflect technical difficulties as much as lack of government commitment. In any case, an effective credit program for herders may be inconsistent with the policy of regional stratification if it enables herders to postpone sales of young stock. Taxes and Subsidies 5.19 Taxes - In addition to the marketing and slaughtering fees and commercial licenses described earlier taxes on the livestock sector can be divided into direct and indirect taxes, of which indirect are by far the most important. The revenue from indirect tax on exports is far more important than from import duties. 5.20 Direct Taxes - There are two types of direct taxes-- those levied on livestock (e.g., cattle taxes) and those levied on livestock producers (head tax, or "impot forfaitaire"). Both are based on the number of live- stock owned. Tax rates and potential revenue from the first direct tax-- that on livestock--are: Type of livestock CFAF per head Potential revenue: 1978 3/ __________________ ______________ (CFAF million) Cattle (over 2 years old) 200 1/ 357 Sheep and goats exempt 0 Pigs 100 2/ 19 Donkeys 50 10 Horses (over 2 years old) 350 17 Total 403 1/ Except for Gaoua and Diebougou where the rate is 150. 2/ Except for Ouagadougou and Bobo-Dioulasso, where the rate is 150. 3/ Based on the 1978 livestock population given in Table 2 and its geo- graphic distribution given in Table 3. For cattle and horses, 70% of the herd is assumed to be 2 years of age or older. No reduction was made for horses owned by Government authorities, which are exempt.  - 48 - Draft oxen are exempt, reflecting the government's desire to encourage the adoption of animal traction, but donkeys--an important secondary source of draft power--are not exempt. Cattle are by far the most important potential source of revenue from the livestock tax. 5.21 Collection rates for this tax are very low. The national budget generally estimates expected revenue at CFAF 137 million, of which about one third is actually collected (see Table 12). In comparison to the potential revenue, only about 10% is recovered. Such tax fraud is hardly surprising, in view of the extensive nature of most cattle herding and given that the penalty is a fine equal to only double the tax. It may be that the tax should be discontinued since its meager revenue may not warrant its administrative costs. Although difficult to quantify, such costs include collecting expenses, inaccurate statistics of herd size, and perhaps a disincentive to more intensive cattle production in mixed farming systems. 5.22 The other direct tax--or head tax--is intended as an income tax on producers, and the rate varies according to herd size: Size of herd 1/ Tax per producer Tax per head of Tax per head of (or herd) (CFAF) cattle (CFAF) 2/ sheep or goats 1/2/ Less than 26 800 31-800 3-80 26-50 3,500 70-135 7-14 51-250 6,000 24-118 2-12 More than 250 10,000 0-40 0-4 Per unit rates, as well as total taxes, vary by herd size. Although the variation is not uniform, the tax is clearly regressive--small herders are taxed more heavily than are large herders. The potential revenue from this tax is almost impossible to estimate, given the lack of information on the distribution of herd sizes. Purely as an order of magnitude, if herd sizes were universally 50 head, the 3.1 million cattle or cattle-equivalents in 1978 would imply a revenue of about CFAF 200 million, which compares to total estimated receipts for that year from all head taxes of CFAF 1.3 billion. The amount actually recovered is also impossible to discern since revenue from the different types of head taxes are not reported separately. 5.23 Indirect taxes - Since the meat tax was suppressed in 1969, the only indirect taxes are levied on livestock and livestock products when they enter or leave Upper Volta. The system of taxation generally relies on assessed values ("valeurs mercurials") instead of actual values, which both 1/ Herd size is measured by numbers of cattle. Ten small ruminants are considered equivalent to one head of cattle. 2/ These figures give the minima and maxima per unit rates. For very large herds, the minimum approaches zero.  - 49 - simplifies the process of collection and gives the Government two ways to change the effective tax rate--altering the assessed value and adjusting the nominal rates. In the past, the Government has usually changed the assessed value. 5.24 Export taxes - Export duties bring in the largest proportion of revenue from the subsector--perhaps over three fourths,l/ and revenues from export duties on livestock constitute the bulk of income from all export taxes (see Table 12). 2/ Export duties consist of five categories of taxes: exit duty, research and conditioning tax ("taxe de recherche et conditionne- ment"), statistics tax, stamp duty (levied on the 3 previous taxes), and a fee that is earmarked for the Conseil Voltaique de Chargeurs to strengthen Voltaic commerce (the "COVOC" tax). These duties are summarized in Table 14. For all goods the research and conditioning tax (0.5%), the statistics tax (3.0%), the stamp duty (6.0%), and the COVOC (0.5%) are identical. Only the exit duty varies; it is 16-17% for livestock and 6-7% for poultry, meat, and hides. Because the tax is levied on an assessed value for all items except,meat, the nominal and real tax rates usually diverge. Nominal tax rates range from 11-22%, but effective rates are between 11-13%, except for sheep and goats. For cattle, the assessed value is less than the real value, which lowers the effective rate. For sheep and goats--and to a lesser extent for poultry--the real value may be less than the assessed value, which would raise the effective rate. Between 1975, when the last major adjustment in assessed values was made, 3/ and 1980, the amounts of taxes on livestock have increased by 250 to 600%, as shown below: Type of Livestock Total increase Annual increase (%) (%) Cattle 261 29 Sheep 623 49 Goats 478 42 Pigs 443 40 Poultry 279 31 These increases have vastly exceeded the general rate of price increase, and they reflect the Government's policy of taxing the export of livestock more heavily to discourage their export. For meat and hides, however, there has been almost no change in effective tax rates, and this stability is also 1/ Revenue from import duties on imported livestock products is not considered as income from the livestock subsector. 2/ Some government revenue from exports accrues through channels other than export taxes (for example, SOFITEX profits paid into the Caisse de Stabilisation). 3/ Nominal tax rates were the same, except that the COVOC did not exist and the stamp duty was not instituted until the middle of that year; it is excluded in calculating the tax in 1975.  - 50 - consistent with Government policies to encourage exports of livestock products rather than animals. As a further incentive to export meat, ONERA is exempt from paying exit duties on its meat exports. 5.25 The effects of these export taxes have never been studied, but in theory they act to depress domestic prices--for both producers and consumers. These lower prices both encourage more domestic consumption of livestock, and depress domestic production, thus tending to reduce the surplus available for export. While helping the Government budget, the duties may on the margin be injuring the country's foreign exchange earnings. Moreover, the shift in relative prices represents a transfer of income from livestock producers to meat consumers --more generally a deterioration in the rural-urban terms of trade. These effects will occur at the margin, and their aggregate impact may be small. For example, given a price elasticity of supply of 0.1, 1/ the elimination of the export tax on cattle would call forth an annual increase in output of only about 1%, or some 3,000 cattle, worth about one-third of the foregone revenue. On the other hand, export taxes from livestock are a small share of total Government revenues (2.5% during 1976-79; see Table 12). Thus, it is unclear whether the certain reduction in Government income would be justified by a problematic increase in output. 5.26 These export taxes also tend to worsen the regional distribution on income to the extent that they depress producer prices and that the poorer northeast and central plateau rely more heavily on the sale of livestock products for income. To illustrate, the seven ORDs of Dori, Fada, Ouahigouya, Kaya, Koudougou, Ouagadougou and Koupela contain about three-fourths of the cattle and presumably supply about the same proportion of government revenues from the livestock subsector. The Sahelian region alone appears to furnish about one-fourth of export revenues from livestock. 5.27 It is also probable that the need to pay export duties encourages clandestine exports. Undeclared exports reduce Government income, and reduce the reliability of statistics, but they do not reduce national income or foreign exchange earnings. The extent of unrecorded trade is unknown in any case, which makes it difficult to estimate the gain in Government revenue from the more stringent border control that is often proposed. 2/ Reduction of export duties would improve statistics, but at a cost of less in Government revenue. 5.28 The recovery of potential export taxes through better control is also difficult to estimate, both because of unrecorded trade and because 1/ The figure is mainly assumed, pending further study. In any case, the long-term elasticity is unlikely to be high if physical constraints are important. 2/ The Customs Service is much less concerned about clandestine livestock exports than is the Livestock Service.  - 51 - revenues from all export taxes are combined in budgetary statistics. Esti- mates of potential revenue shown in Table 12 suggest that for declared exports the non-recovery might be important (e.g., in 1976, estimated tax revenues from livestock exports alone exceeded those actually recorded for all exports by about 50%). 5.29 Export prohibitions - Since 1957, exports of steers and bulls less than 5 years old and of productive cows less than 12 years old have been banned. The prohibition presumably was to protect herd productivity and assure maximum output from the national herd. In fact, it largely accords with the realities of the cattle market and trade: the physical risk and difficulty of exporting smaller, younger stock and the usual practice of herders to keep cows until they reach an age of 11-12 years. However, the slaughter of young animals for domestic consumption makes larger and older animals available for export. Even if enforced, the law probably has little real impact on Voltaic livestock production and herd structure. The export of beef from slaughtered young animals--which other Government policies tend to encourage--actually undermines the purpose of the prohibition. 5.30 Transit taxes - Traditionally, the Government has imposed a tax on livestock transiting through Upper Volta, at the rate of CFAF 500 per head of cattle and CFAF 50 per sheep or goat. Since the change in marketing routes this tax has provided almost no revenue (see Table 12), and as of 1978 it is no longer inscribed in the provisional budget. In principle, transiting livestock with a certificate of foreign origin pay no import or export duties, only the transit tax. 1/ But because a foreign herd must pay export taxes in order to obtain a certificate of origin before leaving its native country, herders may be encouraged to leave neighboring countries fraudulently and to pay Voltaic trade duties. Recent reductions in import taxes on livestock and plans to open customs stations at Markoye and Deou are meant to encourage the legitimate importation of foreign cattle rather than their transit. When these cattle are re-exported, they would be charged the normal export tax. The scheme will increase Voltaic revenue at the expense of neighboring countries depending on export duties in neighboring countries and the physical advantages of passing through Upper Volta. There is also a tax on cattle making their transhumance, but this is col- lected by local authorities. It probably corresponds to the grazing tax ("taxe de stationnement"), which is only CFAF 25 per head for cattle and CFAF 10 for sheep and goats. 5.31 Import taxes - For most livestock products, Government policy has been protectionist. Through 1979, import duties on all types of animals 1/ There appears to be a confusion about transit taxes. Herman cites the increase in transit duties as one reason for the shift in trade routes, and customs officials sometimes claim that foreign herds pay both import and export taxes. In fact, the transit tax has not been altered, and foreign cattle pay Voltaic trade taxes only when they do not have a valid certificate of foreign origin.  - 52 - and meat were much higher than comparable export duties. Meat production is especially protected--import duties are eight times higher than export duties. A major change in policy was adopted in 1980 when import duties on live ruminants were reduced from 37.68% to only 3.68%, a change explicitly intended to encourage declared imports of cattle from northern countries. In addition, the assessed value on imports of cattle was not augmented along with that of exports. As a result, total import duties now amount to only one eighth of the export duties, and livestock production is no longer protected as is meat production. An increase in livestock imports is probably perceived as a means both to augment the size of the Voltaic productive herd as well as a means to augment official exports and Government revenue from them. 5.32 The importance of import duties to the Voltaic livestock economy is probably not great. The high tax previously levied on cattle imports (over CFAF 11,000 on animals worth perhaps only CFAF 50-60,000) is cited as a major reason for changes in trade routes to skirt Upper Volta after the mid-1970s. Cattle imports, at least those declared, can be expected to increase now that this tax has been nearly eliminated. The earnings from the import levy on milk may be important because of the quantity imported but the amount of protection it offers to local producers is small (CFAF 3.5 per liter, or about 5%). Furthermore, most milk imports are food aid, which enter tax-free and distributed without charge. Any project to encourage local milk production and distribution would probably require a change in policies to increase the protection for local producers. Import duties on meat could provide significant protection to local producers if foreign prices are tow enough. At local retail beef prices of about CFAF 500 per kg and with a commercial margin of 30%, the import duties mean that foreign meat must arrive in Upper Volta at only about CFAF 200 if it is to be competitive. This price implies a cost in Abidjan of about CFAF 150, whereas imports of chilled or frozen beef there currently cost roughly twice this amount. If the import duty did not exist, foreign meat could begin to be competitive. Although the duties probably do not cause overall local prices to be much higher--given the small amount of imported meat on the market and the importance of exports in total production--they probably do limit foreign meat imports and thus increase domestic consumption of Voltaic beef. 5.33 Subsidies - Although government policies sometimes cause the transfer of income to livestock producers from other segments of the popu- lation, the government budget provides direct subsidies to the subsector only on the delivery of health services. Currently, all travel costs, equipment, and salaries of animal health (and extension) agents are fully paid by the Government. In theory, vaccines are payable at cost, but the gap between expected and actual receipts of the Livestock Service (see Table 12) indicates that there is less than full cost recovery. Although willing to charge producers for some travel costs, the Government firmly believes that equipment and salaries should continue to be subsidized. In addition, feed produced by the two mills is sold to producers at subsidized prices,  - 53 - but quantities involved are negligible. Improvement in livestock water supplies is also subsidized, but this is mostly an indirect result of the government policy to improve village water sources. 5.34 In terms of total inflows and outflows, the livestock subsector is taxed by the government, although foreign aid more than redresses the balance. During 1976-79, fees and taxes levied directly on livestock provided between 2.9% and 4.2% of central government revenues; consumption taxes paid by herders would raise the percentage even higher. Yet the sector received in return only about 1% of the operating budget (see Tables 12 and 16). Moreover, the degree of taxation has increased. In 1968-69, expenditures were one-third of receipts; 10 years later they were only one-fourth. 1/ 5.35 Investment financed by foreign aid, however, more than offset the outflow of domestic public funds. During the period 1977-79, tax and service receipts projected by the government averaged about CFAF 1 billion per year, 2/ while operating expenditures, subsidies to parastatals, and foreign aid to the subsector averaged about CFAF 1.6 billion annually. 5.36 Trends in domestic financing indicate a weakening ability of the government to finance the subsector. During the 10 years following 1968-69, the real value of expenditure fell by 20%. During the four years 1976-1980, the operating budget grew at the rate of 8.9%, while at the same time civil servant salaries increased by almost 7% per year and annual GDP inflation was around 12%. Over the same period, livestock's shares in both the total central budget and that of the Ministry of Rural Develop- ment also declined, as shown below: 1976 1980 livestock as % of total budget 1.3 1.0 livestock as % of rural development budget 18.4 16.1 5.37 Furthermore, government trade policies now offer little or no protection to the sector except for slaughtering and, by taxing exports, tend to transfer income to urban consumers from producers. The combination of taxes on exports of animals and taxes on imports of meat act to encourage domestic consumption of Voltaic beef, to the detriment of exports. It is unclear that this taxation has significantly depressed livestock production; and as a major economic sector, livestock should be expected to make a net 1/ Data for early years taken from World Bank, The Economic Development of Upper Volta," Vol. 3, "Livestock," Washington, 1970, p. 41. 2/ This figure excludes indirect consumption taxes paid by herders, which are proportionally probably less than the subsectors' 10- 12% share in national income.  - 54 - contribution to government revenue. 1/ Subsidies provided on health services have assuredly increased productivity, although the extremely low output indicates that these subsidies may be insufficient or poorly used. In any case, except for the control of epidemic cattle diseases--for which the social gains generally exceed private gains, improvements in the productivity of herds should be sufficiently profitable to private producers so that subsidies are largely unnecessary. Prices 5.38 Prices are one of the keys to understanding the livestock economy, and, in the absence of solid information on quantities, offer a valuable means for interpreting conditions affecting the subsector. The following paragraphs discuss current prices, their evolution, and efforts by the Government to control them. In general, it should be noted that the Govern- ment has little effective control over prices, and that domestic livestock prices are strongly influenced by foreign prices. Thus, an analysis of the price data, to the extent it can be accurately assembled, probably provides a good picture of the livestock subsector. 5.39 Price levels - Table 8 gives current prices for most livestock and livestock products and Table 7 gives similar information for livestock feedstuffs. These prices are subject to considerable variation, and the ones given represent modal values for the current period. In general, they can be taken as the low end of the range. Sheep and goat prices especially will more than double during celebrations. In most cases, published prices for livestock appear to understate prices given by producers and merchants. To summarize livestock prices, oxen sell for CFAF 170-220 per kg liveweight depending on age, and beef retails for about CFAF 500 per kg. An adult ox thus sells for as much as CFAF 75,000. Small stock sell for much less--under CFAF 10,000. 5.40 With respect to cattle, a few important observations should be made. In general, cattle prices rise with weight and age, the increase per kg between immature and adult stock being as much as CFAF 50, or one-fourth to one-third the price of young animals. This pattern is in marked contrast to the one in more developed countries, where young stock are paid more dearly to compensate for the higher risk and fixed costs of their production. The low prices for which young male stock are sold may mean such stock is considered a byproduct of a herding enterprise concerned with milk output and herd expansion, or that herders have insufficient resources to keep them. 2/ 1/ If the indirect taxes paid on goods consumed by herders were included as revenue from the sector, the sector would appear even more heavily taxed. 2/ For example, both Vengroff and Herman report that the major reason (about 70% of the responses) for selling cattle is to purchase grain; most of the other sales are for clothing and taxes. In Vengroff's study, 80% of the sales are male stock averaging 5 years of age.  - 55 - Low price may also mean there is not much demand for them. In Upper Volta, this price pattern works at cross purposes: while making the use of draft oxen and cattle feeding more attractive--which is consistent with stated government policy--it tends to discourage production and sale of feeder stock by herders--which is essential to the Government's goal of regional stratification of production. 5.41 It also appears that the prices of livestock often double in the coastal markets, relative to Upper Volta. The differences in price are accounted for partly by taxes (10-15%), the high costs of exports, and--in some cases--excess commercial profits, or rents (see para 5.08). Nonethe- less, these large differences suggest that Voltaic livestock are competi- tive. However, SODEPRA (Societe de Developpement de la Production Animale) in Ivory Coast has been selling slaughter oxen at CFAF 250-270 per kg, which may tend to undercut the market for Voltaic exports if Ivorian output expands significantly. 5.42 With respect to different meat prices, there do not appear to be consistently large variations among the various types, given the poor quality of the data. There is some evidence that pork and fish are cheaper meats, especially if most mutton and goat meat is eaten when prices are highest. Unlike prices in developed countries, poultry appears to be one of the most expensive sources of meat, reflecting in part the small scale of production units. 5.43 The fifth quarter of cattle appears to contribute at least 10% to the value of the carcass (before commercial margins). However, the market for offals is fairly inelastic--ONERA's effort to export meat to Nigeria in 1978/79 (equivalent to about 150-175 cattle per week) reportedly reduced the value of the fifth quarter by half. As Staatz, has shown, if meat exports cause the fifth quarter to fall in value in the exporting country, while increasing in value relative to meat in the importing country, meat exports become less profitable. On the other hand, there is currently little dif- ference in the value of the fifth quarter in Upper Volta and Ivory Coast, which tends to make meat exports more attractive--at existing quantities. The problem needs to be carefully analyzed. 5.44 Trends in prices - As seen in Table 10, livestock producers--to the extent their incomes are correlated with movements in consumer prices, have faired better than other sectors of the economy. Over the past decade, beef and mutton prices appear to have gone up more rapidly than both cereal (millet and sorghum) prices and general Voltaic consumer prices. As would be expected, Voltaic beef prices have somewhat followed movements of world beef prices, although domestic prices continued to rise in the mid-1970s when world prices were depressed (see Chart 3). The recent rise in world prices may be one reason explaining why exports of Voltaic cattle have picked up again. More importantly, Ivorian retail beef and poultry prices rose by about 190% between 1972 and 1978, compared to a rise of only about 140% for Voltaic beef over the same period. Thus, prices in the major export market have been rising more rapidly than either world or Voltaic beef prices. Milk prices have fallen in real terms, reflecting the large world surpluses. Thus milk producers will not have benefited from general price increases--although local milk is often bartered in equal volume for millet and sorghum and will have  - 56 - maintained its terms of trade with cereals. Moreover, it appears that local milk often sells for less than the official price for imported, condensed milk, which may give scope for increasing returns to milk producers. On the other hand, high costs of collection may explain most of this differential, which implies that a milk scheme would have to reduce marketing costs, or increase import taxes. Trends in actual producer prices are more difficult to discern owing to lack of reliable statistics, but some indications are available. Average annual change Type of Livestock (percent) Period covered Poultry 7 1970-80 Live cattle, Ouagadougou market 11 1970-78 Ouahigouya market 17 1970-78 Dori market 14 1970-78 Bobo-Dioulasso market 19 1970-78 Draft Oxen 13 1972-79 Live sheep, Ouagadougou market 8 1970-78 Live goats, Ouagadougou market 9 1970-78 Compared to the inflation of general consumer prices over the same period (9% per year during 1970-78), it appears that livestock producers have probably at least held their own. Rising local demand will probably continue to cause local prices to rise, a phenomenon that could eventually force Upper Volta out of the export business. 5.45 Since the rupture in Sahelian supplies of livestock after the drought and the coincidental increase in chilled beef imports from elsewhere, the West Africa coastal market is now fairly strongly influenced by world beef prices. The gap between world beef prices and Sahelian livestock, which have traditionally fetched higher prices in coastal markets, is now diminishing. This suggests that in the future Voltaic export production will be affected not only by trends in world prices but by their level as well, unless the Conseil de l'Entente decides to impose a common external tariff to protect all producers in the union. According to systematic price projections, prospects for world cattle prices are moderately bouyant--in real terms, the world price of chilled beef during the next decade is expected to be slightly higher than in the last half of the 1970s, which was a period of relatively low prices. 1/ The CRED Study of livestock in West Africa concludes: 1/ World Bank, "Price Prospects for Major Primary Commodities", Report No. 814/80, Washington, January 1980, and Memorandum from EPDCE, "Com- modity Price Forecasts - Updating," Washington, July 12, 1982. See also Chart 3. Of course, these projections are sensitive to small changes in domestic policies of major producing and consuming countries, but they are not inconsistent with broad cyclical patterns.  - 57 - "The Sahelian countries may have a relatively good competitive position for a few years and then be severely challenged once again. Sahelian livestock exporters should expect keener competition in the future". 1/ However, others feel that livestock prices are bound to rise, owing to rising world demand. Given the long gestation period for livestock development, it is these long term trends that are most important. If the subsector is to remain profitable, especially in the face of constant or rising costs in Upper Volta, it is imperative that herd productivity be raised. 5.46 Seasonality - There are clear seasonal patterns in livestock prices, although their causes vary by type of livestock. For cattle, prices per kg are lowest at the end of the rainy season, or harvest time (September-October) reflecting the abundant supply of well-fed cattle and efforts to get rid of poorer stock before the dry season. It is also at this period when the value of cattle in terms of grain appears to be highest, and thus the time when most cattle are sold. 2/ Prices peak near the beginning of the rains (May-June) reflecting both the shortage of well-fed stock and the higher demand of herders seeking to rebuild herds in anticipation of better grazing. These price movements are important in making dry-season cattle fattening a profitable venture. For small ruminants, price movements are caused more by changes in demand, rather than changes in supply. Prices rise precipitously before major Muslim feasts (primarily Ramadan and Tabaski). Changes in poultry prices are also caused by similar factors, although the peaks seem to occur at different times. There is less evidence of seasonal variations in prices of livestock products, such as milk and eggs. Feedstuffs (especially fodder) also vary in price, being lowest during the rainy season and mounting steadily as the dry season progresses. 5.47 Government price control - The Government undertakes direct price control by routinely setting official prices for most commodities, but it seldom has the wherewithal to enforce its regulations except when items are bought and sold by Government agencies. However, no livestock commodi- ties fall under the main price control agency--the Caisse Generale de Perequation des Prix des Produits et Marchandises de Grande Consommation. The Government also exercises indirect control by its taxation and subsidy policies described above (see paragraphs 5.19 to 5.37). The official prices reported to be currently in effect are given below: 1/ Edgar J. Ariza-Nino and Charles Steedman, "Synthesis Report", Livestock and Meat Marketing in West Africa, Vol. 1, "Synthesis - Upper Volta", CRED, Ann Arbor, 1980, p. 11. The study projects to the mid-1980s. 2/ See Vengroff, p. 99.  - 58 - Commodity Retail prices Date of adoption (CFAF per kg) Beef-with bones 120 1979 Beef-with no bones 150 1979 Mutton/goat meat-with bones 150 1979 Mutton/goat meat-with no bones 175 1979 Cattle hides 160 1/ 1978 2/ Sheep skins 700 1/ 1978 2/ Goat skins 900 1/ 1978 2/ Eggs 25 1978 2/ Milk-fresh 3/ 130 4/ 1974 Milk-reconstituted 3/ 95 4/5/ 1974 The official prices of skins, hides, and milk appear to be fairly close to market prices, probably because of the importance of milk imports and Government purchases of hides and skins for export. Those for meat and eggs are vastly beneath market prices (see Table 8), reflecting both strong demand and the Government's inability to intervene successfully in the livestock market to control prices. 1/ Producer prices paid at collection markets. 2/ These prices were paid in 1978-79. They may have been adopted earlier. 3/ For milk sold in plastic containers by the dairy at Koubri. 4/ Price per liter of whole milk. 5/ The price of condensed milk (both sweetened and unsweetened) is also controlled, and its current price is virtually identical to this one, account taken of the sugar content.  - 59 - VI. INTEGRATION OF LIVESTOCK AND AGRICULTURE 6.01 Livestock and agriculture have been associated in most of Upper Volta for decades, but integration of livestock with agriculture -- in the sense of mixed farming -- has only reached an embryonic stage. However, the possibility of further integrating livestock, in particular cattle, with crop farming is very good because cattle are fairly well distributed through- out most agricultural areas (for example, the density of cattle around Kaya is almost as great as near Dori -- see Table 1) providing a good basis for a joint development of both animal and crop husbandry. Admittedly, the task will not be easy nor rapid, but the future of Voltaic livestock production probably depends in no small measure on achieving its real integration with agriculture. As the rural population expands, agricultural production will claim an increasing share of the available land, and unless crops and cattle are integrated, it is likely that livestock production may be pushed to the fringes of areas with the best pastures and water. The association of agriculture and livestock currently takes several forms, and the extent of it is rather impressive. Types of Integration 6.02 Cattle parking. The most commonly known, though not necessarily most common type of association, is the interaction between crop farmers and migratory herders on seasonal transhumance. In Upper Volta, there are few true nomadic herders, except for the Sahelian zone, and much of the transhu- ments come from Niger and Mali during the dry season. A simple type of association of cattle with crop farming takes place when transhumance herds eat crop residues on fields following the harvest. Farmers benefit because the dung left by the cattle fertilizes the fields. However, this associa- tion also leads increasingly to conflicts when migrants herds damage crops or eat harvest stores. Ethnic differences as well as the continued exis- tence of transhumance rights make it difficult for the local inhabitants to resolve these differences by themselves. In any case, this type of live- stock production continues to be based on extensive grazing, and farmers and herders do little more than coexist. 6.03 Village herds. The second type of association involves village herds, which are sometimes taken care of by the herding population (Peuhls) even though owned by crop farmers. Increasingly though, farmers take care of their own cattle (see section 3). It is estimated that perhaps 30% of the national cattle herd is owned by farmers (a percentage that was probably greatly increased during and after the 1968-73 drought). In this instance, the herders are usually permanent, or very long-term, residents in the village. They also cultivate crops -- although their primary activity remains herding. Because cattle are kept nearer the village, especially during the crop season, the incidence of conflicts between herds and farmers are greater. Such conflicts usually result from livestock eating crops, from farmers restricting access to watering points by planting fields, and from disputes over whether bottomlands are to be cultivated or reserved for  - 60 - grazing. In the Sahelian zone, where herders predominate, efforts to avoid these conflicts often involve fencing crop fields with thorn bushes to keep out cattle and leaving clearly defined paths to watering points. In the cropping zone, however, the solution has usually been for agricultural exten- sion agents to assess fines to be paid by herders, forcing them to keep cattle farther from unprotected village fields. The struggle for access to pasture near villages leads to continual conflicts between herders and cultivators over crop damages. The fact that the herders are usually charged and forced to pay fines and adjust their herding practices reveals the dominance of the cropping sector in all but the northeast corner of the country. The fact that the conflicts continue to occur reflects in part the lack of a well- defined and enforced policy regarding land use, including the designation of certain areas as access corridors, pasture lands, and stock routes. 6.04 Draft Animals. 1/ The third type of association -- the use of draft animals -- is the only major form of integration between crops and livestock. Farmers own draft oxen and donkeys for use in plowing, weeding and ridging and transport. Available data on draft livestock and sales of equipment indicate that some 50,000 adult male oxen are used for draft purposes (see Table 1). This figure, probably a low estimate, represents about 10% of all adult male (bulls and steers) oxen and 3% of all adult cattle. The use of animal traction (both oxen and donkey) has grown steadily since the mid-1960s. The number of ox teams appear to have increased from a mere 1,000 nationwide in 1965 to at least 26,000 in 1979, which gives an annual growth rate of about 25%. 2/ Lack of credit may be a major constraint to expansion, but as much as 35-40% of credit from the newly formed CNCA is expected to be used for animal traction. 6.05 Keeping draft animals obliges farmers to ensure a proper feeding of the livestock throughout the year, which is a first step toward mixed farming. Some type of fodder cultivation and preservation is necessary in order to keep the animals fit for work; feed supplements are usually re- quired during the working season, and careful veterinary care is needed. As a direct result of the growth in the use of draft animals and feeding small ruminants, fodder (in the form of legume tops and cereal stalks) is now sold in all the major agricultural markets. Some farmers may be planting late- season sorghum expressly to sell or use as fodder, although many observers argue that farmers usually lack labor, and often land, to grow much fodder and 1/ A more detailed discussion of this issue is in the Agricultural Issues Study, World Bank, Washington, October 29, 1982 (Report No. 3296-UV). 2/ Based on data in Chapter 1 of the Agricultural Issues Study. Not all studies are so optimistic however. See for example Christopher Delgado, "An investigation of the Lack of Mixed Farming in the West African Savannah: A Farming Systems Approach for Tenkodogo, Upper Volta," Discussion Paper No.74, Center for Research on Economic Development, University of Michigan, Ann Arbor, November 1978.  - 61 - that most efforts to encourage it have been negative. 1/ Other agricultural by-products in the traditional sector -- such as cereal bran and draff from millet beer (dolo) -- are already often being marketed as livestock feed. The animal traction center at, Bulbi, 2/ south of Ouagadougou, demonstrates how to store and use forage crops and how to make silage out of brewers' grains. This initiative merits attention as it could furnish valuable information for an eventual cattle feeding and fattening project. 6.06 Cattle fattening. The fourth type of integration is intensive cattle feeding. This effort is part of of the strategy to stratify live- stock production, making the north the zone for raising feeder stock that are then grown out in the south where feedstuffs are more plentiful (see Section 9). Although fattening old draft oxen is a logical outcome of using draft animals, there is also a small but growing interest in growing out and finishing cattle during the dry season by small-holders, although no estimates of numbers are available. These household cattle (boeufs de case) are an entirely indigenous effort, carried out without either knowledge of or support from the government. This activity has also been coupled with the "manure stables" (etables fumiers) funded by the RDF. In contrast to traditional, dry-season feeding, the RDF program lasts for 1-2 years. ONERA currently operates one modern feedlot, which purchases rice straw and agro-industrial by-products, but its impact on the integration of farming and livestock is both negligible and indirect. 6.07 Small Livestock. The fifth type of integration involves the ownership and feeding of small stock by farmers. Ownership of sheep and goats is almost universal, in part because small ruminants offer a first stage in a strategy for reinvesting profits from crop production. Some estimates place the average herd size at as many as 10 per farm family. Chickens and pigs are produced largely by farmers, although some urban households also engage in this activity. While an important source of family income, small stock are probably less important to mixed farming than are cattle. Advantages 6.08 There are numerous advantages to a shift from extensive to inten- sive livestock raising, for both the livestock and crop subsectors. Weight losses during a long dry season would be turned to weight gains by a regular, daily feeding (for example, total annual gains could be as high as 100-125 kg, compared to 25-30 kg estimated for traditional grazing). Elimination of 1/ For example, see R. Vengroff, Upper Volta: Environmental Uncertainty and Livestock Production, International Center for Arid and Semi-arid Land Studies, Lubbock, 1980, pp. 74 and 140. He also notes, however, that one-fourth of farmers or herders providing such supplementary feeding buy a part of their feed supplies. 2/ Projet Assistance aux Cultures Attelee, a FAO-Danish supported project for animal traction.  - 62 - these dry seasons weight losses would appear to represent a net physical gain in national resources and could be an important step in raising the produc- tivity of the livestock subsector. It would also alleviate malnutrition, leading to better animal health and fewer losses from disease. However, the costs of feed for the dry season are high (see Table 7), and the economic profitability of dry season feeding must be carefully calculated. 1/ The existence of the activity in the traditional sector is some evidence, though, that the activity raises economic productivity as well as physical productivity. 6.09 The availability of manure would increase crop yields, and already there is a nascent market in manure in some areas. A draft ox produces about 2.5 tons of manure per year, or 50 kg of N -- worth over 10,000 CFAF at current world urea prices would and more than twice this amount at local market prices for manure. Although the evidence is ambiguous, cultivation with ox-drawn equipment may help reduce soil erosion and increase crop yields. 6.10 Farm family labor could be better utilized throughout the year, because herding is a year-round activity. 2/ Although the evidence is again unclear, draft animals may save as much as 40 days of farm cultivation labor (plowing and weeding), worth perhaps 14,000 CFAF. In addition, transport is easier and may even augment family income. Promotion 6.11 Further integration of these activities requires that several constraints be overcome: (a) Cost (in terms of labor time) of guarding animals, which rises as more areas are cultivated, and dry-season crops are increasingly grown. Three specific actions may help cope with this problem: designating limited, crop-free areas--especially village cattle paths; providing cheaper fencing materials (including plants like sisal) for corrals and fields; and encouraging use of animal traction to reduce labor needed for weeding and for transport at harvest. 1/ For example, the compensatory weight gains that are put on quickly after the first rains bring forth new pastures cost very little, especially if pastures are abundant and dry-season feedstuffs are expensive. 2/ Some observers argue that keeping oxen is too labor-intensive and not worth the value of family labor -- especially during the cropping season. For example, see Christopher L. Delgado, "The Southern Fulani Farming System in Upper Volta: A New Old Model for the Integration of Crop and Livestock Production in the West Africa Savannah," CRED, Ann Arbor, September 1978.  - 63 - (b) Relatively low return on intensive cattle raising. To increase productivity, focus should be placed on the following: (i) additional returns from ox traction (including higher yields, labor saved in weeding and harvesting and use of ox-drawn carts), as a way of offsetting higher herding costs; (ii) trade-offs between herding cattle and other ruminants; (farmers already have herds of sheep and goats that they care for); (iii) extra value that can be realized from increased use and sale of cattle manure; (iv) profitability of fattening or finishing cattle both with and without ox traction; (v) prospects for expanded consumption and sale of milk, as well as prospects for improving milk productivity of herds; and (vi) viability and costs of forage production, as a principle crop, as improved fallow, and as a by-product of cereal production (relevant issues include the need for and cost of fencing, availability of suitable forage varieties, relative labor requirements of herding, collecting forage and cultivating fodder, and problems of conserving hay). 6.12 Except for the promotion of animal traction, the government has no clearly formulated policy for associating animal husbandry with agricul- ture. The livestock service is concerned mainly with prophylactic treatment of cattle, and the ORDs concentrate on promoting crops and animal traction. There is currently no viable livestock extension service, and none of the agricultural projects is actively promoting a mixed farming system. Nor are experiments and research dealing with aspects of mixed farming. However, some attempts are undertaken by an ongoing FAO/CILSS project. The spread of a mixed farming system, including farm fattening of cattle and perhaps dairying, can only be achieved through an efficient livestock extension service with emphasis on training of livestock farmers. In addition, relative prices must be such to make intensive livestock production profit- able, and farmers must recognize the value of mixed farming in improved soil fertility and reduced farm labor. 6.13 As more livestock are kept on farms, especially for growing out, fattening, and draft power, the livestock extension service must be re- oriented to provide both curative veterinary treatments and advice on animal husbandry practices. There will also be a growing need to strengthen the integration of livestock feeding with agriculture, by both improving the  - 64 - quality of forage on fallow land, maximizing the use of agricultural by- products and crop residues, and increasing the use of manure as fertilizer. The agricultural extension service must also be reoriented to furnish the appropriate advice. Finally, the system of agricultural cooperatives needs to be altered to encourage the participation of herders.  - 65 - VII. SUPPORT SERVICES Government Institutions 7.01 There are numerous governmental institutions that provide impor- tant services to the livestock subsector. Some of these deal solely with livestock, while others are primarily concerned with agricultural or rural development. Most of them furnish services and inputs--such as extension and veterinary services, training, research, and credit--that would probably not be provided by the private sector. However, the national veterinary pharmacy and the national livestock marketing office play roles that are largly commercial and as such may compete with merchants and producers in the private sector. Although it is difficult to generalize, public institutions and management are inexperienced and lack authority, funds, and information; their poor condition constitutes an important constraint to programs designed to improve livestock production in Upper Volta. Furthermore, some reorgani- zatioii of governmental livestock institutions will probably be necessary to adapt the services they provide to producers. 7.02 The National Livestock Service. The Department of Livestock Ser- vices and Animal Industry, 1/ which is under the Ministry of Rural Development and located in Ouagadougou, has been traditionally concerned with livestock development at the national level, but has dealt mainly with animal health-- limited largely to the vaccination of cattle. The department contains several services such as (see the organigram, Annex I): - a secretrial and accounting service, - a statistical and library service, - a training and extension service, - a planning and studies group, - an animal health service, - a diagnostic laboratory including veterinary research, - a veterinary pharmacy (PHANAVET), - an animal production and research service, - a poultry development service, - a trypanosomiasis research section, 1/ La Direction des Services de l'Elevage et des Industries Animales.  - 66 - - a service to carry out agro-pastoral and hydro-pastoral works, and - a service to assure sanitary conditions in livestock industries. The director of livestock executes instructions from the Minister of Rural Development and has under him all the staff of the service. He is respon- sible for all relations with his ministry, the local administration, the national technical services and international technical organizations. He assumes the coordination on the national level of programs concerning livestock research, animal production development, meat and food inspection, and animal diseases. 7.03 The livestock department also assists and helps supervise various organizations that deal with animal production, such as the ORDs, AVV, the livestock school, the poultry stations, the tannery, the hides and skin companies, ONERA, the West Volta Livestock Project (the Bank's first live- stock project), and IEMVT. 7.04 Veterinary Field Service. The service is oriented towards pro- phylactic treatment of cattle and deals little with the other problems of livestock production, which should include all species of domestic animals, their proper feeding, and the curative treatment of diseases. Although the present focus of the service is a natural outcome of its past concern with controlling epidemic diseases, it is important to set up in the future an efficient livestock extension service dealing with all aspects and all species of domestic animals. 7.05 Before 1974, the veterinary field service was directly under the national director of livestock services. The country was then divided into seven veterinary districts ("circonscriptions"), which were headed mainly by assistant veterinarians. Each veterinary district comprised several veteri- nary sectors which in turn comprised several posts headed by veterinary nurses (infirmiers veterinaires). Following the reorganization in 1974, the eleven ORDs took over the responsibility for the field services. The ORDs were to provide administrative and logistical support; and funds for the operation of the veterinary service were to be channeled through the ORDs. The national livestock service would continue to furnish the technical direction and supervision of the field services. This change was made in order to bring the veterinary administration in line with the general administrative structure for rural development. The present veterinary districts, sectors and posts are shown in Table 17, and the distribution of posts is also shown on the attached Map. 7.06 In mid-1980, the livestock service was again given direct respon- sibility for its veterinary field services, putting animal health care out- side the ORD system. Although the livestock service has responsibility for technical supervision of animal production and extension programs, these will, however, continue to be implemented through the ORDs.  - 67 - 7.07 The quality of the livestock service under the ORD system has been disappointing. It has been unable to produce either reliable statistical information or annual reports that even describe the work of the service. It has had difficulty in creating an efficient service for livestock owners. Except for a few ORDs where rural development projects are financed by foreign donors (e.g., Bobo-Dioulasso), the service probably has little impact on livestock development other than through vaccinations. In some ORDs (e.g., Fada) the service has not even been able to maintain adequate vaccination coverage of cattle and regular vaccination campaigns against contagious diseases have ceased. In addition, the integration of the livestock service into the ORD system has apparently not been fully carried out, as the live- stock services remained separated from the other ORD services. The livestock field services have alleged that fewer funds and resources were made available to them after the responsibility for budgetary allocations was shifted from the national livestock director to the ORDs--with their bias toward crop cultivation. 7.08 Besides poor institutional arrangements--which have perhaps now been ameliorated--there are other reasons for these deficiencies, including inadequate funds, poorly trained staff, and inadequately developed techniques. Budgetary allocations to the livestock service have been disproportionately small, and there are insufficient funds for investment and operating expenses. Between 1976 and 1979, total personnel increased by only 13 field agents, or 6%; in 1980, the number declined. The government's policy of charging herders only for the vaccines--and not their delivery--means that investment in and operation of all veterinary equipment must be wholly subsidized by the govern- ment. Except for foreign-financed investments, it has been unable to pay these costs. In most ORDs (or circumscriptions) the service lacks almost everything needed for veterinary services, from vehicles to medical equipment. As a result, the present veterinary field staff is grossly under-employed. 7.09 Even if funds were available to improve the veterinary service, most field staff need retraining to carry out livestock extension work other than vaccinations. The staff lack diagnostic capability, are unfamiliar with most animal production aspects, and have no knowledge of range management. Staff at higher levels (in veterinary medicine) are almost all concentrated in the capital and only a few are based in livestock areas. 7.10 Even if funds and staff were not constraints, extension programs would probably be limited by the lack of past work on how to improve animal husbandry in traditional settings and how to improve pastures. But at the margin, some improvements in curative treatments and animal husbandry would be beneficial. 7.11 National Veterinary Laboratory. The laboratory--located in Ouagadougou--was created in 1960 as a technical division of the Livestock Service in order to diagnose animal and related human diseases and to produce vaccine for rinderpest. Financial austerity measures introduced in 1962 reduced both the autonomy of the laboratory and its ability to diagnose diseases. In addition, new vaccines against rinderpest were developed that  - 68 - the laboratory could not produce. Because its operating budget is insuffi- cient, the laboratory can no longer carry out diagnoses in the field and much of its activity is now limited to analyzing samples sent from the hospital in Ouagadougou. A staff of fourteen includes 3 veterinarians and 1 biologist. Major problems cited for the laboratory include its small and poorly adapted facilities, its lack of a staff with a full range of compe- tencies, its low operating budget (estimated to be at only 10-20% of what would be needed to assure reasonable operations), and its lack of vehicles to enable staff to travel where there are outbreaks of disease. In 1979, German aid financed a 2 year project (115 million CFAF) to equip a veteri- narian to study parasites in small ruminants. Another German project will improve and enlarge the laboratory's physical facilities and provide two more persons, making the staff complete except for a virus expert. Future plans include the production of poultry vaccines, for which financing from the FED has been requested but not yet received. Despite recent aid, the problems of inadequate operating funds and means of travel remain. 7.12 PHANAVET. German Aid gives support to the livestock department through a project at the national veterinary pharmacy, PHANAVET. The project, which started in mid-1978 and will last until April 1981, is headed by a German veterinarian. Total project costs, including an important stock of drugs, are CFAF 210 million. Each veterinary district (one per ORD) has been provided with a stock of veterinary drugs equivalent to CFAF 3 million. Later, it is intended to provide each veterinary post with a stock of drugs. A revolving fund is supposed to be established by revenues from sales of the veterinary supplies. Because farmers and livestock owners are supposed to pay for drugs the veterinary pharmacy should operate in a sound business manner at no cost to the Government. But there is little evidence that workable revolving funds have been established or are being maintained--in part at least because of poor management and funding pressures at local levels. 7.13 The Animal Production and Research Service. Although always part of the livestock service, the headquarters was located in Bobo-Dioulasso before 1972-73, after which it was consolidated with the rest of the service in Ouagadougou. The service has no separate extension agents of its own, relying on the personnel of the livestock health service. In general, the service is charged to carry out the following activities: - improving livestock breeding, husbandry and feeding practices; - promoting feeding programs for ruminants; - introducing foreign races of livestock; - providing technical assistance to ORDs in animal production programs; and - managing livestock research stations.  - 69 - Since the 1968-73 drought, the animal production service has been given greater emphasis, and within the service itself the focus is much more on better feeding than on breeding improvement. The service is currently organized into three subsections: cattle, small stock, and research. The service has several shortcomings. It does not have, nor is it expected to have, subsections for animal traction. Without separate agents trained in animal husbandry, it is unlikely to be able to carry out its stated activities. Finally, because there are no funds to carry out research, the service does little more than simply look after the herds on its research stations. 7.14 The Poultry Development Service. The service is responsible for all aspects of both traditional and commercial poultry production. The service also has responsibility for the operation of the poultry stations in Ouagadougou and Bobo-Dioulasso. Parastatals 7.15 The National Poultry Station, (Centre National Avicole), is located in the vicinity of Ouagadougou. The station was created in 1965 and given a parastatal status. The station's objective is to provide day-old chicks, feed- stuffs, and advice on modern poultry raising to farmers in the vicinity of Ouagadougou. The station also produces eggs and broilers for consumption. In 1978 it produced about 32,000 day-old chicks. These were not enough, however, to satisfy the demand; and 43,000 more chicks were imported from France. The station also operates a feedmill which produced about 1,000 tons of con- centrates and feedstuffs in 1978. The feed is mainly for its own chickens and layers, but an additional 375 tons were produced and sold to poultry farmers. The poultry station in Bobo-Dioulasso was rebuilt in 1979 with a hatchery, new poultry houses, and a feed mill. The main purpose is to improve the genetic quality of poultry in the southwestern part of the country. The station also provides day-old chicks to poultry farmers near Bobo-Dioulasso, and the staff advises farmers on modern poultry techniques. 7.16 ONERA, an autonomous agency, responsible to the Minister of Rural Development, was set up by Government in 1976 to regulate and develop live- stock and meat marketing nationwide. It deals with agricultural byproducts and animal feeds, studies the meat trade and prices and the market situation outside Upper Volta, and gives assistance to cattle traders through training and information. It is responsible for the ownership and operation of the modern marketing infrastructure, modern processing facilities, and intensive feeding operations. As such, ONERA manages stock routes, improved markets, and holding grounds. Recently it has taken over ownership and management of the Ouagadougou and Bobo-Dioulasso abattoirs and cold storage facilities, the Banfora feedlot that was started through a German-assisted project, and, since 1979, the feedmill in Bobo-Dioulasso. ONERA is divided into several divisions for animal production, industries and marketing. It has been given a legal monopoly on the distribution of agro-industrial byproducts-- especially 3rd quality cotton seeds, which are in short supply. These are distributed via its animal production division to various livestock projects  - 70 - and via the ORDs to livestock owners. ONERA receives technical assistance from Dutch and German aid; and the Dutch have financed a new headquarters in Ouagadougou. Other studies undertaken recently by ONERA include an FAO poultry marketing study and a dairy study. 7.17 It is important to note that ONERA increasingly appears to be playing a more direct role in livestock production and marketing (ownership of cattle) as well as duplicating the animal production functions of the Government Livestock Service. By competing with rather than working with the private commercial sector, it is seen by some local merchants as going beyond its original primary mandate to assist livestock marketing. As a result, ONERA is turning into a state enterprise with power to divert public funds (both budgetary and foreign aid) from the Livestock Service to its own ends, which are not necessarily consistent with the livestock strategy elaborated by the government services. As a minimum, the role of ONERA in the livestock sector should be clarified. Preferably, the Livestock Service should retain responsibility for planning and coordinating the use of public funds for animal production, and ONERA's activities should be limited to improving the existing private commercial sector. 7.18 Mineral Block Factory. A small factory (Atelier de Pierres a Lecher) for manufacturing mineral blocks for livestock has been built recently in Kaya under a project financed by Belgium through the AIDR agency. The factory will be handed over to the Government in 1980 and managed by ONERA. Total invest- ment costs were about CFAF 13 million (US$65,000). Daily production is 500 kg in one shift, and annual production in 1979 amounted to about 100 tons. Annual production could be doubled, but the production is still far below the annual demand of about 2,000 tons, which is met by importation of Dutch mineral blocks. The actual consumption only covers the need for mineral salts for about 250,000-300,000 head of cattle, a small fraction of the total livestock population. Two qualities of mineral blocks are produced and sold at cost (CFAF 110-130 per kg). Blocks are resold by ONERA at a small profit. The quality of the locally produced blocks has been inferior to the imported blocks due to some manufacturing problems that caused them to crumble; imported blocks seem to be preferred by herders even though they are more expensive (CFAF 160/kg). Data were unavailable to assess the profitability of the operation. If domestic production of mineral blocs is to be expanded in the future, careful consideration must be given to the type of manufacturing facility that can produce a high quality product without subsidies. Despite the attractiveness of creating more decentralized small-scale workshops, a single installation may be better for technical, as well as perhaps economic, reasons. 7.19 Feedlots. The one feedlot currently in operation near Banfora was established through a German financed project and is now managed by ONERA. It has a capacity is 600 head of cattle 1/ and the average fattening period is 6 to 7 months, although some animals are kept as long as 9 months. 1/ In four paddocks.  - 71 - In 1979 the feedlot was stocked with 300 to 320 head and the annual through- put about 750 head of cattle. ONERA intends to increase the annual through- put to 6,000 head. Cattle are fed molasses, cottonseed and rice straw. 1/ The feedlot has only a small 500 ha pasture area, but the main feed supply is agro-industrial byproducts. This operation has apparently had no problems marketing its finished steers. 7.20 Another modern feedlot with a theoretical capacity of 3,600 head per year was operated for a few years by SODEXPAD but is now closed. The operation allegedly failed because there was insufficient demand for fat- tened animals. Two other reasons provid,e more likely explanations for its failure. By attempting to operate through the rainy season, the feedlot suffered higher losses, failed to achieve high daily rates of gain, and was unable to benefit from seasonal price movements. The alleged marketing problems resulted in part from offering animals for sale at the end of the rainy season when there was an abundance of grassfed cattle on the market. Other Agencies 7.21 Regional Development OrganizatLons (ORDs). Eleven ORDs were created in 1965 as independent entities--paralleL to the government's political struc- ture--to promote rural development in their areas. At the central government level, the Ministry of Rural Development gives technical support to the ORDs and coordinates and supervises the main activity, which is the agricultural extension service. In 1974, the focus oE ORDs was changed to community development; and between 1975 and 1980 the ORDs had responsibility for carry- ing out the veterinary field services. The creation of the ORDs represented an effort to decentralize rural development efforts. They were designed to be self-financing through their marketing of crops at official prices. Both because of poor financial control and because government price policies are established by the central government irrespective of ORD finances, this principle of financial autonomy for the ORDs has not succeeded. Although they receive some funds from the central government budget to pay for basic facili- ties and administrative personnel, the ORDs usually try to supplement these funds with project financing to pay for equipment and operating costs. Because of the unpredictable nature of these various sources, most ORDs continue to encounter severe financial d:Lfficulties which constrain their effectiveness. 7.22 The Government has had under consideration for some time a reform of the ORD system to improve extension for agricultural production, to alleviate and prevent heavy indebtedness, and to control the existing bad management. The required reform has 3 major elements: (a) ORDs are to give unequivocal priority to increasing both vegetable and animal production; (b) the legal status of ORDs is to be changed to one oi- public administration, eliminating their financial autonomy; and (c) formal village groups are increasingly to provide the link between producers and the ORD technical services. With 1/ Daily ration per animal is 2 kg cottonseed, 4 kg molasses and 4 kg rice straw.  - 72 - respect to livestock, veterinary services are to be completely separated from the ORDs by the recreation of the livestock circumscriptions that will be wholly controlled and financed by the central Livestock Service. Animal production will remain an activity of the ORDs, although the Livestock Service will be expected to provide programs and supervision to assure that targets for animal production are realized. At least 3 difficulties must be worked out before the reform can have much effect: (a) how to coordinate animal production programs with agricultural production programs; (b) whether the Livestock Service itself must be reorganized to handle the dual structure proposed for health services and animal production; and (c) how to provide sufficient budgetary resources to carry out the ORD programms and to allow the central services to exercise adequate control. 7.23 The AVV was created in 1974 to organize settlement in the White, Black and Red Voltas in areas freed of onchocersiasis. At present about 1,300-2,000 families have been settled, mainly in the White Volta area. AVV has a livestock division which deals with livestock development among farmers interested in livestock production, including animal traction and fattening; it receives technical assistance mainly from Holland and France. There are livestock extension programs in various areas of the Voltas such as Linoghin (40 km south of Ouagadougou), Mogtedo (south of Koupela), Bane (south of Tenkodogo), Kaibo (near Manga) and in Bougouriba (10 km from Diebougou). In Bougouriba there are 10 extension agents who look after 1,200 to 1,500 pair of oxen. AVV's livestock service operates independently of the existing regional veterinary services and it carries out all veterinary treatments including vaccinations and deworming of cattle. 7.24 RDF. In the past few years, the Rural Development Fund has provided credit for constructing manure stables, usually at only 5% interest, repayable over 2-3 years. The goal of the RDF, however, has been less to provide credit than to give technical assistance, and its envelope has not been large. In the future, most of its credit will be channelled through the CNCA. 7.25 CEBV, the Economic Community for Meat and Livestock, was created in 1970 in Abidjan by the Heads of State of the Conseil de l'Entente coun- tries (Ivory Coast, Benin, Upper Volta, Niger and Togo). The main purpose of CEBV is to deal with all problems concerning production, processing and marketing of livestock and meat within member countries, which finally should lead to a common market for livestock and meat. The member countries would provide the community with all statistical information on livestock, animal production, commercial movements of livestock across borders, the evolution of supply and demand, processing, and the animal health situation. The collected data would be used in formulation of programs for improvement of marketing conditions and the animal health situation in member states. A number of agreements on marketing, payments, and similar matters have been set up between member states; and custom, health and tax legislations have been harmonized. The Executive Secretariat, which is located in Ouagadougou, is in charge of carrying out the community's work program. Main donor agencies are FAC and USAID.  - 73 - 7.26 CILSS. Following the Sahelian drought, the CILSS was created in 1973 to coordinate the development and drought control activities of its 8 member states 1/. The Executive Secretariat of the organization is located in Ouagadougou; it is responsible for preparation and financing of projects and programs. The secretariat has a number of divisions dealing with various subjects such as plant production, forestry and ecology, human resources, transport, livestock and fisheries. Later, in 1976 a "Club des Amis du Sahel" was created to support the work of the CILSS and to inform the international community on the prospects and needs of Sahelian development. The CILSS currently is receiving technical assistance from a number of countries, and two livestock specialists are working with livestock development projects. A FAO agrostologist is presently carrying out fodder crop research in the Sahelian zone. 7.27 Liptako-Gourma Authority was created in 1970 as an interstate regional development authority for the 370,000 km2 in the Liptako-Gourma area. 2/ The entire region lacks all types of basic infrastructure, and the authority has completed a number of studies for roads, railways, telecommuni- cations, agriculture and livestock. One of the main economic resources is livestock, and one study, "Developpement Integre de la Region Liptako-Gourma', deals in detail with livestock development. So far, no projects have been financed. Training 7.28 Training at the higher level is biased towards veterinary medicine. There are only a few livestock specialists (Ingenieur d'Elevage) and very few animal husbandry specialists. For example, in 1980 there was only one animal husbandry expert among the 31 veterinarians and livestock specialists employed by the livestock service. Also, training in agrostology and range management is almost nil. If productivity is to be increased--a goal of government strategy--much more emphasis must be given to local and foreign training in animal husbandry. And if pasture resources are to be resources are to be preserved and improved, more training will be needed in agrostology. 7.29 In the past, livestock training at the lower level (Infirmier Veterinaire) has also been biased towards disease control, in particular prophylactic treatment of cattle. No aspects of animal husbandry or feeding were included in their training. However, the agricultural extension workers from the Matourkou agricultural school received a polyvalent training that 1/ Cape Verde Islands, Chad, The Gambia, Mali, Mauritania, Niger, Senegal and Upper Volta. 2/ The area is composed of parts of Upper Volta (the east), Mali and Niger. The Niger River forms the major portion of the drainage system and the region extends from the Saharan zone (100 mm rainfall) to the Sudanian-Guinean zone (1,100 mm rainfall) through the Sahelian and Sudanian zones.  - 74 - included animal husbandry subjects. Despite the sufficient quantity of agents that are being trained, the impact on livestock husbandry has been negligible, in part because the training has been inadequate. Consequently all the present veterinary field staff need retraining to improve their competence in animal husbandry, care, and feeding. 7.30 At present, there are 27 Voltaic veterinarians, all of whom were trained abroad (mainly in Canada, France, Belguim, Bulgaria and Senegal). In 1980, about 5 new veterinarians will return from abroad (2 from Russia, 2 from Senegal and 1 from Belgium), but the number will decline in the coming years. There are a number of expatriate veterinarians working in the country, mainly in Bobo-Dioulasso, where they are involved in tse-tse research. 7.31 The Ecole Nationale de l-Elevage et des Industries Animales (ENESA), or Livestock School, dates back to 1957 when the school was created in Oua- gadougou under the name Ecole des Infirmiers Veterinaires. In the beginning, a one-year training course was given; it was later extended to a two-year training course with 10 registered students. In 1968, the school was trans- ferred outside town and moved into new buildings which included a clinic for practical training. In October 1979, the school was reorganized and renamed ENESA; instead of training only livestock nurses (infirmiers veterinaires) the school now has two divisions and trains Assistants d Elevage (formerly carried out in Niamey) and Agents Techniques. Students for Assistant d'Elevage are enrolled directly when they receive a BEPC (Brevet Elementaire Premier Cycle) or are Agents Techniques with a 5-year practical training. Yearly enrollment is 10 students, and the training lasts three years. Students for Agents Techniques are enrolled from secondary schools when they receive a BEPC. There are 20 students enrolled annually and the training lasts two years. Training in animal husbandry subjects, which was formerly done at the Matourkou school, has now been transferred to ENESA. Due to the reorganiza- tion, the school is overcrowded; at present there are about 60 students in a school originally constructed for 20 students. 7.32 The Institut Superieur Polytechnique, created in 1973, also teaches livestock subjects in a 5-year course leading to a degree as Ingenieur d'Elevage. The training is more production-oriented than at ENESA and graduates may get jobs with ONERA and other development institutions. So far, only a few students have graduated; in 1980 the number is expected to reach five, in 1981 only one, and in 1982 about 9 students. 7.33 The Matourkou agricultural school, "Centre Agricole Polyvalent de Matourkou" near Bobo-Dioulasso, has since its creation in 1963 given an interdisciplinary training in agriculture and livestock. Its third year training used to include livestock subjects, but training in livestock will be phased out in 1980 owing to the reorganization of the livestock school in Ouagadougou. 7.34 ELAT, the tse-tse control school in Bobo-Dioulasso, trains live- stock agents from francophone African countries in tse-tse control. Courses last 9 months, and the first batch of 20 students began in 1977. The school is operated and financed jointly by France, Germany and Upper Volta at a  - 75 - yearly cost of about DM 1 million (CFAF 100 million). The curriculum includes classes on theoretical subjects, as well as laboratory and practical field work such as surveys and the organization of eradication campaigns. Research 7.35 Total research effort for livestock in Upper Volta amounts to some 215 million CFA francs per year (in 1980), over 90% foreign, and about 15 professional researchers, of which four-fifths are foreign. 1/ Of all agri- cultural research in the country, livestock accounts for roughly 15%, somewhat less than its relative share in national income. Over the years much research on improving animal breeds and forage was started by Government and various aid agencies, but follow-up at the farm level is lacking and consequently impact on farms has been small. 7.36 Markoye Livestock Station is located in the Sahel north of Dori. The 3,000 ha station was set up in 1966 with assistance from USAID at a high cost of CFAF 200 million. The purposes of the station included working with traditional livestock owners in the area to improve their animal husbandry practices and carrying out experiments on pasture management and fodder crop production. Due to lack of funds in recent years, all research on pasture and livestock has ceased and most infrastructure has deteriorated. Current work is the multiplication and dissemination of breeding stock (Azaouak cattle and Maradi goats imported from Niger) to livestock owners. This initiative has not been very successful as only few livestock owners have purchased breeding stock from the station. Unless the stat:ion can be utilized for research purposes it should perhaps be abandoned;. its use as a breeding center alone is probably not justified. In conjunction with a well-defined research program, it could also serve as a support for livestock extension in the area. 7.37 Samandeni Livestock Station, located north of Bobo-Doulasso, was esablished to study the behavior of cattle in the southwestern part of the country, including a breeding program for N'Dama cattle. The station was neither given sufficient funds nor adequate staff. At present, the station's trypanotolerant herd of N-Dama and West African shorthorn cattle serves the IEMVT trypanotolerance project. 7.38 Banankeledaga Pig Station, also located north of Bobo-Dioulasso, was once a pig breeding station for disseminating improved pig breeds (Large Whites) to local pig farmers. In 1976 the station was closed by a brucellosis infection that forced destruction of the herd. Lack of funds has prevented its reopening. There is a need for improving the feed conversion of Voltaic pig breeds. If research could be defined and funded, the station might be reopened, which would also enable it to serve the pig farmers in the Bobo- Dioulasso area. 1/ These figures cover the existing livestock stations, the veterinary laboratory, the research on trypanosomiasis, and the CILSS/FAO fodder project.  - 76 - 7.39 Despite the underutilization of the three existing research stations, the government has proposed the creation of three new stations: one for animal husbandry at Sonore-est and two for small ruminants, one each in northern and southern regions. Until present stations function adequately and long-term financing can be assured for these new stations, they should probably be discouraged. 7.40 Biological Tse-tse Control. Since 1974 Upper Volta has been host country for biological control of tse-tse flies, a program which started in the Central African Republic. The program is located in Bobo-Dioulasso at the Research Center for Animal Trypanosomiasis (CRTA) and carried out by IEMVT under financing by France and Germany (CFAF 900 millions over 5 years). The technique of releasing sterile males is environmentally acceptable but the cost of producing sterile males on a large scale is still too high. A new strategy that uses a combination of biological and chemical control is expected to be tested in the context of a proposed pastoral development scheme at Sideradougou, beginning in 1980. 7.41 Trypanotolerance Project. A new four-year trypanotolerance project, financed by France and Germany (CFAF 500 millions) and carried out by IEMVT, was started in 1978 at CRTA in Bobo-Dioulasso. The purpose is to study the phenomenon of trypanotolerance and to find ways of enhancing it by immuniza- tion or breeding. 7.42 Fodder Crop Research. During many years, forage research has taken place at the IRAT agricultural experimental station at Farako-Ba near Bobo-Dioulasso. Seeds of various leguminous plants and grasses are produced and sold to livestock projects throughout the country. Production in 1978 included about 400 kg of Brachiaria seed and about 1,000 kg of Stylosanthes seed. Research with irrigated fodder crops is carried out by CERCI, based at the experimental station with field experiments in the Kou valley. Feed trials with work oxen have also taken place in the Kou valley where oxen, during the dry season, were grazed on improved pastures. 7.43 Fodder crop research in the Sahelian zone is at present carried out by a FAO-supported CILSS project (CFAF 11 million over 1978-81 covering Mali, Upper Volta and Niger). The purpose of the project is to increase the availability of forage, in quantity and quality, in the Sahelian zone by the introduction of forage crops in the crop rotation on farms keeping cattle or work oxen. The project tests at the farm level the various species of leguminous plants and grasses available nturally as well as from experiment stations. Selected varieties are then multiplied by farmers and introduced in the crop rotation. Similar efforts should be repeated throughout the country since future increases in livestock production will depend on the extent to which livestock and agriculture are integrated.  - 77 - VIII. PAST LIVESTOCK PROJECTS 8.01 Almost all investment in the subsector is financed by foreign funds. For the period 1977-81, CFAF 8.7 billion has been allocated to livestock, of which 7.3%, or 84%, has been firmly committed by foreign donors (see Table 13). Of all planned investments, however, the livesock subsector receives less than its share in the national economy and exports. Roughly 2.5-3% of commitments are allocated to livestock, even though the subsector accounts for over 10% of GDP and from one-third to two-fifths of export revenues (see Tables 4 and 13). Of investments in the rural sector, roughly 15-20% go to livestock, a share clearly beneath the subsector's one-third share in the primary sector. Bank Projects 8.02 The Bank's first livestock project in Upper Volta (Cr. 557-UW) was passed by the board in May of 1975. The components provide customary veterinary services--including vaccination yards--to livestock herders in the southwest part of the country, improve marketing and processing (construction of cattle paths, a market, and holding grounds, as well as rehabilitation of an 'existing slaughterhouse), and promote the creation of group ranches, which were to improve the management of some 30,000 cattle. According to supervision reports, the health component of this project has appeared to proceed satisfactorily, with most vaccinations and treatments largely exceeding appraisal estimates. The two exceptions are CBPP--now at 20-30% of appraisal levels--and anthrax at less than 10%. The absolute number of vaccinations has fallen (see Chart 4)--a normal trend as more of the adult herd becomes immunized. Earlier problems with trypanosomiasis now seem under control, owing both to better treatment and to adaptation of the herds. It has been possible to carry out two more vaccination campaigns that the 4 financed under the project by using the counter-part funds generated from sales of drugs furnished by the project. But it is unclear if these health services will be maintained after the end of disbursements because the personnel and operat- ing costs of the animal health service may not be provided for adequately, although the government has recently decided to allocate 5 million CFAF to each livestock circumscription. The abattoir has been fully operational since the beginning of 1980 and two of the stock routes have been delineated. The access to and creation of the holding ground, which was part of a forest reserve, was an unsolved problem for a long time; however, a solution has finally been reached by a satisfactory agreement between the forest reserve and the project. The adjoining cattle market has been constructed and taken in use. Implementation of the group ranch scheme was slower than expected although the situation appears to be improving now. There was an initial emphasis on the construction of buildings and other infrastructure needed to support the ranches; as the scheme expands, these works may not turn out to be excessive, although they could have been simpler. Herders were initially reluctant to participate in the schemes; but now there is a waiting list. Trypanosomiasis appeared to be a problem; but that is now under control as the  - 78 - result of treatment, adaptation, and better management. Also, a new control technique, using screens treated with insecticide, has been tested and given good results. There have been legal problems over ownership of ranch land; but there is now better cooperation between farmers and herders, and critical dry-season pastures are being demarcated. As a result of initial problems, the scope of this scheme was reduced from 9 ranches to 3 in 1978, although two more may soon be established with credit to herders. The project is scheduled to terminate in December 1982, but further support to the extension services for the group ranches is being provided under the recently-approved Hauts- Bassins Agricultural Development Project (Credit 1285). 8.03 The group ranch scheme now appears to be functioning reasonably well. There ranches--which are best thought of as a delimited area enclosing camps of enrolled herding families--may now be in a position to deal directly with their major objective: increased productivity of herds and pasture land. The bulk of investment (routes, watering points, and dipping tanks) on the three ranches--covering almost 100,000 ha, or about three-fourths of the area originally appraised--has finished. Over 75 families with 10,000 cattle are enrolled, which takes up all of the estimated stocking capacity on these ranches. At least 35-40 families have expressed interest in joining a group ranch, and they have indicated a willingness to pay for the required invest- ments themselves since the project cannot finance any additional ranches. It is expected that the National Development Bank (BND) could provide the requisite credit to these herders. These 5 ranches would bring the total area covered to 200,000 ha, some 50% more than appraised. Thus, after a slow start and several modifications, the ranches appear to be functioning and to have the support of the local herders, who seem primarily interested in formally joining the ranches because of the security of land tenure it provides them. But they also receive the benefits of better health protection for their herds and of better herd management. There is some evidence that cattle on the ranches can be marketed earlier at heavier weights and that fecundity is higher while calf mortality is lower on the ranches. The critical test of the merits of this innovative component will be the ability of project authorities to work with traditional herders in improving animal husbandry and the manage- ment of herds, developing and managing pastures, and controlling typanosomiasis. First steps have been taken in these areas. Over the next several years, both technical assistance and additional financing will be needed to help the scheme through its difficult gestation and to learn from the experience. Continued support of the Group Ranch Development Center would allow it to serve as a focus for further research and training. Only after more time will it be possible to judge fairly whether the concept is valid and the initial investment worthwhile. 8.04 There are, however, a number of lessons that could already be drawn from the first years of the project. The most important is that ranching schemes, meant to substitute for traditional migratory herding, are difficult and time-consuming to implement because they require numerous social and institutional changes. Therefore, herders must be involved from the beginning  - 79 - in the design of the ranches, and their ability to manage their own herds in this new setting must grow with the project--not be taught later as a replac- ement for departing technical assistants. Investments, especially early ones, should be small and simple--like watering places, dips, and squeeze chutes-- that correspond to the immediate needs of the herders and are within their capacity to finance and manage. At the same time, more fundamental changes in social organization and regulation must be pursued, especially land tenure rights and procedures for controlling herd size and pasture use on the ranches that are created. Improvements in animal husbandry and pasture lands are likely to be incremental, and they may demand additional technical research before good approaches can be found. Discrete, but valuable components like animal health care and feedlots may not need the structure of group ranches at all in order to be implemented, although they can be complementary adjuncts. It is still unclear if these ranches can be economically justified or if they can be replicated. This uncertainty and the inherent delays all indicate that the Bank should be prudent in pursuing this type of livestock project, although recent positive signs also suggest that it would be premature to abandon the efforts already begun. 8.05 Livestock components--comprising animal husbandry activities--are included in all five of the agricultural projects approved by the Bank's Board of Directors since early 1981. 1/ These components comprise general animal health measures, specific efforts to reduce calf mortality by dealing with parasites and diseases, provision of credit for purchase of animal traction equipment, and/or actions by the extension service to improve the integra- tion of livestock with crop farming--including the use of feed supplements, minerals, fodder, and pasture management. Other Aid Agencies 8.06 Other main aid donors in the livestock sector are Holland, Germany, France, FED and USA. Dutch aid is giving strong support to livestock market- ing through ONERA. Its activities include extension of the Ouagadougou abattoir, construction of the ONERA headquarters, and provision of technical assistance for livestock marketing and processing. German aid includes the creation of and assistance to a national veterinary pharmacy (PHANAVET), a modern cattle feedlot in Banfora, a project to study and delimit the areas infested with tse-tse, other research on trypanosomiasis (eradication. immuni- zation, and training), and a feedmill in Bobo-Dioulasso. French aid includes a tse-tse eradication project (eradication, immunization, and training) carried out by IEMVT, technical assistance to the national Livestock Service, and studies (e.g., mapping pastures in the east and research on small ruminants). FED is carrying out projects in the Dori, Yatenga and Banfora ORDs. USAID has completed pilot village livestock projects in Kaya, Fada and Koupela ORDs. 1/ Projects include Bougouriba II, Hauts Bassins, and Volta Noire Agricul- tural Development Projects, Rural Development Fund III, and Koudougou Pilot Agricultural Development Project.  - 80 - 8.07 FED is by far the main aid donor in the livestock sector. At present it is carrying out three livestock projects, two in the northern part of the country and one in the southwest. The livestock project in the Sahelian zone started at the end of 1979 and covers the northern part (5,000 km2) of Dori ORD. The project is concerned with regeneration of pastures and with improving the livestock extension service, aiming at better balance between fodder resources and fodder requirements. The project includes a marketing component to reduce stock numbers. The project period is three years and total project costs are US$2.0 million. 8.08 FED is also undertaking a four-year livestock development project in Yatenga ORD mainly to give support to small stock, but the project also will construct wells and set up a workshop for manufacture of farm equipment. 8.09 Finally FED is carrying out a four-year livestock project in support of traditional cattle raising in Banfora ORD, in the southwest. The project includes fattening of draft oxen, cattle marketing, and the creation of a grazing scheme for transhumant cattle. 8.10 The USAID Village Livestock Development Project aimed at developing the capability of the national Livestock Service and three ORDs to plan and implement village livestock management systems that can maintain the integrity of the environment. The project was implemented by a team of American con- sultants consisting of a rural sociologist, a livestock specialist, and a. range management specialist. The project area was restricted to a few selected villages. Of the total project costs of US$2.3 million, about US$1.5 million was a US grant, mainly to cover salaries of expatriate staff. The project was originally conceived as the initial phase of a two-stage effort, but financing for the follow-up has not yet been assured. 8.11 The BOAD (West Africain Development Bank) project is designed to use livestock to help regenerate abandoned rice fields in the Kou Valley irrigation scheme. Two of its major components are the development of Irri- gated pastures using a mixture of grasses and legumes and trials for feeding and fattening steers. It is being implemented by the pasture division of CERCI. 8.12 A general evaluation of these projects is beyond the scope of this review. Most are recent, thus any definitive evaluation would be premature, but indications appear to be that donors have not yet developed interventions that can have a significant, widespread input on either total production or productivity in the livestock subsector. Except for the national vaccination campaigns against major epidemic diseases, it may be most appropriate to view these existing efforts more as pilot projects, with a large element of applied research.  - 81 - IX. GOVERNMENT LIVESTOCK DEVELOPMENT STRATEGY Description of Current Strategy 9.01 Past Strategies. The Government's strategy for developing Voltaic livestock is based mainly on the preparatory work for the 1977-81 Plan carried out by the livestock service. 1/ Although this document was heavily influenced by conditions created by the 1968-73 drought, its major focus remains largely unchanged. The Livestock Service has reviewed its past strategy and has produced two documents updating livestock strategy which have been communicated to the Bank. 2/ The following description of government strategy draws heavily on these documents but is not limited to them. 9.02 As early as the interim plan of 1967-70, the three major goals had been identified for the livestock sector: better animal health, higher productivity, and maximum local processing of livestock products. By the second plan (1972-76) these goals had changed only slightly, although a stronger emphasis was put on increasing total production--rather than simply raising productivity--to recover from the losses of the drought and on promot- ing Voltaic processing of animal products. In particular, the government proposed a 2% annual growth rate in cattle numbers through 1985, with a 1% rate thereafter, to achieve a herdsize of 3 million animals by 1990. 3/ To increase the production of cattle, three general orientations were pro- posed. (i) Regional stratification would use the Sahel for breeding and producing calves and the Sudanian and Guinean zones for growing out and fattening them. This stratification would require an integration of crop cultivation and livestock in the south, which would be made possible by animal traction and farm livestock raising and by modern, intensive feedlots using agro-industrial byproducts. (ii) Permanent animal health programs would combat major contagious diseases, with a growing awareness of the need to treat illnesses in specific animals like draft oxen. (iii) Better 1/ Republique de Haute Volta, Ministere du Developpement Rural, Sous- Commission de la Production Animale, "Programme de Production Animale pour le 3eme Plan 1977-1981," Ouagadougou, probably 1976. 2/ These documents are: Republique de Haute Volta, Ministere du Devel- oppement Rural, Direction des Services de l'Elevage et des Industries Animales, "Note a l'Intention de la Mission Banque Mondiale," Ouagadougou, 1980; and "Note d'Observations Relative A l'Etude sur le Sous-Secteur de l'Elevage par la Banque Mondiale", December 1981. 3/ From FAO, Perspectives du Developpement Agricole a Long Terme de la Haute Volta, Rome, 1977 (?), p. 24.  - 82 - animal nutrition would be based on the use of agricultural byproducts, cultivation of fodder crops, and management of natural pastures. For poultry, the goal was to modernize and expand production, relying largely on improved health measures and better feedstuffs. Interventions for small ruminants were to be limited largely to controlling parasites. 9.03 To carry out this strategy, the plan proposed actions that can be grouped into six major categories: (i) an increase in the financial and material resources for the government livestock service, including the number of agents and veterinary posts; (ii) border control to reduce clande- stine exports; (iii) land reform to transfer control to herders over some areas that are currently controlled exclusively by cultivators; (iv) better management of pastures; (v) provision of credit to herders; and (vi) reacti- vation and improvement of livestock research stations. The most important were considered to be increased resources for the livestock service, land reform, and livestock credit. In addition, the desire to increase livestock processing within the country has led to the creation of modern facilities to slaughter meat for export, as well as to the institution of a state monopoly over the use of agro-industrial byproducts to aid intensive cattle fattening. 9.04 Updated Strategies. In recent efforts to update its livestock strategy, the government has changed little from the major elements identi- fied for the last plan, except that the problem of land tenancy is now seen to be an extremely important constraint to further rational livestock devel- opment. In addition, there appears to be a growing recognition of the need and desirability of integrating livestock into agriculture, including pasture improvement and promotion of small ruminants and other small stock (chickens, pigs). In broad terms, emphasis is now placed on three elements: health, production, and exports. Improved animal health is considered a precondition to increase both production and productivity. The government wants to main- tain permanent protection against the major contagious diseases, with par- ticipation by herders limited to the costs of vaccines and perhaps to the transport component of delivery costs. In addition, the livestock service calls for more attention to be given to controlling parasites, combating trypanosomiasis, and improving veterinarian care of individual animals. To bring about a better health program, the livestock service feels it needs more staff and equipment, better structures such as laboratories, and increased training of staff. 9.05 The second element of the updating focuses on animal production. In order to increase output, the government: (a) continues to adhere to the doctrine of regionalization in order to improve extensive livestock herding; (b) wants to organize and equip pasture areas--viewed more as an administra- tive than an investment problem; (c) thinks that priority allocation of agricultural byproducts must be given to livestock feeding--regardless of their alternative uses; (d) wants to adapt short-term credit to finance livestock feeding; and (e) stresses the need for a larger, more highly trained and better equipped livestock service.  - 83 - 9.06 The third element is a strong bias favoring exports of meat over live animals, although the most recent statement of objectives conspicuously excludes the promotion of meat exports and local livestock processing. 9.07 Views of Herders. In order to complement the views of the admini- stration on livestock strategy, the mission contacted numerous herders and livestock producers directly. 1/ Even though most of the views come from herders in the east, they represent an important segment of livestock production in the Sudano-Guinean zone. From the herders points of view, their major problems include: (i) lack of water, which forces transhumance during the dry season; (ii) shortage of pasture, which is exacerbated by bush fires and the reported influx of cattle from the north since the 1968-73 drought; (iii) and conflicts with cultivators over the placement of fields and destruction of crops by cattle. Disease problems are not felt to be severe, although they become worse--especially trypanosomiasis--when cattle are taken south. 9.08 There was general agreement that the drought had brought large numbers of herders and cattle from the north, and that this influx has perturbated the equilibrium of the herders already living in the area. In many cases, the shortage of water was attributed to the larger number of cattle from the north that have to be accommodated. Localized migrations have both turned more toward the south and increased in distance during the last several years. The local populations express discontent at this change--in part because the south is viewed as ridden with disease. Yet they feel they should not try to stop the influx from the north. Lack of pastures was sometimes attributed to influx of northern cattle, but most herders eventually claimed that there would be no real shortage--even with the larger cattle population--if bush fires could be controlled. No one had suggestions of how to do this, although the incidence of bush fires tends to diminish as population density and cultivation increase. 9.09 The most immediate--and to some, the most serious--problem was felt to be the conflicts between farmers and herders. Such conflicts arise when cattle herded by Peuhl eat standing or harvested crops, and their resolution usually means that the herder has to pay for damages as assessed by extension agents of the Agricultural Service. The farmers charge the Peuhl with negligence, and the herders charge the Gourmantche with failure to respect the herders- rights of access to the village and to watering points. It appears that conflicts flare up only between ethnic groups, and that punitive actions are taken only against the Peuhl. 9.10 A number of measures might be taken to reduce these conflicts, but the most frequent have been to increase penalties for crop damage, so that herders are encouraged to pasture their animals farther from the fields. In 1/ Herders were visited in Samorogouan, Nassougou, Matiakoali, Gayeri, Kantchari, and Diapagou. Livestock feeders were visited in Kaya and Ouagadougou.  - 84 - addition, farmers sometimes give greater protection to grain stores. Among the herding populations in the Sahelian north, cultivated fields are commonly enclosed using spiney bushes. Fields of high-value crops like fruits and vegetables are protected in the south, but only during the dry season when divagation is accepted. More careful herding is alleged to be in- feasible by the Peuhl, presumably given the high cost of herders. 9.11 Despite the poor quality of veterinary services and infrastructure in the country, as well as the widespread incidence of disease and parasites (see Tables 17 and 18), herders do not seem to perceive animal health as the major problem. Control of major epidemic diseases means that catastrophic losses are unlikely, and the relationship between some illnesses and pro- ductivity is perhaps not well appreciated. However, there is a clear demand for proper diagnosis and curative treatment of individual illnesses like trypanosomiasis. The demand is great enough that herders are apparently often willing to travel to Ouagadougou themselves to purchase medicines unavailable in the regional livestock offices and to pay the travel costs of veterinary assistants who come to treat their animals. Evaluation of Government Strategy 9.12 In much of its general emphasis, the government's strategy is sound and conforms to the major orientations identified by this review: integration of livestock and cultivation, diagnostic and curative veterinary services, and improvement in pastures and fodder. But as far as presenting a rational program of action, government planning could be improved. There needs to be a clearer distinction between objectives and means, with more well-defined and realistic objectives. For example, despite the unimpressive performance in the past, the Plan optimistically asserts that meat production would rise by 70% between 1977 and 1981. Moreover, it fails to make the correspondence between its general objectives and what steps must be taken to achieve them. There are virtually no estimates of how much various interven- tions will raise productivity--the sine qua non of expanded Voltaic production. Hence there is no way to judge whether increased expenditures on research, training, staff and facilities will actually increase both physical and economic productivity. The few departures from themes established by the colonial veterinary service have largely been forays into modern, self- contained parastatal facilities, such as feedlots, feedmills, and slaughter- houses. Besides not helping small livestock producers, who appear to be relatively efficient, these activities both waste investment funds and are probably a recurrent drain of scarce public revenues. In general, if might be said that the strategy is unduly composed of preconceived programs and not based sufficiently on what producers want and actually do. The paragraphs that follow assess specific elements of the government's strategy. 9.13 Stronger Livestock Service. Despite the strong priority accorded to a strengthening of the livestock services by the planning commission, adequate budgetary allocations have simply not been forthcoming. In terms of domestic public funds, the livestock subsector traditionally receives less than it contributes to the national budget. Although foreign funds  - 85 - more than compensate for the outflow of government funds, the subsector still appears to receive less than its share of such investments. Moreover, the situation has deteriorated, not improved, since the commission's report in 1976 that called for a stronger livestock service. Staffing remains inadequate (less than 250 field agents), poorly distributed among regions, and limited largely to activities related to animal health (see Tables 14 and 15). The government's goal that each of the arrondissements has a livestock post by 1982-83 is unrealistic in light of current staffing levels and past performance. 9.14 The inadequacy of budgetary allocations could be partly offset by increased participation of the herders and other producers in the costs of the livestock service. Despite the growth in the service, however, the proportion of receipts actually recovered appears to have declined (see Table 12). The livestock service insists that animal owners should be charged only for veterinary products while the cost of delivering these services should be fully subsidized. Estimates based on figures from the Bank's West Volta Livestock Project suggest that these delivery costs are 3-4 times the cost of vaccines for major cattle diseases and one-half to two-thirds the cost of treatments against trypanosomiasis (see Table 9). But because it is unable to finance the subsidies, it cannot provide many of the services demanded and needed by the herders. Material inadequacies and poor financial management of the livestock field services mean that herders must already often bear substantial extra charges anyway in procuring vaccines and arranging for agents to give the treatments. In light of the existing situation, the government is considering augmenting vaccine costs by 10% to cover some of the costs of transport and travel. The establishment of an effective revolving fund at the national level to cover the costs of veterinary products and their delivery is a priority to improve animal health services. 9.15 The government's fear that higher treatment costs will drive herders into neighboring countries where vaccinations are often free is probably overstated, except for areas near the borders. It seems unreason- able to expect that an additional few hundred francs per animal will cause substantial shifts in cattle movements, especially when some herders already pay extra charges. Even if temporary movements were to occur, there may be a net gain for the country since animal health costs would be subsidized by neighboring governments. 9.16 The livestock service generally blames its poor performance on its integration with the agriculturally-dominated ORDs in 1975 and feels it will be more effective if autonomous. Autonomy, however, in no way assures that adequate budgetary support will be forthcoming. More seriously, it would ignore the conclusion here that increases in the productivity of Voltaic livestock can probably occur in the long run only if livestock and agriculture are integrated--not separated. 9.17 Regional Stratification. The government recognizes that this concept has been poorly realized, despite its investments in structures-- like markets, stockroutes, feedlots and slaughterhouses--to handle cattle from the north. It blames this slow progress largely on the mentality of herders who are considered reluctant to sell their stock--especially young  - 86 - stock, While the decision to sell cattle is affected by many factors--such as the need for money, the use of cattle as a store of wealth and sign of status, and sentimental attachment to animals, the existing structure of prices also encourages herders to hold livestock until adult age, providing they have sufficient funds and pasture. Available market data clearly indicate that price per kg rises with the size and age of animals (see Table 8). Given that the risk of mortality is fairly low once a calf reaches 1-2 years of age, herders would probably maximize returns by keeping animals longer so long as pastures, water, and herding labor are not constraints. 9.18 However, the evidence is fairly strong that young cattle are being sold by herders despite what appear to be relatively low prices for them. For example, there are about 25% fewer male than female calves (see para 5.01), and it may be that herders in the north just do not have much more of a surplus of young stock to sell. A large share of these are being slaughtered before the age of 3 years--despite the apparent attractiveness of growing out (and possibly fattening) cattle. This movement may be forced in part by deteriorating economic conditions in the Sahelian north--reflecting a failure of herders' incomes to recover after the drought, and in part by the strong, growing demand for young oxen for animal traction and the demand for smaller, cheaper animals by local butchers. It is also consistent with the alleged preference of Peuhls to produce milk rather than to grow out bulls or steers for beef, although it conflicts with statements by herders that they prefer not to sell young stock. 1/ 9.19 The fact that large numbers of these small animals are slaughtered --rather than grown out--further shows that the government has been unable to implement its policy of regional stratification. There may be two principle reasons for this failure: (a) the unprofitability of feeding out cattle, given the costs of feed and the buying and selling prices; and (b) the lack of credit to finance feeding operations. In any case, the government's modern feeding lots have had no significant impact quantitatively; and the government has given little assistance to private producers who might undertake growing out and feeding. 9.20 Land Reform. Although the Livestock Service, along with herders and cultivators, now consider this issue as perhaps the most severe con- straint to improving livestock production, virtually no action has been taken--beyond radio proclamations against bush fires. The Livestock Service views the first step as simply increasing awareness of the problem among other government agencies and obtaining financing for a master plan of land use potential. But the issue will not be simple to resolve, in part because it goes beyond the competency of the Livestock Service and the Ministry of 1/ As one put it, "A country without children is not a country."  - 87 - Rural Development, in part because it calls into question the existing priority of land use given to cultivation over grazing, and in part because any reform will require some control over cattle movements which traditionally have been unfettered. Despite the recurring mention of land use as a problem, it may be that economic and demographic pressures are not yet strong enough for reforms to be accepted. 9.21 Until the mission's insistence in early 1980, there had apparently been no collaborative effort among government agencies to define a land-use policy. Even now, the prospects that a policy will be elaborated are not great--in part given the reluctance of the officials to broach a sensitive issue that will necessarily require a reallocation, of sorts, of resources among groups. Even were a coherent policy to be elaborated, there is little evidence to date that the government would have the commitment and means to enforce it. At the village level, farming and herding groups who coexist seem unable to reach mutually satisfactory accords, except insofar as farming asserts its traditional dominance over herding by pushing it farther from the village. It may well be that effective land reform must largely await changes in economic, social, and demographic conditions that make more intensive livestock production relatively more attractive than extensive herding. 9.22 Beyond the usual claim of the government that it owns all land, and therefore has the legal right to reallocate it, there is no stated land tenure or land use policy. Within the traditional legal system, however, all land seems to be owned or at least claimed by members of the resident populations--whether the land has ever been tilled or not. Since herders do not constitute a part of this resident population, except perhaps in the Sahelian northeast corner of the country, farming always takes precedence over grazing. Most conflicts appear to be judged in favor of farmers, and herders must adjust. A definitive reallocation of land in favor of live- stock, with accompanying restrictions on the decisions of the farming population, represents a departure from the traditional and de facto land tenure system. Even when the government has taken a legal position to reserve land for herders, as for the group ranches at Samorogouan, it has offered virtually no enforcement and has only reluctantly--if at all--backed up the efforts of herders to keep farmers off the land that the government itself has given to them. 9.23 A land use policy of some sort is probably a prerequisite for the implementation of a grazing scheme--especially if it involves watering points and improvements to pastures, in order to assure that the resources provided to the herders are not over utilized. The policy will have to address two different problems: (a) control of the use of grazing land for agricultural purposes, because in almost all areas cultivation and herding are intermingled (the problem of zoning); and (b) control of the use of pasture and water by non-resident herders, because transhumance is a common practice for nearly all herders (the problem of ownership). Both herders and farmers sense the need to organize their respective rights to land, and both claim that outside support from the government will be necessary to establish a system that will both preserve existing rights and reduce  - 88 - conflicts. Resident herders admit that the arrival of herders and livestock from the north, especially following the drought, has impinged on their traditional pastures and water, sometimes forcing them to migrate themselves to less favorable areas in the south. But no herder wanted to limit foreigners' traditional rights of access, in part because they themselves migrate to other areas, and in part because of familial and commercial links. 9.24 Credit. Despite the government's belief that a shortage of credit constrains the development of livestock production--especially growing out and fattening--it has not had the funds to improve the situation. The National Agricultural Credit Bank (CNCA) offered an opportunity for the government to implement a major tenet of its livestock strategy, but the CNCA provides virtually no credit to the livestock sector except for purchasing draft oxen. The credit, as in the agricultural sector generally, is sub- sidized, and it is quite possible that livestock--as well as crops--would be unable to support the real capital and administrative costs of rural credit. The Livestock Service has apparently not explored the feasibility of a program for livestock credit. 9.25 Livestock Feedstuffs and Intensive Animal Production. While this is a valuable component of a long-term development strategy for livestock, government's efforts at implementation are not always well-advised. The major part of the government's program has been focused on modern feedlots using mostly agro-industrial byproducts. Because the operations--both cattle fattening and poultry and egg production--have been generally unprofitable, the Livestock Service and ONERA have tried to acquire monopsonistic control over all agro-industrial feedstuffs. It argues that intensive feeding operations are unprofitable because feedstuffs are too expensive; therefore, the government should control their use and make them available to the livestock service at low or no cost. The Livestock Service's policy would forbid any alternative uses (including exports) of such byproducts and ignores the fact that prices of some consumer goods (such as oil and flour) would have to rise if no longer offset by returns from the sale of byproducts. Finally, it overlooks the fact that private cattle feeders appear to make a profit using feedstuffs that are available to them on local markets at much higher prices than the Livestock Service and ONERA now pay. Despite the presence of backyard livestock fattening, the Livestock Service has done almost nothing that could aid these private efforts, and there are only recent, inadequate moves to include animal husbandry in livestock extension activities. Finally, the recent government approaches seem to miss the point that supplemental feeding, coupled with grazing, is probably the best way to get marginal gains from feedstuffs that are high enough to support their costs. For cottonseeds, however, there does appear to be a case for increasing the share allocated to the livestock subsector in general. 9.26 Exports of Processed Livestock Products. One of the most strongly expressed and specific elements of the government's strategy has been the export of processed products -- mainly beef and tanned hides. The government argues that local processing adds extra value to the products and thereby  - 89 - increases national income and export earnings. 1/ While seductive in prin- ciple, the strategy may not make economic sense because: (a) the processing industries are capital-intensive, relying heavily on imported equipment and energy; (b) installations are large and capacity tends to be underutilized (the case for the slaughterhouses, cold storage facilities, and tanning facility); (c) transport costs are usually higher for processed goods (the exception may be poultry carcasses); and (d) valuable byproducts may be wasted owing to the inability of the local market to absorb them (the case of the fifth quarters of cattle). If the processing is inefficient -- and there is not much evidence to the contrary -- implementation of this component could actually reduce potential income and lower foreign earnings. 9.27 Border Control. The Livestock Service's desire to prevent unde- clared exports can be justified only if the increase in tax revenues justifies the additional administrative cost. This accounting has not been done. In any case, investment in border control would have virtually no impact on national income, livestock production, or productivity. Moreover, to the extent that it raises the effective tax rate on cattle exports, it lowers the return to herders and could conceivably depress production. Given the pressing needs for productive investments, this element of strategy may be an unwarranted luxury. In any case, it is not supported by the Customs Office. 9.28 Pasture improvement. Despite the government-s view that better pastures are important to improve the livestock subsector, it has done little that reflects this interest. There is no separate division in the Livestock Service for pastures. There has been no agrostologist on the staff of the Livestock Service, and there is no land-use policy that would encourage pasture improvement by either farmers or herders. But there are some encouraging signs. Various foreign donors are financing some studies and pilot schemes related to pasture improvement; and in 1980 the reorgani- zation of the Livestock Service included a separate division for pasture and water development. 9.29 Research. Contrary to the Livestock Service-s stated belief that more research on livestock is vital, none of its three research stations carries out any research. Despite the priority assigned to research by its strategy, the government has, in fact, watched the situation deterio- riate, not improve. 1/ Its argument that hide exports are an important source of STABEX revenue is, however, false. For the four years 1976-79, hides and skins exports received no compensation.  - 90 - X. DISCUSSION 10.01 The purpose of this section is to discuss the options open for the government and for donors to intervene in the livestock subsector. Given the lack of reliable data and the relatively limited experience with live- stock projects, recommendations are necessarily tentative; and in some cases only pros and cons are considered. The following paragraphs set out the major constraints in the subsector and discuss major orientations for govern- ment and donor action, including specific project ideas and appropriate changes in policies. 10.02 The fundamental problem in the subsector is the low productivity of the herd, reflected in low offtake rates and low herd growth. 1/ What is unclear, even after this review, is why productivity and output are not higher. Clearly, there are physical, technical constraints, including poor animal health, lack of feed, and poor management. There are also economic constraints resulting from what appears to be stable or perhaps falling world prices and rising costs of production. There are institutional and organizational constraints that may retard changes. Finally there are direct constraints on government action, the most important of which are shortages of funds. Perhaps most importantly, it is unclear that the govern- ment has the knowledge to attack these various constraints, even if it had sufficient resources and power to do so. Physical Constraints 10.03 Health. Loss of livestock from contagious diseases now appears to be quite limited, reflecting the success of past major national vaccination campaigns. Thus, epidemic diseases are no longer an immediate constraint to greater livestock production. But because the threat of outbreaks continues to exist--especially given large international movements of cattle and poor border control, these vaccination campaigns should be maintained as a preventative measure. And since the actual gains to the nation from con- tinuing to suppress epidemics (e.g., from rinderpest and CBPP) usually exceed expected gains for private producers, these campaigns must usually be subsidized. Although current government policies, as well as the bulk of the livestock service, are designed to achieve this end, implementation is increasingly constrained by the shortage of funds to pay for the vaccines and the operation of the veterinary service. 10.04 Although major epidemic diseases are largely under control, poor animal health--including parasites and trypanosomiasis--is probably one of the causes of low fecundity in cows and high calf mortality. There are probably large potential gains still to be won in this area--especially in 1/ It is interesting that this same problem was pinpointed in 1964 by the World Bank in The Economy of Upper Volta, Washington, p. 7.  - 91 - wetter regions where the tse-tse challenge is severe--but the animal health service must be somewhat reoriented to deliver necessary treatments. The Livestock Service recognizes the need to provide health care for individual animals, but to be able to do this, it may require certain assistance, including: (a) funds for acquisition and operation of material (vehicles and veterinary equipment); (b) retraining of veterinary nurses and livestock agents; (c) financing for initial subsidies on some products to help encourage their use; and (d) improvement of certain infrastructure necessary for health services. Because improved health of herds should lead to greater wealth and higher incomes for producers, these services ought in principle to be self supporting. New approaches to providing these services must be considered, which could include greater involvement by village groups and reliance on private vendors, with technical support from the Livestock Service. 10.05 Feed. In the short run (say 10-20 years), the quantity of these resources and the efficiency of their use will probably increase very little, if at all. Despite existence of some unused pasture areas in the eastern sudanian zone and southern tse-tse infested areas, available pasture is probably a limiting resource and it may be that lack of feedstuffs means that few increases in herd size and output can be expected from these areas by simply improving animal health. 10.06 There are some tse-tse free areas--notably the northern half of the Fada ORD--where pasture resources appear to be presently underutilized because of water shortages. Grazing these pastures could be an initial step in helping to feed more livestock. If cattle density could be doubled-- making it about the same as on the central plateau--the cattle population could rise by some 300,000, or 10%. But cultivation in these areas is already widespread, especially in the low-lying areas that would have made the best dry-season pasture. Thus it may be too late to implement this option simply by reserving tracts of grazing land for pastoral associations --even if complex institutional problems could be overcome. Yet without these associations, donors may be reluctant to improve water resources. 10.07 Additional livestock production in the zones currently infested with tse-tse flies may offer the greatest potential source for expanded output in the short-term. There are three basic approaches for achieving it: treatment of Zebu against trypanosomiasis, eradication or suppression of tse-tse flies, and multiplication and introduction of trypano-tolerant cattle. Only the first option both has significant scope for immediate implementation and appears very attractive from economic and managerial perspectives at the present. But its future viability depends importantly on continued research into drugs and vaccinations to prevent tryps from building immunities. The group ranches currently supported by the Bank have become a test of the acceptability of this approach.  - 92 - 10.08 Control of tse-tse flies by artificial means is usually considered to be prohibitively expensive--both financially and administratively-- initially as well as on a recurrent basis, except for highly intensive live- stock operations; and these operations are probably unlikely to be profitable (see the discussion on feedlots). It may be that real control of tse-tse must await the land clearing associated with higher population densities, which also implies that livestock may have to be produced in relatively more intensive systems integrated with agriculture. On the other hand, it is also possible that new techniques will soon be developed to combat tse-tse more effectively and at lower costs--for example, sterilization of male flies. 10.09 Greater use of trypano-tolerant taurin cattle is an alternative that can be implemented only slowly, but it would appear to warrant in- creased attention, given the fact that 30% of the national herd already consists of these cattle. Since the major physical constraint to expansion is not the quality of the breed, but rather the small size of the taurin breeding herd, active efforts to purchase taurin breeding stock from neigh- boring countries may be one immediate step that could be taken. Such an effort would coincide with the component of the recently approved Guinean livestock project aimed at exporting more taurin breeding stock. 1/ 10.10 The objective of using existing feed resources more efficiently is to obtain more livestock products from existing feedstuffs. Although better feed conversion may eventually require breed improvements, in the near term it will depend essentially on improved herd management. It is here that greater intensification may offer the potential for significant gains. The major questions are under what conditions will intensification be attractive to producers, and what constitutes better herd management? Population growth alone is probably the major impetus for intensification because it makes extensive livestock herding more difficult and costly and increases the attractiveness of more intensive crop cultivation that relies in part on livestock activities. 10.11 Available feed resources might be increased by shipment of agro- industrial by-products from the more productive southern areas, by increased utilization of agricultural by-products produced in the region as livestock feed, and by expansion of fodder production. Although attractive in theory, there are problems with this approach. Agro-industrial products are limited, often have a fairly high value in alternate uses, and are expensive to trans- port. Increasing the cattle population in the south where more agro-industrial by-products are available is limited by tse-tse infestations and by the slow increase in the taurin herd. Traditional agricultural by-products appear to be already largely used for livestock. Methods that are attractive to farmers for producing and conserving fodder are still in pilot stages, 1/ See World Bank, "Revolutionary People's Republic of Guinea - Livestock Development Project - Staff Appraisal Report", Report No. 2959a-GUI, Washington, August 19, 1980, pp. 21-22.  - 93 - although in the long run this action might be very important. Institutional difficulties (rights to graze cultivated pasture) must be worked out and the economics needs to be examined. Thus, in the short run, none of these tactics appears promising. Economic Constraints 10.12 Principal economic constraints can be summarized: (a) lack of any firm evidence that beef prices are likely to rise in major export markets, (b) high transport costs, (c) high costs of feedstuffs and fattening stock, (4) rising costs of herding labor, and (e) cheap milk imports, including lack of tariff protection. There appears to be little evidence that herders behave in ways that are not economically rational--they respond to changes in prices, costs, and investment opportunities; the constraint is not, as sometimes suggested by the government, the need to change mentalities, but rather to offer appropriate incentives. 10.13 Although Voltaic beef is currently competitive in coastal markets, world beef prices are expected to remain constant or even to decline in the next 5-10 years. There is probably little scope for relying on new methods to expand production if they raise costs, and major efforts should be directed at actions to reduce costs of production, marketing, and transport. Production activities have already been discussed under physical constraints above, and the review finds little scope for reducing marketing costs other than some institutional arrangements to aid information exchanges and transfers of money. There may be interventions to reduce transport costs, including: improved stock routes (demarcation posts, grazing reserves and holding grounds), refrigerated rolling stock to ship meat by rail, and slaughter facilities for poultry. However, transporting meat rather than livestock induces other costs that could offset gains, and data are not yet reliable enough to draw any firm conclusions. In the absence of cost reductions, and in the face of growing domestic demand, it could well be that prices in Upper Volta will rise, which will encourage more production but will tend to reduce exports. In short, there seems to be a conflict between paying higher prices and maintaining exports. 10.14 The growing cost of herding labor and of litigations with farmers seems to be shifting the relative profitability of cattle production away from extensive methods to move intensive activities. This trend is likely to strengthen as social programs and agricultural expansion make herding less attractive, and as better transport, better animal health, improved  - 94 - breeds, and more agricultural by-products make intensive production easier. The implication of this shift is that a major production input is becoming dearer and that livestock production will tend to become less attractive unless these costs can be passed along or offset. Downward pressure on world prices limits the scope for passing on costs. Interventions that might com- pensate for higher labor costs appear to require the availability of unused pasture lands or the evolution of mixed farming. 10.15 Most feedstuffs other than pasture currently sell for prices that make it very difficult to convert them into beef except under special, somewhat restricted conditions where such feedstuffs are complements to grazing. They do, however, enable some producers to take advantage of sea- sonal price changes by feeding cattle during the dry season. The situation is unlikely to improve in the future, although the real costs of forage production have never been carefully estimated seriously. Changes in government policies could also allocate more cottonseed to herders. More intensive feeding programs will probably succeed only if they produce high- quality beef and are very short-term to take advantage of seasonal price increases. Given past Voltaic experience, there appears to be no economic justification for promoting large-scale feedlots, and there is not yet suf- ficient information to determine how to encourage small-scale finishing-- although the activity is already occurring. In general, the high value of feedstuffs means they must be viewed as complementary feeds, not as a sub- stitute for grazing. Although the ratio of fattening stock prices to those of slaughter cattle is still favorable, comparative information from other countries and strong evidence of a growing demand for young stock--induced in part by the lack of herd growth in the past--suggest that prices for fattening stock are likely to rise significantly faster in the future. 1/ As a result, operations that rely on feeding alone are likely to become less profitable, at least for export. 10.16 The high volume of cheap milk imports, mostly foreign food aid, limits scope for programs to expand domestic commercial milk production. Moreover, local fresh milk can neither be stored nor transported so easily as imported condensed or powdered milk. To the extent that milk sales are vital to make the integration of livestock with agriculture profitable, low milk prices could pose difficulties and would require both a marketing program and trade protection. A program to increase milk production might also be justified by increased consumption by producers themselves rather than by substituting for imports. Institutional Constraints 10.17 Two major institutional problems appear to reduce the efficiency of the sector: structural imperfections in marketing, and conflicts over 1/ The current relatively low prices for young stock may also reflect pressures on herders to sell, perhaps in part because grazing lands cannot feed the new animals. The effect of this selling of young stock will be to reduce herd offtake, leading to future increases in their price.  - 95 - land use. While trekking is well organized by traditional traders, it may be possible to help it function better by improving the cooperation of local officials along the routes, by investing in public facilities (resting places and forage) along routes, and by regularizing (or rediscounting) monetary claims on importers outside the franc zone. Private trucking also appears to be reasonably well organized, although poor roads increase costs and isolate some areas during the rainy season. These problems cause difficulties mostly in the market for small ruminants, since they cannot be walked long distances. Arrangements for rail transport may be the least satisfactory, in part because much of the system is outside the control of traditional cattle merchants. It may be possible to encourage institutional changes that would improve rail transport, including better assembly operations--which might be aided by holding grounds near rail lines and by improved rail scheduling (lack of rolling stock does not appear to be a problem but should be examined more carefully). 10.18 Problems with land tenure arrangements are widely seen to con- strain both expansion of extensive herding and increased intensification of cattle production. Livestock activities are seen to be at a disadvantage because grazing land is regarded as a communal, not individual asset. The issue, then, is not one of zoning by the government but of conferring effective control of land use to pastoralists who traditionally have the right to graze land only if it is not cultivated by resident farmers. Ownership, as such, would both guarantee to herding groups control over the use of land and make them responsible for its good management. By clarifying who has the right to use what land, this approach would help reduce uncer- tainty and costs of livestock herding. 10.19 For such a scheme to work, several conditions probably have to be met. First, the government must show a willingness to institutiona- lize and enforce the arrangements eventually worked out in the grazing areas. The review found a reluctance even to consider the issue, consistent with the government's slow action in procuring titles to land needed for the Bank's project. Even were the government solidly behind the reforms, it would doubtlessly need an increased administrative capacity to carry out cadastral surveys, to help to assure that decisions are respected, and to handle litigation. 10.20 Second, there must be a demand for land ownership by herders, and here the evidence is confusing. Although herders often agreed they would rather not be nomadic, they also recognize the value both of their being able to go elsewhere and of receiving others into the areas they customarily use. Thus, while settlement and control over one's own area is viewed favorably, loss of migration rights is seen clearly as a disadvantage. Although it still seems premature to think of widely replicating ranches and grazing associations, several factors are slowly converging that may make these or other options feasible. These factors include the increased dif- ficulty of finding grazing during transhumance, the higher cost of herding (both labor and litigation costs), social pressures encouraging sedentary livestock production (e.g., schools), and more efforts to provide stock water in the dry season and to improve and to protect pastures--making migration less necessary.  - 96 - 10.21 Third, cultivating groups must be willing to relinquish their standing claims over land currently used for grazing. As livestock becomes more closely integrated with farming--which is the trend and which is being encouraged by the government's promotion of animal traction, cultivators will have less need for symbiotic relationships with herders and more need for pasture lands of their own. Perhaps more importantly, continued rural population growth will increase the demand for crop land, which will steadily eat into many grazing areas. It may be necessary to envisage paying compen- sation to some groups before conferring ownership to herders. Herders are also likely to rely more on farming--in fact, it is possible that they should be the initial focus of efforts to promote mixed farming. 10.22 Fourth, organizational issues must be solved, including the type of land ownership by herders (individual or cooperative), the size of the area needed for each herd, and how to take and enforce decisions controlling individual herd size and disposition. The Bank's current group ranch scheme may eventually provide insights to resolutions for these issues. Constraints on Government Action 10.23 If the problems discussed above are those that are primarily responsible for physically limiting herd expansion, what is preventing public agencies from taking appropriate actions to overcome them? Given the importance of livestock to the Voltaic economy and trade, it would be difficult to argue very strongly that the government does not see the value of increasing livestock production. Admittedly, it may give priority to crop production, which is only consistent with the dominant importance of crop cultivation. Lack of action by the government probably indicates the presence of constraints other than low commitment. 10.24 There are three other major constraints on better action by the government: insufficient funds, poor coordination, and conflicting policy considerations. Even if past public expenditures on livestock have not been out of line with the importance of the subsector in the economy, it does appear that sufficient funds have not been directed at major constraints on increased output. To illustrate, applied research, animal husbandry exten- sion, and operating costs of the veterinary service have been neglected while funds have been spent on some investments that probably will have no major impact on animal production such as commercial feedlots, slaughter- houses, cold storage facilities, and group ranches. 10.25 Part of the blame for insufficient funding rests squarely with the government because it maintains unnecessary subsidies, its agencies are not always very efficient, and funds are often misallocated to the wrong programs. The government may also be simply unable to mobilize an adequate level of domestic public resources to fund livestock development at levels needed to increase production in the face of constraints. It must therefore rely heavily on foreign aid, which means that much of the blame for poor funding must also rest with donors, who are responsible for starting ambitious programs (e.g., the Markoye research station) or presently unprofitable  - 97 - ventures (e.g., feed mills) that require a level of recurrent public fi- nancing the government can scarcely be expected to maintain. Donors may need to accept a long-term obligation to provide continuing finance for the operation of inherently deficient operations (like research, vaccinations against epidemic diseases, and training). However it is important to note as a final caveat that lack of public funds is a real constraint only to the extent that these funds could be productively used. 10.26 It appears that governmental arrangements have made it difficult to coordinate interventions in the livestock sector. For example: (a) Until mid-1980, livestock agents were paid by the central live- stock service but supported logistically by ORDs; (b) Units responsible for animal husbandry and development of forage production are separated from the agriculture service; and (c) The parastatal ONERA, in addition to managing specific assets of the government (like slaughterhouses), is building a parallel structure that increasingly duplicates and competes with the admin- istrative service. This lack of coordination, illustrated by the fragmented, piecemeal programs of different foreign donors, reflects the difficulty that the government has had to devise a workable strategy that contains clear priorities and addresses major constraints. 10.27 The subsector review did not really show that current public finance policies of the government significantly impede livestock develop- ment, in part because the most regressive are never implemented. Nor do they aid it very much because expenditures called for by policies are seldom made. Subsidized prices on the delivery of veterinary services encourage their use, which is desirable, but the livestock service is not sufficiently funded to assure their delivery to meet demand. Price controls on meat cannot be enforced, mostly because the government has no way of affecting either supply or demand. Large imports of milk do not directly compete with local production, largely because imported milk--especially food aid--meets a demand that fresh local milk would not. The sector is clearly taxed by the government--almost wholly through exports; under reasonable assumptions it does not appear that such taxes have much impact on production despite their tendency to reduce producer incomes. Public investments more than compensate for the tax outflow, but they do not always appear to be very useful. Such misallocation is often the fault of donors. Actions to Consider 10.28 Since livestock has been and is a major economic sector in Upper Volta, the objective of donors' interventions should be to expand, or at least maintain, output from the sector. In particular, if its importance in earning foreign exchange is to be maintained or improved, and if domestic  - 98 - consumption is allowed to increase with population and income growth, national production must rise. According to several projections, it is unlikely that livestock will continue to be an export sector unless actions are taken to augment productivity and perhaps to restrain domestic demand. Owing to physical constraints on available feedstuffs, much of this growth must come from increased productivity, not herd expansion. A reasonable growth target is perhaps no more than 1% per year. 10.29 Any action by donors is also constrained by two important factors. First, there is no clear indication of major programs--at best there are a number of marginal, rather small investments that suggest themselves. There are also some ideas for limited pilot schemes as part of a search for viable interventions that could be widely expanded. Thus, the first step would be to help the government elaborate a strategy, which could be used to assign priorities to projects and to form the basis for proposing alternatives for reorganizing the livestock service and forging links with related agencies. This might best be done in the context of a technical assistance project or a sector loan coupled with various other components. Second, most of what is needed cannot be directly financed by donors; actions focused on institutions, organizations, policies, and expenditure priorities demand a close, continuing dialogue with the livestock service, perhaps coupled with financing. 10.30 A few general orientations for future actions emerge from this analysis. There is only limited scope for expansion and improvement under traditional, extensive techniques of herding, but two attractive possibili- ties are: (i) to strengthen curative animal health services (which would be mostly self-supporting) that will improve rates of growth and feed conver- sion; and (ii) to expand production in zones where the tse-tse challenge is strong. In tse-tse free zones where agriculture is dominant, as well as perhaps in areas with trypanotolerant cattle, the long-term orientation must be intensification. The country is not ready for major actions because relative prices have probably not yet sufficiently changed to make it attractive except under certain conditions. But incentives should increase, and pilot schemes might help elaborate workable interventions, including fodder production and conservation, better use of ox traction and manure, dairying, and fencing. It is unclear whether lack of credit is a binding constraint to intensification; technical factors and costs seem more critical. 10.31 There is a clear need for basic research in such areas as drugs to control trypanosomiasis and for further applied research in areas such as animal husbandry and feeding, production and conservation of fodder, and control of animal parasites and other health problems in intensive opera- tions. Effort to improve local herds by breeding probably should have low priority--given the apparently under-used genetic potential of both Zebu and taurin cattle; in any case they are long-term and might be limited in the near future simply to developing better criteria for selection. However, in the long-run, genetic improvements will be essential to raise herd productivity and improve prospects for dairying. Socio-economic research is needed to provide more insights into methods for land reforms and more intensive live- stock production.  * - 99 - 10.32 The specific project ideas that might be financed are neither abundant nor major. There may be a case for poultry processing and more stock routes as designated on the attached map. There is a need for more research and although this does not necessarily imply simply reopening the defunct stations. If nothing else, current research programs need to be better coordinated. There is clearly a need for training especially to teach agricultural extension agents about animal selection, feeding, and husbandry, and about fodder cultivation and conservation. Veterinary agents need to be trained more in curative medicine, as well as in how to teach livestock producers to treat their own herds with modern medicines. Plans have already been elaborated to expand ENESA. 10.33 Most of the interventions that directly affect livestock produc- tion--except for on-going national vaccination programs against epidemic diseases--might best be carried out in agricultural development projects, given the large degree of integration between livestock production and crop cultivation. The focus of the livestock components would be to help support veterinary services and to develop methods to assist and promote integrated farming. 1/ Such pilot schemes might be first tried on the more densely populated Central Plateau and could include: (a) the elaboration of training programs for existing agricultural extension agents to give them technical knowledge about animal production; (b) continued experimentation to develop better ways to cultivate fodder, to improve the pasture quality of fallow land, and to conserve hay; (c) financing for animal traction and, possibly, short-term feeding schemes for cattle and small ruminants; (d) efforts to encourage the use of manure as fertilizer; and (e) efforts to make feasible the production of more milk. Such a project would consist mostly of studies, training, and extension, but financing may be required for fencing, village cattle paths, sheds, and credit funds. 10.34 Probably the most sweeping donor action would be a sector loan for livestock. This type of loan would have three purposes: (a) to encourage institutional reorganization; (b) to finance inherent recurrent costs asso- ciated with research, training, and the delivery of veterinary services against major epidemic livestock diseases; and (c) to compensate for the budgetary effects of policy adjustments that appear necessary to increase livestock productivity and to maintain exports. 10.35 Such a sector loan could help the government to reorganize and to finance its livestock services in such a way that: (i) logistical means are adequate for the veterinary service to function, which may also require efforts to strengthen the accountability for the livestock service for revolving funds (e.g., those furnished by PHANAVET); 1/ It is interesting to note that promotion of mixed farming was also one of the main recommendations of the Bank's 1970 economic study on Upper Volta.  - 100 - (ii) animal health services provided to livestock producers are no longer subsidized, except for epidemic diseases; (iii) livestock producers are taught increasingly to provide health services for their own animals; (iv) animal production and husbandry is more closely tied to agricul- tural extension; (v) fodder production programs are better integrated with the agricul- tural service; (vi) livestock research is integrated with the national agricultural research authorities; 1/ (vii) the role of ONERA is more narrowly limited to providing support services for livestock processing and marketing; and (viii) various foreign-financed livestock programs are more closely coordinated and placed in the framework of the development policy to be elaborated. 10.36 Donors could provide recurrent financing for the research programs, training, and vaccinations against epidemic diseases. Such work is probably vital for the future economic strength of the sector, but because the impact is likely to be significant only on the long-term the government can hardly be expected to finance it out of current budgetary revenue. 10.37 Finally, donors could help finance the consequences of some major policy reforms that are considered necessary to rejuvenate the sector. Some of the adjustments might include the following: (a) Export taxes could be reduced to raise domestic livestock prices. The price rise would have two effects: an incentive to expand production (which is unlikely to be very great) and a disincentive to local consumption. At the margin at least, foreign exchange earnings would be won by a sacrifice in govern- ment revenue, which the sector loan would offset. In addition, the higher domestic prices might make some new, more intensive methods of livestock more interesting, thereby encouraging their adoption. (b) Payments could be made to agricultural groups to encourage them to relinquish their standing claims on land in favor of herding groups. Such a transfer of effective ownership could assist traditionally nomadic groups to adopt progressively the 1/ Such integration might be carried out under a separate research project.  - 101 - more intensive techniques that physical and economic constraints call for. It may also be necessary to expand cadastral services to implement and enforce the transfers. In addition, various investments like fencing that could facilitate the reform might need to be subsidized. 10.38 There are specific areas that donors should avoid. These include: (a) Additional group ranches, except whatever marginal recurrent financing may be required to allow the Samorogouan project to continue for a sufficiently long period (10-15 years, or until 1990) in order to judge its efficacy; (b) Commercial feedlots, because they are unlikely to be cost- effective--given the high value of feedstuffs--even under special conditions and with careful management neither of which seems likely; (c) Eradication of tse-tse flies, because high initial and recurrent costs most likely make the endeavor worthwhile only for intensive operations, the profitability of which has yet to be demonstrated; (d) Market improvement, except perhaps designation of cattle paths leading to the markets and creation of public grazing reserves if they can be self-financing; (e) Most commercial processing facilities, until excess capacity is utilized and the case for exporting meat is clearly demonstrated. One exception may be a poultry slaughter and packing facility, although this requires further study; (f) Support to parastatals like ONERA, which appears to be currently engaging in unprofitable commercial ventures and duplicating or competing with the government service.  L'PPER VOLTA LIVESTOCK SUBSECTOR ItEVIEW Table 1. C g-raphic Distribution on livTetoch Population and Agricul,tural products 4 5 5 1 6 78 Orgsnisme Regional Are 2 Villages Population Lie-stock Cattle Sheep -d gosts Cotton - MlIlet,orghum.malze Rice Groundnuts Animal 9 de DeveloPPament (OD) 000 k( (