MSU LIBRARIES ’31:- “1.", .573 ”1-9;;qu u'i . A ‘5 {N\ 813qu W‘IEF_1_N_'—§= P1ace in book drop to remove this checkout from your record. [gfl§§_wé11 be charged if book is returned after the date stamped below. ’ :33 is"): C; THE BEHAVIOR OF LIVESTOCK OWNERS IN SENEGAL: A MICROECONOMIC MODEL BY Chelkh LY A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Economics 1985 ABSTRACT THE BEHAVIOR OF LIVESTOCK OWNERS IN SENEGAL: A MICROECONOMIC MODEL 3? Cheikh by The development of the Senegalese livestock subsector faces technical and economic issues that need micro-level research to provide firm factual bases for sound policy recommendations. After describing the subsector and reviewing the competing frameworks of analysis on the Sub- Saharan livestock production systems, the study uses capital theory to develop an analytical microeconomic model of the behavior of livestock owners in south Senegal. Based on secondary information, the decisions on optimal ages of sale for Ndama cattle (male and female), the herd demography and the impact of different management practices are simulated. The results suggest that producers are optimizing and that technical parameters such as calving and mortality rates have more important effects on herd management than economic parameters such as liveweight prices and costs. Finally, research data priorities are recommended in relation to the FSR methodology used to study the livestock component of farming systems in Senegal. To my parents, Abdoulaye and Madeleine who instilled in me the will. desire and fortitude to succeed at such a task and to all my family. 11 ACKNOWLEDGEMENTS I first extend my deepest sense of appreciation to Drs Stephen Harsh and John Staatz. Dr. Harsh served me well as both my Major Professor and Chairman of the Thesis Committee. Dr. Staatz critically reviewed the thesis and offered valuable input. I am extremely grateful for their encouragement,support and advise. I would also like to express my thanks to Dr. Robert Deans for his constructive criticism and many helpful comments as a member of my Thesis Committee. I am also greatly indebted to Dr. Edgar Ariza-Nino who graciously offered his assistance and let me use his computer program while he was at the Center for Research on Economic Development (CRED) at the University of Michigan. Special thanks go also to the US Agency for International Development (USAID) and the Institut Senegalais de Recherches Agricoles (ISRA) who allowed me to pursue my Master's Program. Finally, I express my grateful thanks to all my friends and colleague students at MSU for their support. iii TABLE OF CONTENTS CHAPTER Page I INTRODUCTION .............................. 1 Problem Statement ..................... 1 Objectives of the study ............... 4 Organization of the study ............. 4 II THE SENEGALESE LIVESTOCK SUBSECTOR ....... 6 Trends in Sub-Saharan Africa ..... ..... 6 Place of the Livestock Subsector in the Senegalese Economy ................ 5 Arguments for Livestock Production Development ........................... 10 Projections 1984-2000 ................ 13 The Development Process .............. 13 Livestock Subsector and Development Policy . ........................ . ..... 17 Goals ......................... 17 Policy Responses .............. 18 The Marketing System .......... 22 Organization ............. 22 Prices 0 O I O O O O I O O O 00000000 24 Government Levies ........ 26 The Livestock Development Projects ...................... 25 The Nouvelle Politique Agricole 29 Guidelines for a Livestock Development Policy ............................... 31 Constraints ........ ... ......... 32 The Ndama Production System.... 34 iv Description and Distribution of the Breed 34 Performance.............. 35 Production Environment... 37 Policy Guidelines . ............ 38 Technical Aspects ....... 38 Economic Aspects ........ 40 III CONCEPTUAL APPROACHES OF LIVESTOCK PRODUCTION ............................... 44 Introduction ....... ...... ........... . 44 The 'Stategies for Security' Interpretation ....................... 45 The 'Commons' Framework .............. 47 The Marxist Anthropoloy Analysis .. ..... ....... ............... 51 The 'Repository of Value' Framework ............................ 54 The 'Production Rationales' Framework ........... ... .............. 56 IV ECONOMIC MODELING OF LIVESTOCK OWNERS' BEHAVIOR ......................... 62 Livestock Production Systems Modeling ............................. 62 Decision—Making Analysis of the Livestock Small-Holders .......... 64 Problems With the Assumptions on Decision-Making ............ 64 The Supply Response of Herders. 66 Capital Theory modeling ..... ......... 68 Introduction ...... ............ 68 Theoretical Bases . ............ 71 Mathematical Model ............ 74 V PRESENTATION OF THE MODEL AND THE SIMULATION METHODOLOGY ............................ 78 Characteristics of the Model ......... 78 Description of the Model ........... .. 80 Optimal Age of Component ...... 80 Equation ................... 80 Decision Rule ............ .. 81 Data . ...................... 83 Liveweights ........... 84 Mortality Rates ....... 87 Calving Rates ......... 89 Milk Yield and Values ..... . .......... 91 Meat Value .... ........ 92 Costs of Production ... 94 Interest Rate ......... 97 Herd Structure Component ...... 99 Block Diagrams ...................... 100 Computer program .................... 101 VI RESULTS OF THE SIMULATION AND SENSITIVITY ANALYSIS ................................ 10? Optimal Age of Sale and Herd structure 107 Base Analysis Results ........ 107 Validation ... ..... . .......... 111 Age of Sale Component of the Model .............. 112 Herd Structure Component of the Model .............. 113 Sensitivity Analysis ................ 116 Internal Consistency ......... 117 Mortality Rate Changes .... 118 Calving Rate Changes ...... 118 Price and Cost Level Changes ...... ..... . ....... 118 Rate of Interest Changes .. 125 Evaluation of Livestock Development Projects ......... 128 vi Changes in Mortality and Costs ................ 129 Changes in Calving Rates and Costs ..... ........... 132 Changes in Price and Cost Structures ...... 135 VII RESEARCH IMPLICATIONS AND SUMMARY ........ 139 Introduction ................. . ...... 139 The Senegalese Research Context ..... 140 FSR and Livestock Production Systems ..... ........................ 142 Specific Problems in Research on Animals ........... ..... ... .......... 143 Synopsis of Research on Livestock Production ................. . ........ 145 Summary ............................. 149 ‘APPENDICES 151 BIBLIOGRAPHY 163 vii LIST OF TABLES Page Table 1. LIVESTOCK NUMBERS Average growth rates, Senegal (1960 — 1980) 9 Table 2. MEAT PRODUCTION Average growth rates, Senegal (1960 - 1982) 9 Table 3. PLACE OF THE LIVESTOCK SECTOR IN THE ECONOMY 10 Table 4. LIVESTOCK TRADE Quantities traded in MT, Senegal,1980 10 Table 5. AVERAGE INCOME ELASTICITIES 0F DEMAND FOR SELECTED CROP AND LIVESTOCK PRODUCTS IN WEST AFRICA 13 Table 6. PROJECTIONS 0F BEEF,VEAL,MUTTON AND GOAT MEAT CONSUMPTION AND CATTLE, GOAT AND SHEEP INVENTORY REQUIREMENTS . SENEGAL . 1984-2000 15 Table 7. COMPOSITION OF THE DSPA BUDGET Senega1,1965-1978 21 Table 8. GROWTH DATA FOR THE GOMPERTZ CURVES (Ndama cattle; Male and Female;Senegal) 88 Table 9. MORTALITY RATES (Ndama cattle; Male and Female;Senegal) 90 Table 10. CALVING RATES (Ndama , Senegal) 92 Table 11. PRICES (Liveweight prices; Casamance; Average from 1978-84) 95 Table 12. AVERAGE ANNUAL INTEREST RATES (Informal markets; 1982-83; Senegal) 99 Table 13. WEIGHTS AND EPV OF A MALE (Base analysis results from 1 to 15 years) 110 viii Table Table Table Table Table Table Table Table Table Table Table Table Table Table 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. WEIGHTS AND EPV OF A FEMALE (Base analysis results from 1 to 15 years) HERD STRUCTURE : SEX COMPOSITION (Model output and comparative data) HERD STRUCTURE : AGE COMPOSITION (Model output and comparative data) PRODUCTION PARAMETERS (Model output and comparative data) EFFECTS EFFECTS EFFECTS EFFECTS EFFECTS EFFECTS CHANGES CHANGES CHANGES OF OF OF OF OF OF IN IN IN MORTALITY CALVING RATES LIVEWEIGHT PRICES COSTS MILK PRICES THE DISCOUNT RATE MORTALITY AND COSTS CALVING RATES AND COSTS PRICE AND COST STRUCTURES STAGES IN LIVESTOCK PRODUCTION RESEARCH ix 111 116 116 116 123 123 125 125 126 128 132 138 138 147 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 10. 11. 12. 13. 14. 15. 160 17. 18. LIST OF FIGURES BREED DISTRIBUTION IN SENEGAL WEIGHT CURVES 0F NDAMA (Male and female) MORTALITY RATES (Age specific — Male and female) CALVING RATES (Age specific) PRICES 0F LIVEWEIGHT KILOGRAMS (Males and females) MARKET VALUES (Meat values) FUNCTIONAL FLOW DIAGRAM OF CATTLE PRODUCTION CAUSAL MAP OF CATTLE PRODUCTION DIAGRAM OF AGE OF SALE DYNAMICS DIAGRAM OF HERD DYNAMICS NET EXPECTED PRESENT VALUES EFFECTS OF MORTALITY RATES EFFECTS OF CALVING RATES EFFECTS OF PRICES EFFECTS OF DISCOUNT RATE CHANGES IN MORTALITY RATES AND COSTS CHANGES IN CALVING RATES AND COSTS CHANGES IN PRICE AND COST STRUCTURES Page 37 87 89 91 94 96 103 104 105 106 109 120 121 122 127 131 135 137 CHAPTER I INTRODUCTION Problem Statement The traditional livestock production systems based on pastoral grazing methods are in various phases of disruption. Affected by the general constraints to food production and economic development, they also have a unique set of problems confronting them. They are threatened by weather fluctuations and desertification. In addition, there are periodic fodder shortages, which are aggravated by overstocking and lack of managed grazing patterns. Besides these intrinsic hindrances, the traditional systems are losing in their competition with cash and food crop production systems (Simpson and Farris,1982). Such a situation characterizes the Sahel region and its neighboring regions. These problems have been compounded by the lack of sound regional policies in an environment where the social and economic practices and values that were introduced by the colonial rulers and their successors have been markedly changing. In particular, in Senegal, the livestock subsector is characterized by a price policy which subsidizes the cost 1 2 of meat for urban dwellers at the expense of incentives for producers. Furthermore, despite recent attempts to reorganize the agricultural sector, there is still a great need to be more specific about the policy decisions that need to be taken to encourage the development of livestock production and marketing. In recent years, the structure of the livestock subsector has been changing from nomadism to nomadic pastoralism and mixed farming systems (Eicher and Baker,1982). Research on the economics of traditional livestock production systems is developing, and there is an urgent need to create a knowledge base supported by a structured research program in order to provide a firm factual basis for policy recommendations. Numerous conceptual frameworks of analysis have been proposed for the understanding of African livestock production systems and their dynamics. These frameworks need, however, to be related to the specific environment where research must be done and the policy decisions that are being considered. In addition, besides the ethnological and environmen— tal studies that have been conducted, additional research is needed, especially at the micro level. By addressing the technical and economic issues faced by the livestock owner, an integrated research program will give insight into the micro-macro linkages that have important implications for policy and provide a better understanding 3 of the livestock producer decision-making process. This is of particular importance in Senegal where livestock subsector development is planned on the basis of a stratification policy. Generally, stratification policies aim at a differentiation in land use, management systems and production methods, by geographic area (such as extensive grazing in one area, extensive crop production, mixed farming, reforestation and range conservation in others). In the case of livestock production, stratification is done according to the comparative advantages of the different socio-economic and ecoclimatic zones in the region whose production systems are being reorganized.In beef production,stratification leads to separate layers or stages. These include breeding cow-calf herds, growing out, fattening and processing (Ferguson,1976). It is, therefore, critical to determine the economic differences among the production systems in various zones and their consequences for the livestock producer. Thus, as a contribution to the organization of research, it is useful to build an analytical model of the decision-making process at the producer level. Such a model permits us to determine what data are needed for a better understanding of the current and future economic and social changes in the Senegalese livestock production systems. 4 Objectives of the Study The objectives of this study are as follow: 1-To describe the Senegalese livestock subsector, its current trends and its policy problems from a macroeconomic perspective; 2-To review the various proposed approaches of the Sub-Saharan livestock production systems and compare their strengths and weaknesses; 3-To develop an analytical model which can help economic analysts understand livestock producers' management decisions and examine the impact of different management practices; 4-To use the analytical model and test its usefulness in understanding the livestock component in farming systems and to identify research data priorities;and, 5-To link the findings with the development of policy recommendations for the Institut Senegalais de Recherches Agricoles (ISRA). Organization of the Study The work is presented in 7 chapters. Chapter 2 describes the Senegalese livestock subsector , its development and its policy problems. This chapter also contains a background description of the Ndama production system that is modeled in subsequent chapters. Chapter 3 reviews alternative conceptual frameworks that have been 5 proposed for analysing livestock production systems in Africa and suggests a synthetic framework for the economic modeling that is treated in Chapter 4. Chapter 5 presents the model and data used. In Chapter 6, the results of the simulation and the basis for its validation are exposed. Finally,the research policy recommendations are presented in Chapter 7. CHAPTER II THE SENEGALESE LIVESTOCK SUBSECTOR Trends in Sub-Saharan Africa Livestock production in Sub-Saharan Africa experienced a significant growth between 1950 and 1970. The production of meat and milk increased by 2.3% and 2.1% per year respec- tively. However,these increases did not keep pace with the annual human population growth of 2.4%. This situation led to a decline in per capita production (Jahnke,1982). Studying the same trends from 1960 to 1980,Anteneh reaches the same conclusion: " The livestock output situation in Sub-Saharan Africa is clearly very serious.The situation is doubly alarming in that not only has the rate of output declined over the last two decades, but also the percentage output growth attributable to productivity (i.e.,yield growth) has decreased." (Anteneh,1984b,p.15) Place of the Livestock Sector in the Senegalese Economy An examination of the statistics available shows the same trends in Senegal. During the French colonial administration, the livestock subsector was not a priority item in a setting entirely devoted to the economie de traite of groundnuts. The sector was of minor importance to economic 7 development and the growing involvement of farmers in the cash economy. The drilling of deep—bore wells allowed a movement of the livestock away from the expanding cropping areas, especially in the Sylvopastoral region.Annual vaccination campaigns were organized and resulted in a rapid increase in livestock populations. Research was done on the Zebu gobra at the Dahra station but with no real application of the results. The livestock subsector was further stifled by a government head tax placed on livestock. Following Independence in 1960, no major policy changes have taken place despite the steady annual contribution of the subsector to the GNP. From 1959 to 1971, the value of output was increasing by 2.4% p.a. in constant prices_ while the overall value of agricultural production was increasing annually by only 1.9% with very erratic move- ments (IBRD,1974). During the ten-year period following Independence, the subsector did not exhibit any major change in production or marketing techniques despite favorable conditions and other factors. The population of cattle increased at a rate of 4.9% p.a. while that of small ruminants increased at 5.1% p.a. (Tables 1 and 2). After 1970,the situation became serious, with successive herd contractions due to severe droughts.The extensive system of production based on perennial and 8 annual grass rangelands ceased to perform adequately. From 1970 to 1980, the annual growth rates in numbers of cattle and small ruminants were -1.4% and 1.7% respectively (Table 2). These trends did not really change drastically the contribution of the subsector to the economy ( Table 3). The value added to the overall economy by the livestock subsector was 4.5% in 1960. By 1975,it increased to 7.9% and reached 9.3% in 1980. The share of the value added by the livestock subsector to the total agricultural sector was 24.5% in 1970. It decreased to 19.9% in 1975 due to the successive droughts and then increased to 25.1% in 1980. Such specific trends reflect an increasing supply of animal products. However, Senegal has become increasingly a meat and milk deficit area since the 1960's. Due to a continuous growth in the purchasing power of consumers and the urban population,there has been an important increase in effective demand. This is especially true in Dakar. The domestic production was predicted in the 5th Plan of Economic and Social Development to satisfy only 70% of the meat demand in 1980. The government has been, therefore, spending foreign currency by importing milk, meat and live animals (Table 4). Table 1. LIVESTOCK NUMBERS Average annual growth rates (%),Senegal(1960-1980) Periods Species 1960-1970 1970—1980 1960—1980 Cattle 4.9 -1.4 1.4 Goats,sheep 5.1 1.7 3.9 Pigs 14.2 5.5 13.8 SOURCE: Appendix A Table 2. MEAT PRODUCTION Average annual growth rates (%),Senegal(1960-1982) Periods Species 1960-1970 1970-1982 1960-1982 Beef 7.1 -0.2 3.0 Goat-mutton 3.3 1.0 2.3 Pigmeat 10.0 4.2 9.1 SOURCE:Appendix A 10 Table 3. PLACE OF THE LIVESTOCK SECTOR IN THE ECONOMY -N‘L—riu-u . r " ‘ ' 3" GDP =- . | ... PROPORTION OF AGRICULTURE IN 3 . m I AND PROPORTION OF LIVESTOCK IN ‘3’ « ' : AGRICULTURAL GDP IN TROPICAL _ 5° mi . AFRICAN COUNTRIES I980 g ° : em a 1 l gm 3 1 m: .0" < l -1». LNG moon MLVI MALAwI _. ‘0. "C 9“:wa 0. .suO BEN m M02 Mom ,3, . ...... «I: mm man In» MAI-mum m ‘ | cm can cm NOA mum ‘ . ' CAR mm ROR men .5 1““ L :32.” £3-12: ..-- «"";" ""‘T “’"m ""°"".'ifi"-"I§§’Eo~'ao ' sen "new 9,, ‘ 3 m : ETH ETHIOHA sou sums ‘ T: 1 ' I 5" MS mom :00 Sum ' as GM! we» as mama. anT an 5 ’Pr— UNA owe me mo ‘ 4 : OUI any» INA «nan ‘ - ' Ivc Ramona mw unuwua I : ,1»- Keu mm 2» 2m ‘ I Ll. LIBERIA 2AM ZAMBIA Ow-Gfl : mm ummwnn an zmumn " J o?» I MU mu : “mm“ ' : Wang. 1 i uwmmouRgmgmmymgr "WW ~~f 20 '3'. '3‘ ““107 'is' 's'o'3's 7‘01 'o'f'vo 15 Iowa SOURCE: Jahnke (1982) Table 4. LIVESTOCK TRADE Quantities traded in MT,Senegal,1980 SPECIES IMPORTS EXPORTS NET EXPORTS Cattle 25,000 140 -24,860 Goats-sheep 200,000 10,000 -190,000 SOURCE : FAO Trade Book 1981 11 The data base used to describe the livestock subsector is presented in Appendix A. It has been very controversial with respect to accuracy and consistency. Therefore,the main purpose of the present analysis is to identify overall trends for a better understanding of the problems. As stated by Jahnke (1982): " In summary, experience with livestock development to date is disappointing. Agricultural development as a whole has fallen behind overall economic growth and the shortfall is particularly serious for the livestock sector. The per caput availability of livestock foods in Africa has not improved and whatever increase in production it is mainly due to herd and flock increase rather than productivity increase.Traditional production systems have simply expanded maintaining traditional techniques." ( Jahnke, 1982, p.46) Argggents for Livestock Production Development The priorities and basic requirements in a balanced human diet are subject to controversy. Crawford (1975) argues that many of the development and aid policies that have favored the increase in protein production may not have been based on sound reasoning. He further argues the world food problem is more likely related to energy nutrient shortages than protein shortages. On the other hand, Raimbault (1980) argues that we have a combination of protein and energy shortages which may affect as much as 75% of the world population. Furthermore, the Food and Agriculture Organization (FAO) offers the following assessment: " There is ..... a general agreement that the increase in calorie intake should not always be 12 obtained from an increase in per caput consumption of cereals, of starchy roots,which would still imbalance the diet but should come from those foods which would give a better balance to the present diet, namely, fruits, vegetables, pulses and nuts on the plant side, and still more important,the food of animal origin--milk, meat, fish and eggs." ( FAO, 1976, p.1) FAO also recommends that by the year 2000, individuals in developing countries should be able to consume 2400 calories per day . Total daily protein intake should at least be 70 grams, with at least 20 grams being animal protein (FAO, 1976). These statements stress from a nutritional viewpoint that the production of animal protein is a necessity for developing countries in meeting nutrient needs. In addition, from ecological and efficiency viewpoints, livestock production, especially with ruminants, allows for beneficial use of wastes, byproducts, vegetation and land that are marginal for crop production. Finally, the importance of the livestock industry is supported by the growing demand for livestock products With the increased urbanization in Senegal--32% of the population in 1976 lived in cities-- there has been an increase in family income level and a corollary increase in the demand for livestock products. Josserand (1983), for example, notes that in West Africa, the ratio of total meat consumption per capita in urban versus rural areas is 4 to 1. Furthermore, relatively high demand elasticities with respect to income have been shown, such as 1.08 for meat 13 and 1.23 for milk (Table 5). In fact, Jahnke(1982) believes that the real income elasticities for livestock products are substantially higher than stated. " An open question is the extent to which the income elasticities are overlain by price effects. Particularly in the case of livestock production, the shortfall in supply has led to rapid increases and curbed demand accordingly. There is reason to assume that the income elasticities of demand thoroughly corrected for that effect would be higher, possibly substantially higher." ( Jahnke, 1982. Pp 48-49) Table 5- AVERAGEa INCOME ELASTICITIES 0F DEMAND FOR SELECTED CROP AND LIVESTOCK PRODUCTS IN WEST AFRICA Region Cereals Meatb Milk Eggs Sahel .24 1.04 .53 .90 West Africa .21 1.08 1.2 1.2 West Africa (Less Nigeria) .33 1.08 1.23 1.1 a/weighted according to share in overall consumption of food products ; average of projected elasticities over the period 1975-2000. b/excluding edible offals;including game SOURCE: Jahnke after ILCA (1981) Applying "conservative" coefficients, Josserand (1983) projects the annual rate of growth in per capita demand in West Africa from 1975 to 2000 at : 1.2% for meat, 0.8% for milk and 2.2% for eggs. Such data are not currently available for Senegal and could only be derived from further study. 14 Pro ections 1984 - 2000 It is important from a policy making standpoint to base a long-term vision on projections about possible out- comes. With respect to the serious situation of the Senegalese livestock subsector, such projections enable us to perceive with more detail the long—term trends and relative changes required to satisfy the projected needs. In this study, the projections to the year 2000 are made according to the methodology followed by Simpson (1984). They permit a long term vision that highlights the urgent need for production improvements if, at a minimum, the current level of consumption is to be maintained. Table 6 presents in a summary form the projections generated from the currently available data. The projections show that important change in production patterns and intensity must occur to satisfy the coming needs in Senegal. The Development Process Following Gallais'(1972) viewpoint, it is useful to recall the major phases of the socio-historical evolution of the agropastoral societies in West Africa. 15 Table 6. PROJECTIONS OF BEEF. VEAL, MUTTON AND GOAT MEAT CONSUMPTION AND CATTLE, GOAT AND SHEEP INVENTORY REQUIREMENTS . SENEGAL . 1984-2000 Items Units Values Per capita consumption,1980 Beef and veal kg 5.6 Mutton,lamb and goat meat kg 1.8 Total population 1980 Millions 5.9 2000 Millions 10 Total consumption,1980 Beef and veal 1,000 MT 33.04 Mutton,lamb and goat meat 1,000 MT 10.62 Total consumption,2000 Beef and veal 1,000 MT 56.00 Mutton,lamb and goat meat 1,000 MT 18.00 20% increase in per capita (0.9% annually) Beef and veal 1,000 MT 67.00 Mutton, lamb and goat meat 1,000 MT 21.60 Indigenous production,1980 Beef and veal 1,000 MT 29.00 Mutton,lamb and goat meat 1,000 MT 8.00 Ratio of indigenous production as a proportion of total consumption 1980 Beef and veal ratio .88 Mutton,lamb and goat meat ratio .75 Indigenous production per head of inventory,1980 Beef and veal kg 13 Mutton,lamb and goat meat kg 2.5 Inventory,1980 Cattle 1,000 Head 2,238 Sheep and goats 1,000 Head 3,170 16 Table 6 cont'd. Items Units Values Inventory required to keep 1980 ratio of indigenous production to total consumption, 20003.No change per capita consumption Cattle 1,000 Head 3,800 Sheep and goats 1,000 Head 5,400 20% increase in per capita consumption Cattle 1,000 Head 4,500 Sheep and goats 1,000 Head 6,300 Indigenous production per head of inventory required for no increase in inventory to meet consumption projection,2000b,No change per capita consumption Cattle kg 22 Sheep and goats kg 4.3 20% increase in per capita consumption Cattle kg 26.3 Sheep and goats kg 5.1 Annual growth rate in production per head of inventory required for no increase in inventory to meet consumption,2000. No change per capita production Cattle % 2.6 Sheep and goats % 2.6 20% increase in per capita consumption Cattle % 3.6 Sheep and goats % 3.6 a-Total consumption in 2000 divided by indigenous production per head and the result multiplied by the ratio of indigenous production to total consumption. b-Total consumption in 2000 times the ratio of consumption to indigenous production divided by 1980 inventory. SOURCE:Simpson(1984);FAO Production Year Books (various issues); IBRD(1983) 17 Until the end of the last century, there was a permanent adjustment in the human and animal populations through wars, epidemics and droughts. The Sudanic agropastoral economies were first dominated by the Sudano-Sahelian empires of Mali and Sonray (XII to XVI century A.D).Their decline led to a strengthening of the pastoral societies living at their margins. These pastoral societies were typically Fulani, having a high degree of organization in land use and a close integration of pastoralism and agriculture (Swift,1979). During the reorganization of the economy that accompanied colonization various changes occurred. First, crop production areas were extended without an integration of crop and pastoral techniques.This situation led to ecological disasters such as bush fires, deforestation, reduction of fallows, wind erosion and uncontrolled drainage. Second, one-sided development attempts (animal health and water supply) without concommitant land and pasture improvements created a disequilibrium in the‘ pastoral systems. Finally, an important development in human and animal growth was made possible by medical breakthroughs, especially the control of the rinderpest. The major economic, social and political problems faced by herders reached their most intense level with the 1970 drought, whose consequences are still being felt. However, as Ware (1979) points out, herders in Senegal 18 have adapted to the drought by devising innovative strategies. In particular, they have diversified their traditional productive activities with crop production and there has been an increased desire to sell milk through non-traditional distribution channels. Is has also been shown that migration southward and toward the urban areas are phenomena of growing importance (FAO,1977). A broad historical review would not be complete without special attention being given to two major factors besides climat that affect the livestock subsector: government policy and development projects. Livestock Subsector and Development Policy Goals The government orientations defined in the 5th Plan of Economic and Social Development for 1977—1981 have been retained in the current 6th Plan (1981-1985).The goals related to the livestock subsector include - a reduction of the strong dependency on inter- national markets to satisfy the internal demand for meat and milk products; -the development of an integrated system of animal production (mainly cattle and sheep) based on the complementarities and comparative advantages of the eight ecological zones found in the country; 19 -an increase in the per capita consumption of meat from about 13 kg in 1974 to 15.7 kg in 1985. For such a goal,the desired rate of growth per year was set at 2.5% for cattle, 4.6% for small ruminants and 11.5% for poultry; also in 1985, the meat production objective is set at 100,200 MT, with 54,900 MT beef and 10,000 MT mutton-goat, the remainder being poultry and pigmeat (Societe Africaine d'Edition,1983). These goals were assumed to help satisfy the growing urban demand, reduce the protein gap for the poor and assure a reasonable farmer income. Policy Responses Surveying the problems raised by African government policies toward the livestock and pastoral subsectors, Sandford (1981) concludes that they have been defective because of either wrong, contrary or non- existing policies with respect to the land use, the pricing of commodities and the overall organization of the sector. The Senegalese case confirms that general critique since the livestock sector has not been among the priorities in terms of investments, policies and organization. For instance, from 1965 to 1971, the subsector received only 1% of the government's current budget.In constrast, the other rural sectors received 7% of the budget (IBRD, 1974). Futhermore, although the operating 20 budget of the Direction de la Sante et des Productions Animales (DSPA) grew at a rate of 6.6% p.a. from 1965 to 1979, only 2.1% of the government's investments was for animal health and production related expenditures in 1979, even though in that year the share of livestock in the contribution of agriculture to the GDP was 21%. Paradoxically, a sharp decrease occurred after 1973 (in the drought period) when taxes on cattle at the producer level were abolished. As can be seen by studying the composition of the DSPA budget in Table 7, there has been a significant decrease in non-staff expenditures from 1965 to 1978. Operating and investment expenditures represented an extremely low percentage of the total national expen- ditures in those categories. Operating expenditures for livestock were 0.8% of the total operating expenditures and investment expenditures for livestock were 0.1% of total investments (Anteneh,1984a). In the 6th Plan, 17.5% of total investment in agriculture has been planned for the livestock subsector with 0.1% going to the traditional government livestock, agencies and 16.4% going to the development projects (Societe Africaine d'Edition,1983). Such trends have placed impossible financial burdens upon the public agencies despite their important roles in providing yearly vaccinations. Furthermore, the lack of equipment and personnel leaves the DSPA unable to intervene adequately 21 at the level of the producers and, most important, give strength to their cooperatives. Table 7. COMPOSITION OF THE DSPA BUDGET Senegal,1965-1978 Years Staff budget(%) Non-staff budget(%) 1965-66 69 31 1968-69 65 35 1969-70 77 23 1970-71 75 25 1971-72 70 30 1972-73 74 26 1973-74 80 20 1974-75 83 17 1975-76 84 16 1976-77 85 15 1977-78 85 15 SOURCE : Anteneh (1984a) The producer cooperatives have been in existence since the early 1950's, but they remained very marginal until the 1960's. At that time, they developed with the Mouvement cggperatif inspired by the government. In 1971, 6,355 individuals belonged to 86 cooperatives.However, these cooperatives did not have any real impact on the supply and marketing of meat. N'Diaye and Ba (1972) attribute the failure of the livestock cooperatives to several different reasons. Their structure and functioning were not appropriate since they were replications of the crop cooperative schemes. At the beginning, created under the control of the Office National 22 de Cooperation et d'Assistance au Developpement (ONCAD), they became the responsibility of the DSPA in 1972. A major constraint was the cooperatives' difficulty in obtaining credit, unlike the crop—oriented cooperatives, which were fully supported by the credit system embodied in the Programme agricole. Moreover, the Banque Nationale de Developement du Senegal (BNDS), the main supplier of credit to the rural sector, had a guarantee system for cattle raisers' cooperatives, butchers and traders identical to that for small entrepreneurs in the industrial sector (IBRD,1974). This system required important collateral assets and charged higher interest rates than those charged crop cooperatives; these practices greatly reduced incentives to invest in the livestock subsector. Finally, the cooperative participants were unable to create a lobby for their interests because of their dispersion throughout the country and their lack of information. Since the 1970's, more efforts have been directed toward the development projects and the parastatals mana- ging them (Societe de Developement de l'Elevage dans la Zone Sylvopastorale (SODESP), Projet de Developpement de l'Elevage au Senegal Oriental (PDSEO), Projet USAID de Bakel). These institutions are expected to play an important role in the stratification of the livestock production system thanks to a differential pricing system. Besides these projects, there is a heavy reliance 23 on foreign aid for the supply of feedstuffs in case of emergency. For example, in 1984, the Operation Sauvegarde du Betail permitted the government to provide 68,000MT of feedstuffs in emergency areas at a price of 15 FCFA, which represented solely the cost of transport. Although this program provided temporary relieve for drought stricken livestock producers, such actions are not a serious remedy and are not a viable solution in the long run because they do not address the basic feed constraint in the country. In terms of price policy, price control measures are employed at the meat retailer level. That aspect of the public sector policy is analyzed in relation to the marketing system in the following section. The Marketing System Organization One of the research priorities of ISRA is to conduct an extensive study of the livestock marketing system. Except for some regional analyses (SODESP,1984; SOMIVAC,1982), there is a lack of information on the performance and efficiency of the marketing system, its structure and the conduct of its participants. However, because conditions throughout West Africa are similar, the available description of the livestock marketing systems in the rest of the region may be used 24 when studying Senegal (Shapiro,1979;Herman,1980).Particular emphasis is given here to the factors affecting producers. The traditional marketing system satisfies 90% of the demand in the main consuming area in Senegal, Dakar, and entirely meets demand in the other areas of the country. Numerous problems affect the efficiency of the system. Conflicting relations between the government and the meat suppliers have created many boycotts by the latter during the last decade. The strong interventions of the Societe d'Exploitation des Ressources Animales du Senegal has, without doubt, various effects on the competition and conduct in the sector since the beginning of the 1980's. An intense development of black markets has been noticed as well as the subsequent worsening of health hazards for the consumers. Each year, there is also an explosion of prices and a great strain on the marketing system for sheep at the Moslem holiday Tabaski. A brief analysis of the system shows also different levels of integration and complementarities among the producers,the intermediaries ('diula','tefanke') and the butchers.The 'diula' or bush collector is supplied directly by the producers or at small collection markets. These weekly markets play a growing role nowadays. The animals are then driven to regroupement markets , from where they are then sent to the terminal markets of the major consuming areas. The 'tefanke' plays a more 25 important role in the last parts of the marketing Channel. All slaughterings must in theory be conducted under government supervision. The meat is provided to wholesale butchers, retail butchers and peddlers of raw and cooked meat. Complex interrelationships exist among the supplies of beef, lamb, mutton, goat meat and fish, a major substitute. Analyzing the marketing of small ruminants, Josserand (1983) shows that supplies of mutton and goat meat have a countercyclical pattern when compared to the supply of beef. Due to more frequent offtake, sheep and goat meats usually "take up the slack in domestic beef supply" (Josserand, 1983, p.7). Prices This analysis is based on the main consuming area, Dakar. After a period of stability, the prices of cattle, goats and sheep increased by 60 to 80% between 1967 and 1971 because of feed shortages during the drought period and the decrease in livestock coming from Mauritania at that time (IBRD, 1974). That period corresponds with the lowest producer prices ever recorded, 50 to 55 FCFA per kg as producers tried to liquidate their dying herds. However, the important rates of mortality greatly reduced the number of animals supplied to the Dakar market and caused shortages that led to high consumer prices. With respect to the current situation, the most important 26 characteristic of the marketing system is the official ceiling on retail prices for meat that is set by the Ministry of Commerce and the Ministry of Rural Development. Two sets of prices are fixed for beef and mutton: wholesale prices ('cheville') and retail prices (SODESP, 1984). Due to religious factors and consumer's preferences, there is a premium for lamb and mutton, which are more expensive than beef. Conversely, there are no price guarantees at the producer level. Therefore, the price fixing and the carcass weight force the wholesaler to buy live animals at very low prices to obtain any safe marketing margin. In 1984, prices averaged 350 FCFA/kg liveweight for fattened animals and 250 for regular animals at the terminal market of Dakar. Such prices determined producer prices estimated at 250 and 200 FCFA, respectively (SODESP, 1984). At the retail level, traditional butchers are officially allowed a margin of 12% over the wholesale price. SODESP (1984) considers, however, that the value of offals covers the processing and storage costs at the wholesale level. The modern butchers, who are allowed higher retail prices (averaging 2050 FCFA/kg) get a margin of 46% over the wholesale price which is fixed at 1100 FCFA/kg. An important research topic is to determine how these sets of prices affect the supply of both live animals and meat. In effect, following Girardot-Berg 27 (1982), it can be expected that the official retail prices are too low in relation to the free market prices and their market clearing levels. Such a situation is very likely accompanied by other problems: - the depression of the producer prices is correlated with disincentives for increased supply and production improvements; . - there is a transfer of revenues from producers to urban consumers and a concommitant excess demand with respect to the available supply; - retailers, especially traditional butchers, invest less in hygiene and quality because of low profits from trade and processing; and, - defined weights and standards of quality aiming at improving the efficiency of the system are generally opposed by the participants because they may reduce their marketing margins. Government Levies Cattle head taxes were abolished after the 1970 drought due to, according to the government, welfare considerations. However, other probable reasons included the practical administrative difficulties and irregu- larities in the collection procedures; and the counterpro— ductive effects on censuses, vaccination campaigns and other development efforts. Anteneh (1984a) shows, however, that regardless 28 of the reason, the abolition gave livestock services a weaker claim on available funds and, thus, caused a sharp drop in the financing of the subsector.Meanwhile, other government levies are still in practice. They include import and export duties on live animals; trade licence and market fees for butchers, wholesalers and traders ; slaughtering fees; sanitary taxes; veterinary certificates; and treatment costs and vaccination charges except for rinderpest and bovine contagious peripneumonia. Even though a systematic study of these levies has not been done in Senegal, it is very likely that some have adverse effects on the efficiency of the marketing system, as shown by Herman (1981) for a very similar situation in Burkina Faso. Beside the retail price ceilings, the livestock development projects are major elements of the government intervention in the subsector. The Livestock Development Projects At present, the livestock development projects are a major tool in government policy. Since the six-year period from 1968 to 1974, there has been an increased interest on the part of international donor agencies in providing technical, managerial and financial assistance in range and livestock development. A very abundant literature refers to the characteristics of and evolution strategy for the resulting livestock programs in Sub- 29 Saharan Africa (Teitelbaum,1975;Stryker,1984;Teele, 1984; Atherton, 1984;Jahnke, 1983;Josserand, 1983;Behnke, 1983). Unfortunately, the projects have been inherently top-down intervention schemes aiming at an increase of commercial livestock production (mainly cattle) through the transformation of subsistence pastoralists into commercial ranchers via herd modifications and improved animal nutrition. The general consensus is that those projects constitute an "unrelieved failure" (Behnke,1983) due to various reasons. For instance, basic misconceptions at the design stage have resulted from focusing solely on land and animals instead of the human populations. In addition, highly capitalized interventions were expected to create social changes and a new economic behavior thanks to a process of "social engineering" that would suspend the normal functioning of the indigenous economic systems and replace them with more modern ones. Two more reasons for failure were the lack of concern on the part of planners for existing and potential interactions between animal husbandry and cropping and the bias they showed for strategies that favored high cattle and meat output rather than more integrated benefits. A fifth reason was their lack of interest in small ruminant development, even though, as Josserand (1983) notes: " One should recognize that investment in small 30 ruminants will show much quicker payoffs [due to biological parameters] than in the case of cattle." (Josserand, 1983) Furthermore, land use schemes were inadequate due to the communal land tenure and the high spatial mobility involved in herding activities (Styker, 1984). Finally, wrong assumptions were made with respect to the role of women in livestock production and marketing. Safilios- Rothschild (1983) shows from various case studies the importance of including women in project design, and the factors related to the women's role in agriculture and 'animal production. The productivity-oriented interven- tions often overshadow the family welfare requirements in the subsistence function of the herd, the returns on women's labor and the sexual differentiation in claims and ownership rights over the herd (Sollod,Wolfgang and Knight, 1984). The Nouvelle Politique Agricole Since 1983, major changes have been occurring in the Senegalese agricultural sector under the name of Nouvelle Politigue Agricole (NPA). The NPA is similar to the emerging policy responses to food problems in many African countries (Christensen and Witucki, 1982). Unfortunately, these policies are due to pressures from international financial institutions, a deep crisis in public finances and a deterioration of the balance of payments rather than strategic policy choices. As in 31 other countries, they are characterized in Senegal by - a reorganization of the marketing system for cash and food crops, with a larger role granted to the private sector in marketing agricultural inputs and products ; - a new definition and role for the producer cooperatives, with less reliance on public rural credit agencies;and - a pricing differential in favor of production and consumption of local cereals that includes increased retail prices , a phasing out of subsidies and increased fficial producer prices for cereals above world market prices. However, the available literature and information make no reference to the livestock subsector, despite its importance. It appears that, in terms of policy definition, crop production remains the sole preoccupation despite the need for more consistency with respect to the overall agri- cultural sector and each of its components and despite the potential that integrated farming could have on crop pro— duction. More attention must be given to the livestock sub- sector due to the constraints slowing down its development, its current dynamics and ,above all, the goals set in the government planning. An attempt is made here to suggest a consistent global perspective on the issues related to the subsector based on the international experience as seen through the literature on livestock development. 32 Guidelines for a Livestock Development Program Four basic goals must be set for any national food policy according to Timmer, Falcon, and Pearson (1983). First, an efficient growth in the food and agricultural sectors must be assured. Second, an improved level of income and employment ought to be guaranteed. Third, the nutritional status of the population should reach a minimum acceptable level. Finally, food security has to be properly handled. These goals imply considerable policy trade-offs between producer income and consumer welfare, between fiscal restraints and government subsidies, and between public interventions and market forces, especially for the livestock subsector (Girardot-Berg,1982). The complexities of the trade-offs increase when combined with the differential weighting of short-run and long-run requirements. It becomes crucial to develop tools and an analytical framework for doing the sectoral analyses that are the foundation of a consistent food policy (Timmer, Falcon and Pearson, 1983). It is, therefore, important to concentrate in this study on the development constraints in the Senegalese livestock subsector. As background context, the Ndama production system of the Casamance region is presented with its main characteristics. This information is used later in the simulation exercise to verify the validity of the microeconomic model proposed in this 33 study. After describing the Ndama production system we will address ways of improving livestock policy and actions that could be initiated for the necessary development of the subsector. Constraints " Livestock production and livestock in Africa rest squarely on a combination of ecology, economics, biology, and sociology." (Pratt, 1984, p.39) We cannot analyze here all the constraints that are specific to the Senegalese livestock subsector. A general overview is presented , based on the literature on the livestock subsector in West Africa. A general summary of the important constraints facing the subsector is presented by Atherton (1984). The specifics of how these constraints are manifested in any other country requires further investigation. The constraints cited are: - inadequate government policies and programs; - an inappropriate information base for technical packages, particularly with respect to range management; - a paucity of animal feed; - inadequate animal health services;' - difficult access to markets; - limited skills for new technology adoption ; a bad understanding of the existing system rationales; 34 - the lack of clear definitions in economic terms of feasible technical packages applicable on a general basis; and - poor institutional capacities. Other constraints related to the overall situation in the agricultural sector may be added. There are increasing labor shortages. The general low productivity of the agricultural base combined with the large exogenous capital inflows creates relatively high returns to urban labor, thereby favoring a fast rate of urbanization, such as in Dakar, with a growth rate of 4.8% per year. Urban lobbies and other organized political forces favOr agricul- tural taxation and income transfers from the rural sector (Dumont and Mottin, 1982; Delgado and Mellor, 1984; Christensen and Wituki, 1982). Besides these constraints, new phenomena are occuring that introduce greater complexities. Especially in the case of the Sahel production systems, changes in the structure of the herds are the results of difficult reconstruction phases between the drought years. Also, there is a new geographical distribution of the herds, with more pressure on the southern boundaries of the dry ecological zones. The potential of mixed production enterprises is growing and it gives more importance to animal traction and fattening operations. With the strengthening of the commercialized economy, small ruminants are increasingly a source of cash. In greater 35 numbers, nontraditional owners, who have strong profit- maximization motivations, are entering the subsector. Finally, changed price ratios between animal products and cereals may lead to a new pattern and nature of exchanges in the rural economy (FAO,1977). These general constraints mark the Senegalese livestock production systems, in particular, the Ndama production system existing in the south of Senegal. An overview of this specific system is provided in the following section. This description will be used as the background for the microeconomic modeling of the Senega— lese livestock owners‘ behavior in Chapter V and VI. The Ndama Production System Description and Distribution of the Breed. The Ndama is a Bos taurus breed derived from the Hamitic longhorn ancestral type. Its present origin is the Futa Djallon. Despite some phenotypic variability between 'Ndama petite' and 'Ndama grande' in Senegal, the Ndama is a single breed (ILCA, 1979). According to Doutressoule (1947), the Ndama increased rapidly in the periods following the rinderpest epizootics of 1890-91 and 1918. At these times, the Ndama replaced herds of other types in Senegal and the French Sudan. The actual total population in Africa is estimated around 3.5 million head (Starkey, 1984). In the Gambia, the south and southeast of 36 Senegal, the Ndama is the predominant breed, after an intermediate belt of Djakores, a mixed breed of Zebus and Ndamas. The distribution of Ndama is superimposed on that of the Tsetse flies (Glossina spp.) (Figure 1). It is estimated that 46% of the Senegalese cattle population is composed of Ndamas and Djakores. Performance. The main feature of the Ndama is its reduced susceptibility to trypanosomiasis, hardiness, survivability and water conservation ability (Starkey, 1984). In comparison with the Zebu, the Ndama shows significant production potential,,more biological efficiency in its dispersion areas and a high economic profitability in both traditional low input—low output systems as well as improved ones. Furthermore, it has been used widely as a draught animal due to its natural amenability to training and draught work (Starkey, 1984; Doppler, 1980; ILCA, 1979). However, information gathered under station conditions is much more available than under village conditions (ILCA, 1979; ILCA, 1982; Landais, 1983). A great effort has been provided by ILCA to develop an information network about trypanotolerant livestock. In- depth investigations are just beginning in Gambia and Senegal, in particular (Trail, Murray and Wissocq, 1984). 37 MAURITANIE ( "I . ‘\ l 1‘ \l \ . \‘ ) . :‘y‘fil—“rh ‘ / ‘ ‘E?”I ,; ,’-‘-‘\ ‘) :’ [W II' ‘0‘.“ jaws-1g .... “‘~..’a.‘-‘ '4“ é'fl ____ ;"~ #4“ / ‘ZHI ’ / / f I / f "I ‘~ ’ I, . ~C ‘ l | “I" \l‘--‘-'~ I "- -/ A // . W" , III HIllIIII GUINEE GUINEA BISSAU a. I Bounchry of uotue-Infoulml zone W Mmhry "I III“! whore Zebu predominate — Boundary of tone when: «mum-ed caulo prodmuinnlu N'Dnma x Zebu N' Dune Figure 1. Breed Distribution in Senegal Source: ILCA, 1979 38 Production Environment. Several characteristics of the semi-extensive traditional system with Ndama cattle stand out. First, the system allows free range grazing on communally owned rangeland and pastures or seasonal fallows. The stocking and grazing rates are not yet major issues. In Upper Casamance, only 21% of the area is occupied for crop production and the population density is 18.7 inhabitants per kmz (Landais, 1984). Second, there is no management of rangeland and pasture, nor is there systematic utilization of agricultural byproducts. Instead, the feed needs are met solely through grazing all year round, with short range transhumance patterns before and after the cropping season. The Fulani ethnic production system is dominant, and the cattle are used for a mixed dairy-beef—draft- manure enterprise, which is essential for the subsistence economy. The cattle population is estimated at 280,000 head (Landais, 1984) and a very low offtake rate (Dumont and Mottin, 1982). The productivity and performance are low despite the good genetic potential. In part, this is because of a poor management that increases susceptibility to diseases. These diseases include trypanosomiasis, with various levels of geographical infestation density; fascioloses; brucellosis and anthrax. The annual growth rate of the herd is estimated at 1.5% (Landais, 1984). In addition, besides the yearly vaccination 39 campaigns organized by the government animal health services and some actions toward ox-cultivation by the Societe pour le Developpement des Fibres Textiles (SODEFITEX), there is no real livestock development project in the area. Finally, station research is done by the Centre de Recherches Zootechniques de Kolda as well as some on-farm general investigation (Gueye, Ly and Diallo, 1981). Policy Guidelines For enumeration purposes and a better integration, Jarvis(1982) argues for the categorization of any livestock development policy into two sets of policies: technical aspects and economic aspects. Technical Aspects Ecological constraints remain the most basic and pervasive among the many determinants of the technical problems faced. Effective range management strategies have been tested for protection against drought, for restoration and carrying capacity amelioration (Pratt, 1984). Penning de Vries and Djiteye (1982) state that soil fertility, more than precipitation, is the key factor in rangeland productivity because it is the principal cause of the meagre production of forage and its usual low quality. Furthermore, various options for increasing 40 rangeland productivity while protecting the environment are possible, but under current price conditions they are not economically profitable except in the case of simultaneous intensification with agriculture. The International Livestock Center for Africa (ILCA) also refers to development of animal production thanks to an improvement in the traditional husbandry system, i.e., improved management practices, veterinary care packages and upgraded breeding stock (ILCA,1979). Despite problems such as chronic malnutrition and low quality feeds that permit animal just to maintain a subsistence level of nutrition which restrains the expression of the genotypes, Mc Dowell (1984) shows that prospects are rather promising in terms of appropriate technology. From a technical point of view, numerous developments in animal science, especially those involving use of by-products, are applicable in the Sahel provided that their costs are lowered and constraints are reduced, such as the constraints on labor availability and lack of credit and extension support (Mc Dowell, 1984; FAO, 1982). Hence, economic considerations appear to be the main determinants of which technological improvements already possible are likely to be adopted and of which types of technical research are going to produce applicable results. 41 Economic Aspects Delgado and Mellor (1984) argue that, on the production side, raising prices only "is not likely to be a sufficient condition for accelerated agricultural growth" (Delgado and Mellor, 1984, p.6). Christensen and Witucki (1982) show also the limitations of a policy solely based on the "right prices“. Among the numerous reasons for that, are the generally low elasticity of supply due to the poor state of the agricultural infrastructure, the inefficient input distribution systems, the difficult production conditions and the low level of technologies. Also, while the elasticity of supply with respect to labor is probably unitary in the current conditions of production due to the low productivity, the labor supply for agri- cultural production is affected by shortages at decisive peak-level periods. On the consumption side, food prices are critical short-run links in a general equilibrium framework since they determine mainly the relative and absolute real incomes of the urban and rural participants in the food system. With the intersectorial linkages permitted by the market system, they contribute to the macroperformance of the food system. Futhermore, they influence the allocation of products and factors according to their elasticities of substitution. Also, Delgado and Mellor (1984) consider that a sustained acceleration of agricultural growth can be 42 promoted by economic policies working after a structural change. Thus, growth depends upon technological innova- tions that cut per unit costs, especially labor costs. That has to be accompanied by good input delivery systems, rural infrastructure, a productive agricultural research and extension system and an increased supply of efficient decision-makers. Another perspective is the Induced Innovation Model defended by Ruttan and Hayami(1984). These authors state that economic growth ultimately depends on the flexibilities and efficiency of society in transforming itself in response to technical and economic oppor- tunities. Without rejecting the importance of an exo- genous supply of technology, Ruttan and Hayami emphasize that appropriate technological change is-endogenous to any real development process. However, the attractiveness for the new technological options and the correlative output generation are possible only if the resource endowment and economic environment are favorable. In particular, economic distortions have to be corrected, and also, the allocation and use of resources and factors made more flexible and efficient. Thus, technology, by playing a deterministic role in the growth process, induces and reinforces the necessary institutional changes. Criticizing the technological determinism of that view, De Janvry (1984) argues that scientific and technological capacity are important and necessary but 43 they should be understood as social products determined by social and institutional variables. Therefore, development of any subsector in a developing economy is influenced by the distribution of economic and political power at the national as well as the international level. The deter— minants of the price milieu and the departures from economic efficiency are not only due to imperfect knowledge and foresight. The distortions among prices and the relative scarcities of factors or products as well as repercussions in the market also reflect political preferences and distributional decisions. They are generated by the institutional arrangement, the different "opportunity sets" constituting it and the structure of power it represents (Schaffer, 1984). A complementarity can be found in the divergent emphases of the Induced Innovation Model and the Radical approaches. These models allow us to find an operational and workable determination of the nature of the relation— ship technological changes/institutional changes or, in other words, development of the productive forces/social relations of production. The Induced Innovation Model and the Radical models point respectively to each side of the relationship and show its role in the explanation of rural stagnation and the definition of a development policy. They demon- strate, at the end, their necessary dialectic relationship instead of a monolithic determinism from either approach. 44 For instance, the Senegalese livestock subsector needs a reorganization based on economic rationalty with respect to the technical aspects, the resource endowment and scarcities but also social and distributional considerations that should be clearly defined by the policy makers. Such institutional considerations are beyond the scope of this study. The focus is rather on the understanding of the subsector dynamics and the research needs that can help in the determination of a successful policy for the subsector. CHAPTER III CONCEPTUAL APPROACHES OF LIVESTOCK PRODUCTION Introduction Understanding of the pastoral economies has long been hindered by the "perception gap" (Baker,1981,p.369) of outsiders. The failure to appreciate the many roles of livestock in the traditional economy and society often led to policies that caused a shattering of the traditional ecological relationship and established no economic and social substitute that could provide a safety valve for the growing pressure on the traditional livestock production system (Baker, 1981). The first attempts to develop frameworks for analyzing livestock systems were made by anthropologists, who limited their analyses to the dominant social and cultural anthropological paradigms (Schneider, 1979; Baker,1981;Dahl, 1981; Grandin, 1980). "Pastorology" (Dahl, 1981) concentrated on the pastoral values and attitudes instead of the economic aspects, according to the structural functional theories in the anthropology of the 1950's (Schneider, 1979). This led to the stereotypes, strengthened by colonial administrators, on "cowdolatry","cattle complex","boumanie" and ritualistic associations between men and beasts. 45 46 By the 1960's, the limitations of the structural analyses in social anthropology were reduced with the Malinowskian influences which attributed all social behaviors to economic causes. However, the resulting economic interpretations were limited by their bad handling of temporal and spatial differences among societies. The ethnocentric biases remained strong (Schneider,1979; Baker, 1981). Furthermore, the economic explanation of overstocking and the general low supply of livestock by the traditional producers led to the pre— valent idea of an economic irrationality and backward- bending supply curves of effort and output in the African pastoral societies. Nowadays, different schools present competing frameworks for the economic interpretation of pastoralist behavior. Their discussion allows us to present, at the end of the chapter, a synthetic interpretation. The 'Strategies for Security' Interpretation Dahl and Hjort (1976), after simulating the effects of drought on livestock systems, show that the main characteristics of the pastoral production system are determined by strategies for security and, in particular, adequate food production. These strategies require, however, an intensive labor use that the system is unable to provide nowadays because of a shortage in supply, especially, at the peak periods of watering. 47 The main strategies cited are - mobility. This procedure is declining due, especially, to labor shortages, loss of land to crop production, ecological accidents and national boundaries; - herd diversification and dispersion. A greater reliance on small ruminants, goats, has been noticed as well as smaller herd units under the same management; and, - maximization of size. That explains the permanent building up of herds and the reluctance for destocking. The policy implication of such a framework is that, by increases in crop production and soil conservation practices, livestock development needs will be served, since grains will be substituted for pastoral products. This will create more intense voluntary destocking. However, Little (1983),.referring to a study in Kenya, shows that opposite results can occur. Increased crop production did not result in decreased livestock holding because it did not undermine the investment aspect of livestock. This was particularly true for wealthy herd owners for whom sales went down and grain production was used for supporting livestock holdings. Criticizing Dahl and Hjort's simulation of required animal numbers for milk subsistence production, Schneider (1979) states that pastoralists are not people 48 chasing protein across the landscape and that the rationale of cattle raising cannot be located solely in the maximization of milk production. Rather, he says, it is better to think of production as focused on utility. Utility is defined to mean a combination of all things that are considered goods, whether they are food, repositories of values, or anything else. The 'Commons' Framework According to the 'commons' framework, the central issues in pastoral production systems are the actual land tenure institutions and the derived individual incentives that shape the resource use decisions (Hopcraft,1981; Widstrand, 1975). The producer behavior is, therefore, not culture bound or determined by a cattle-complex irrationality. It is a predictable response to two sets of incentives: —an external incentive system implied by the trade and economic relations with other producers and other sectors of the national economy; and, -a set of internal management incentives whose interrelationship with the first explains the producer behavior. The 'commons' framework has been described by Hardin and Baden (1977) and Foxall (1979) in terms of the inevitability of the depletion of natural resources with an expanding population and an unrestricted access to 49 their use. In such a context, the positive utility gained by an individual from an incremental use of the resource is almost plus one. Conversely, the negative utility imposed on each other user is a fraction of minus one since the deleterious consequences of overuse are widely shared. In the case of traditional livestock production systems, grazing land is a communally owned resource. However, the private individual or familial ownership of milk, offspring and stock enables the individual or family to derive personal benefits from the land. This leads to a strategy of choice based on herd size maximization, whereby the future of the commons is discounted at a very high rate. This strategy of herd maximization may reduce individual risk in case of ecological disaster and allows individual herder to maintain enough stock to rebuilt their herds. There is an opportunistic use of the land by overgrazing and no incentive to reduce stock numbers. Pastures and feedstuffs are not used in an efficient manner by the growing inventories. In addition, for nutritional reasons, their value declines and the overall increase of output from the sector is constrained. Hopcraft (1981) describes such a situation as a poor way of storing the community wealth because of high maintenance costs and negative interest rates. In the long 50 term, the situation becomes a "tragedy" where: "Freedom in a common brings ruin to all." (Hardin and Baden,1977, p.20) This framework has been the guiding line of many livestock development projects in Sub-Saharan Africa (Lawry, Riddel and Bennet, 1984; Hopcraft, 1981;Crotty, 1980). Its policy implication is the development of private exclusive property rights for an internalization of externalities. This approach has been criticized as inapplicable to many livestock management situations since for the vast majority of cattle producers the circumstances, besides the sociocultural background, require maintenance of some forms of communal tenure. These circumstances are: the average small herd size and institutional constraints for matching them with discrete small pieces of land; and the need for a quick response to highly variable rainfall patterns and low productivity range (Lawry, Riddel and Bennet, 1984). According to Runge (1981), there has been also a misdiagnosis in the modeling of the "Tragedy of the commons" and, furthermore, a limitation in the institutional alternatives proposed. The "Tragedy of the commons" is not a N-person variation of the prisoners' dilemma with a non-cooperative assumption, strict dominance of individual strategies, separability of individual cost functions and the need for a compulsory 51 enforcement by an outside authority for any solution. Rather, reconsidering the short-run profit motive as a causal agent, Runge (1981) argues that the problem of the commons is one of decision-making under uncertainty where strategic interdependencies and non-separable exter- nalities determine the marginal conditions for profit maximization in grazing decisions and herd numbers. For Runge, the institutional answer to overgrazing lies in devising mechanisms that guarantee that choices will be made collectively; these choices will lead to a security of expectation or assurance that the cooperative institutions will act as "endogenous responses to the uncertainty of social and economic interaction" (Runge, 1981,p.601). In a case such as Senegal, there are currently no such institutions even though in the past there had been a traditional regulation of access to pasture, especially for the Sudanic agropastoral economies (Swift, 1979 p.59). Unfortunately, that regulation broke down long ago as a result of population growth,technological changes,cyclical climatic disasters and the power shift during the colonial era (Lawry, Riddel and Bennet, 1984). 52 The Marxist Anthropglogy Analysis Bonte (1977) provides an extensive study of Sahelien herders in an attempt to model their behavior and its evolution. The following viewpoint is an application to the herder situation of the general analysis elaborated by the French School of Marxist Anthropology (Eicher and Baker, 1982). Such a Marxist approach is based on the "theory of value" in relation to the herd and its functions or values. In summary, the value of the herd has three compo- nents that influence the herder's behavior toward the creation of a surplus: -a use value, i.e., a value as a consumption good for immediate use, cash need and food security ; -a value as a means of production, in which future production is expected depending on reproduction parameters, diseases, droughts, etc; and, -a social value, expressed through formation and reproduction of social relations, cooperation and cohesion, reflections of social differences and the distribution of power and influence. These three components are interchangeable directly and immediately because of the distinctive characteristic of animal husbandry in contrast to other techniques of production. There is a perfect technical identity between animal products, means of production and social value repositories. That identity permits, at any moment, one of the components to dominate. 53 However, Bonte (1977) insists that the inter- changeability appears only in the circulation process (ex- change of products) among producers. Also the circulation process involves specific production decisions by the livestock owner. These decisions involve various tradeoffs since each use influences the value of others and reduces their possibility of occurring. From these premises, it is argued that livestock ownership do not only involve joint consumption and production decision making but also simultaneous reproduction of social relations within and between families or domestic groups, ie.,the community. Therefore, the ultimate goal of traditional livestock production is not only the satisfaction of the subsistence needs but, by the production of social value, the formation of social relations. The subsequent strategy for the livestock owner is the maximization of his production capacities, which include his herd size and his family-supplied labor. Such a maximization is subject to environmental contraints, the herder's technical ability and the limits in the development of his domestic group. The competition developing between domestic groups for higher social status creates the need for a surplus and what has been described as the irrational behavior of herders. Bonte (1977) also shows that the same economic and social structures are found from pure nomadic pastoralists 54 to agropastoralists. However, for agropastoralists, since the production process is not only determined by animals, there is a preeminence in the social value of the live- stock. Since land cannot be circulated among producers, livestock becomes fundamental as a means of production and in the formation of social relations. These viewpoints are corroborated by findings from Sullivan (undated) and Grandin (1980). Bonte (1977) concludes that in recent years the herder had less and less control over the outcomes of his strategy and that the pastoral production system has been under an irreversible process of change. These changes have been caused by ecological disasters and diseases worsened by a technological backwardness, on one hand; and, on the other, by the insertion of herders into the world and national markets activated by colonial policies (taxes, sedentarization, territorial limitations). This creates a new environment where the herders are more integrated in the market economy. Incentives from the market and ecological risks lead to competition between the strategies of maximizing herd numbers and accumulating monetary values through animal sales, ie., a maximization of revenues. More and more, the importance of the social value of the animals is decreasing and the increasing dominance of monetary relations inhibits the former uses of the herd. Migrations of herders to urban areas are frequent because they lost 55 or have been disposessed from their means of production. This permits a liberation of productive labor that can be used in capitalist relations of production. Finally, there is a trend of sedentarization for nomadic herders and a growing importance of crop production for agropastoralists since it provides less variance in food and revenues. The critiques that are directed to this analysis refer to its strong viewpoints inspired by Marxist ideology and its radical interpretations of social phenomena. These criticisms correspond to those presented by Eicher and Baker (1982): abstract theorizing, neglect of empirical economic research at the micro level and absence of workable policy guidance. The 'Repository of Value' Framework Schneider (1974), the main proponent of this framework, bases his viewpoints on a cross-cultural analysis of money that includes both integrated market systems in Western economies and non—Western economies. Using the concept of "primitive money", which was first developed and analyzed by Einzig (1949),Schneider (1981) shows how livestock fulfill the three most relevant functions of money: medium of exchange, store of value and unit of account. Also, livestock are a good whose high demand often makes them the medium through which all other goods are traded, as has been the case in many places in 56 the world. From these premises, Schneider (1984) critiques the cattle complex hypothesis in its consideration of cattle as the sole expression of wealth for East African herders but not as money. In effect, besides a competing use for subsistence, cattle have financial functions, are repositories of value and should be taken as money in pastoral societies. The monetary role is also emphasized by Crotty (1980), Shapiro (1979) and Konczacki (1978). From that monetary role of livestock, Schneider theorizes a political economy of pastoralism whereby "pastoral capitalism" (Schneider, 1979,p.221) explains the relations of production and the distribution or acquisi- tion of power. In this framework, pastoralists are constantly involved in investment decisions. They manage a pastoral capital for subsistence and make reinvestment decisions and capital gains from sales (Schneider, 1979). This is, however, done in terms of inferior or "primitive money“. Therefore, a shift to beef production, where cattle becomes a simple commodity, corresponds to a radical revolution in the rationality of the cattle owner for whom the animal is not comparable to the modern and superior forms of money (Schneider, 1981). The main critiques to this approach refer to the overemphasis on the monetary role of cattle and the inadequate analogies made between herders and businessmen or specualtors in a stock market. Also, the business 57 cycle type of evolution related to the consequences of both herd contractions and expansions over time(Schneider, 1979) has been considered inadequate (Livingstone, 1977). Shapiro (1979),déscribing the role of livestock in West African pastoral societies, gives a more balanced approach by referring both to cultural maintenance and social relationships and to production and exchange. In these societies, livestock contribute to subsistence and crop production and provide financial services by acting as investment goods that are easily convertible to cash and portable media of exchange, and that also yield a moderate rate of return. Livestock also offer options for decreasing risk through dispersion of herds or building large herds. The 'Production Rationales' Framework The preceding analyses and frameworks can be updated and their strong points included in a synthetic analysis that enables us to shed light on the herders' environment. Due to the nature and circumstances of traditional livestock production in West Africa, in many cases there is no superior alternative to the traditional man-cattle-land relationship established after a long process of adjustment. Wherever changes are expected despite the climatic constraints, it is important for research program and policy decisions to be based upon a 58 correct perspective of the traditional pastoral system. The 'Production Rationales' framework proposed here is an attempt to capture the strengths and avoid the weaknesses of the frameworks of analyses described above. It provides an understanding of the decision-making environment and subsequent behavior of the livestock owner. The starting point lies on the joint household— herd or household-farm decision-making process, whereby the herders allocate their resources to food and production for cash in the context of social relations that determine the producers'own knowledge, their access to productive inputs, the prices for their output, and their abilty to improve their individual or family welfare. Therefore, in the current production systems, due to severe stresses brought about by ecological problems as well as economic and institutional pressures, production rationales are changing. There is a general reorientation of the herders' economic interest away from strict subsistence production toward engagement in a more integrated economic system dominated by world and national market forces (Lawry, Riddell and Bennett 1984; Behnke, 1983). This creates different tendencies that have been described by the frameworks above. In summary, the two main tendencies are: -The dissolution of the traditional regulatory management of the commons, implying a widespread "tragedy 59 of the commons";and, -the increased use of external markets, with an attempt to maximize private rather than collective benefits since accumulation of social value in livestock has been overthrown by monetary appetite (Marxist anthropology). Changes occurring now in livestock production parallel those changes that were brought about by the introduction of cash crops into subsistence agriculture. However, in pastoral production systems, there has been more flexibility in the adaptation and response to economic change due to the biological characteristics of the production process, the social role of cattle and less control from government marketing networks. Lawry, Riddel and Bennett (1984) refer to a greater decision- making autonomy that is leading herders to choose divergent responses or differential production orientations, management styles and investment practices. This is also emphasized by Little (1983), who adds that social differentiation among pastoralists is a determinant factor in the choice of production rationales Such differential production orientations persist because of the uncertainties linked with the ecological variables and an unpredictable commercial environment. The pastoralists manage to exploit the new market opportunities without, however, committing themselves irrevocably to them. This leads to a hybrid management system (combining 60 traditional and modern practices) that characterizes the livestock production system in Senegal, for instance. According to Behnke (1984), such a situation is a reflection of the transitional phase underway in the process of commercial change in pastoral Africa. His model considers the current phase of development as one of open ranching in the continuous shift from the subsistence system to the capital intensive and fenced ranching system. The dynamic variables in such a shift, however, at different levels in different areas, are: - the transformation of livestock into a commodity and a reduction of its social value due to the changes in the economic and institutional incentives faced by the pastoralists; - the development of a commercial livestock production system at the margin of the market dominated economy; and, - the development of new political issues related to the reallocation and control over water and pasturage resources (Behnke, 1983; Behnke, 1984; Lawry, Riddel and Bennett, 1984). This framework has important implications for policy. In effect, the transformation of the traditional production system and the weakening of its mechanisms of biological equilibrium in favor of a more profit-oriented system implies very serious disequilibria. As it has been true for the Sahel, the impact of these disequilibria on 61 producers' welfare is great, especially when the ecological conditions deteriorate. Therefore, important equity considerations are at stake, in terms of the participation of subsistence— oriented pastoralists in the commercial system. Also, in situation where fencing has been proposed as a solution to the deterioration of the productive resources, such as pasture, the question of who will have access to them and/or control them will have important welfare implications. The analysis of the different conceptual frameworks of analysis leads to the conclusion that in order to face its short-term challenges as well as achieve its long-term economic transformation, the dominant traditional livestock production system needs to undergo important changes. Each of the different frame- works reviewed is based on specific problems leading to related policies. An integrative approach guided by the 'Production Rationales' framework suggests the need for urgent changes in the Senegalese livestock subsector. The institutional setting should not be kept from transforming, but guarantees should exist against the dispossession of herders and further deterioration of the environment. Ideally, we seek cooperative solutions that are self- sustaining mechanisms, that permit the maximization of aggregate livestock production from the land over time and 62 also allow a preservation of the pastures and investments for their improvement. In addition, as in the other agricultural sectors, it is required that the policy makers improve, in real terms, the economic incentives faced by the producers. Finally, technological improvements have to be introduced on large scale with an emphasis on the integration of the husbandry techniques with crop production to allow better joint economic returns. CHAPTER IV ECONOMIC MODELING OF LIVESTOCK OWNERS' BEHAVIOR Livestock Production System Modeling Numerous bioeconomic simulation models of animal production have been developed with emphases either on biological or economic characteristics. Basically, these models are a series of simultaneous equations that trace the flow of energy through the plant-animal system. The equations model the feed production, the intake of feed, the maintenance requirements and the production performance of the animals in the system under study (Levine, 1982). Besides their use in analyzing herd structures and making projections, these models focus on beef grazing systems. They aim at an analysis of the consequences over time of improved pastures, higher technological systems or different management and organizational structures. However, these models are demanding of data and, when doing sensitivity analysis, often require wide variations in the model parameters. They have also a very limited value in the absence of a large data base on the agroclimatic environment (Levine, 1982). Chudleigh and 63 64 Cezar (1982) show also the costly nature of building such models and the need for a breakthrough in how the models are structured in order to achieve greater operationality. Various models of African livestock production systems,are available (Abassa, 1984; Doppler, 1980; Dahl and Hjort, 1976; ILCA, 1982). Generally, these models do not include economic components such as cash flow series, investment criteria, net present values, internal rates of return or economic comparisons of different situations. (Exceptions include Maxwell (1980),Powers (1975) and Ngategize (1982). Furthermore, various kinds of simulation have been done in the modeling of small-farm systems. The models used vary from simple computerized budgets to detailed representations of the complex and interrelated biological, economic and social processes in small-farm production (Hardaker, 1979). In contrast to sophisticated biological models, Levine (1982) shows the operationality of economic models describing livestock production components or overall production systems. In dealing with other types of problems,they are less demanding in biological coefficients. Furthermore,simple economic models allow a conceptuali- zation of the research problems before too much data collection is undertaken. That leads us to an improved allocation of research resources, a move toward the decisive steps of getting research underway and, finally, 65 a perspective on later studies focused more sharply on the crucial data. Decision-Making Analysis of the Livestock Small-Holders Hardaker (1979) shows the role of positive and behavioral studies based on a modeling of the decision- making of small farmers. From these studies, predictions on the responses to specific incentives, new technologies and policy changes can be derived. Mellor (1969) also states that policy oriented production economics in low-income countries is concerned with discovering in sequence: what decisions farmers make, why farmers make the decisions they do, and what must be changed in their environment so that it is profitable for them to make different decisions which will increase production? Problems with the Assumptions on Decision-Making Analyzing the farm management research methods for small farm development in developing countries, Hardaker (1979) notes that in addition to the general lack of relevant and reliable data, there is also the important issue of how farm system performance in subsistence and semi-subsistence agriculture should be measured. General problems related to the assumptions of profit maximization and rationality have been raised by Naylor and Vernon (1979). In addition, Timmer, Falcon and Pearson (1983) argue that pure profit maximization is an 66 extreme case of rational behavior not likely to be found in the complicated world in which farm-households make decisions. In addition, Shapiro (1976) insists on the precautions to be taken when assuming the "Schultzian efficiency hypothesis". Thus, the traditional farm development analysis and a significant part of farm management economics have been treated as misleading when applied to "peasant" or small—holder agriculture. Its many differences with commercial agriculture create the need for new approaches in concept and application (Hardaker, 1979). Such views are similar to those of Eicher and Baker (1982) who stress the great potential for building models that more accurately represent farmers' decision- making by including multiple goals, resource constraints and sequential adjustments to stochastic events, particularly rainfall. Reviewing the political economy dimension of West African farmers' decision—making, Berry (1980) proposes a new "dynamic analysis". That analysis encompasses the static concept of maximizing the expected utility based on profit maximization and subject to risk aversion. Berry proposes that West African farmers follow a strategy of economic diversification based on a portfolio of assets and under the constraint that the different categories of farmers have different access to markets. Other concepts, such as targeting and economic 67 strategizing (Little, 1983) can be added to the many behavioral assumptions or models already presented in Chapter III. Furthermore, greater complexities arise when one recognizes that in West African pastoral systems the differences between ownership and management are very unclear. Numerous rights and social obligations exist among the managers of herds and the owners of animals. The main rights are the rights to sell, dispose or slaughter stock and the obligations to make decisions over management, care and herding. The right to milk or to control the distribution of the milk produced daily is essential in the subsistence role of the herd (Dahl, 1981; Hopcraft, 1981; Behnke,1984). Hence, the difference between the owners and the manager of the herd creates difficulties in the modeling of producers' behavior and supply responses. Moreover, that difference adds other constraints to the development efforts when more investments and new management procedures are required. The Supply Response of Herders The aggregate behavior of herders is at the center of the debate about African producers' supply responsiveness to price structure and, in general, economic incentives. Low (1980) presents a systematic review of the question. 68 Contradictory price responses are not exclusive features of traditional African herders since, even in North America where livestock data are more reliable, conflicting price responses have been obtained. Examples include the cyclical price—quantity behavior in beef and hog markets over the past century,which is still persistent in the U.S.A. Such cycles are described as rooted in rational expectations and adjustments of producers to biological constraints on production. These cycles are coumpounded by the changing age distributions in the herds caused by precedent perturbations in the economic variables (Rucker, Burt and LaFrance, 1984). In such a context, when hog or cattle prices change, the ranchers' decisions are determined by the investment demand for a large size herd, on the one hand, and the consumption demand and concommittant sales for profit, on the other hand. These two types of demand lead over time to different paths of inventory build up and supply of beef or hogs and also opposite responses in the short term and long term (Jarvis, 1977). In relation to the African situation, Low (1980) concludes that: " Although lack of adequate data will often limit the univocity of the results obtained, a clearer understanding of the theoretical and data problems involved with the analysis of traditional herders' supply response to price should contribute to less contradictory interpretations of available data. However, even where unambiguous results can be obtained, price response information alone cannot be used to determine the extent to which African cattle 69 owners' behavior is based on objectives other than income maximization." (Low, 1980, p.2) Besides the supply response studies, the literature presents numerous resource allocation studies specific to livestock production, mixed farming systems and animal traction. Different economic analyses are used to support and interpret the results. Among them, capital theory emerges as a valuable economic framework for under— standing the decision making of herders and livestock small-holders. Capital Theopy Modeling Introduction Capital theory modeling is based on the assumption that cattle, or livestock in general, are "unusual capital assets" (Perrin, 1972,p.60). They are treated as durables, in other words, resources whose services are not all extracted during one time period. Such a treatment can be done since it is the time interval used by the analyst that distinguishes between durables and non-durables in economic analysis. After use and passage of time, durables do not perform the same service as efficiently as they did earlier. Thus, the rate of increase in their value slows, and so the time to disinvest becomes essential in a decision-making process seeking an optimization of the 7O economic lifetime of durables. The problem is, therefore, one of selecting the particular production period which, over a specified planning horizon, yields the maximum net present value of future profits in a situation where resource constraints, such as land and pasture avail- ability, are not yet binding and production conditions not fixed. However, when production alternatives are planned on the same fixed resource, such as a private livestock farm or a cattle feedlot, capital budgeting procedures should be implemented as shown by Harsh, Connor and Schwab (1981). These procedures adjust the accumulated net discounted income during the planning horizon with an amortization factor. Thus, the problem becomes one of selecting the most profitable annuity or standardized income generated from the fixed resource Different adjustments are also made for stochastic inputs and outputs, continuous and discrete time processes, or random elements (Burt, 1965; Chisholm, 1966; Perrin, 1972). In the context of a traditional livestock production system, the key question for the decision maker is when to convert his naturally growing piece of capital into a consumer good, i.e, to slaughter or to sell. Reviewing the main economic models in the analysis of the cattle sector, Porter (1979) shows the complexities arising from this approach and its fundamental differences with the wine production model or the tree-growing model. The main differences are: the demographic aspects of the 71 herd,the biological processes involved,the non-homogeneity of the unit of study due to sex, the continual decisions with respect to physical input streams and the biological requirements, the cash flow complications, the uncertainty considerations, playing a crucial role in the raising of livestock, and the influence of the environment (over- stocking, competition with food growing areas). Jarvis (1977) presents a series of micro-models .that treat cattle as capital goods and producers as portfolio managers for a theoretical framework for an econometric model of the Argentina cattle sector. He states that cattle are considered to be capital goods, i.e., potential future monetary values, which are held by producers as long as their capital value in production exceeds their slaughter value. This happens in the context of a large scale cattle production system that is essentially commercial and that has a correct valuation of capital costs. The review of the competing conceptual approaches of the African livestock production systems in Chapter III has revealed a unanimity in relation to the role of live- stock as a capital good, even though that it is not livestock's only role. This implies the need to look closer at a replacement theory framework that allows us to consider in more detail the multipurpose nature of livestock in the traditional African farming systems and the impacts of age on costs, quality of meat, different 72 uses and revenues. Jarvis' microeconomic approach has been also used in the analysis of the status of cattle and the 'commons' problems in Swaziland. In response to Doran, Low and Kemp (1979), Jarvis (1980) uses his model to show how , under a communal range system, herders find it profitable to hold onto animals, therefore creating a negative Short-run price response to cattle slaughter. Crotty (1980), Ariza-Nino and Shapiro (1984) also use capital theory frameworks in their different economic models of decision-making by cattle keepers. Theoretical Bases Livestock are a form of capital asset whose physiological ability allows their manager or owner to consider two strategies: - the accumulation of capital goods creating, through physiological reproduction, other capital goods that are themselves sources of value; and, - the management of an asset that is able to convert feed into meat, milk, draft power, manure, hides, prestige and religious or social values. These functions also require costs in the form of maintenance, mortalities and foregone interest or return on alternative investment opportunities. The net expected present value of the services provided by the livestock asset is a function of the time the asset is held and 73 corresponds to the difference between returns and costs, all discounted and adjusted for death loss. At the beginning of its life, the net expected present value of an animal's services is negative since only costs are incurred . Then, it increases over time until reaching its maximum, which corresponds to the optimal age for sale. After that moment, the net value decreases until the death of the animal. It is assumed that the asset owner compares the opportunity gains from keeping the current asset for another time interval, such as a year, with the opportunity gains which could be realized from its liqui- dation during the same period (Perrin, 1972). The relative value of the future stream of gains is expressed in terms of present earnings by the discounting procedure. The discounting procedure allows the owner to compute the value at birth of the livestock asset as the present discounted net value of the services provided by the asset during the production process associated with it (the livestock asset). This is so assuming perfect knowledge and no risk. The discount rate is the conversion factor of future streams of value to present values. It can be determined by either the cost of capital, the return on the best alternative investment or the preferences in the timing of personal consumption and, finally, the risk involved in the economic enterprise undertaken. 74 As it relates to an economic evaluation based on the marginal principle, such a model suffers from the general criticisms that are raised with respect to modeling decision—making in traditional production systems. Furthermore, the assumption of homogeneity in animals and farms is also a problem in an economic setting exhibiting an important dispersion of types. Also, as pointed out by Jarvis (1982a), the joint product problem faced in the modeling becomes (analyzing the contribution of different products and services produced by the herd to the value of the stock) too complex when a high number of products is involved. In such a case, Jarvis recommends a dissaggregation of the production process and activity analyses to obtain better solutions. These major limitations restrict the model accuracy but the model remains a useful tool for conceptualizing hypotheses concerning herders' decisions, for sketching out issues and for deriving second-best solutions. In addition, these investigations can be done cheaply while still providing indications about what further data need to be collected to test specific hypotheses. Finally, with its examination of the livestock asset profitability, The capital theory approach makes it possible to study the relationship between the structure and sex composition of the current herds and their current uses . Thus, it contributes to our understanding of the livestock subsector dynamics, the economic bases in the 75 pattern of livestock holding and management and the means for their improvement. Mathematical Model In this section, we will focus only on cattle production systems, assuming that simple adjutments can be made for other livestock, such as small ruminants. The mathematical model presented is a partial equilibrium model of a pastoral production system or the cattle component of a mixed farming system. Although it results from a combination of different studies, it is inspired mainly by Jarvis' approach (Ariza-Nino and Shapiro, 1984; Jarvis, 1977; Jarvis, 1980; Jarvis, 1982; Panayotou and Tokrisna, 1982; Porter, 1979). The general mathematical model is as follows: p(i.a).w(i.a) e'ra Price and growth component (4.1) Pr(a) e—rt dt I 0 H é\\\ N Cost component Cv(i,t)/(1+r)t 1 Calf stream component + n (IMO) ad(i.t) e‘rt dt Draft power stream +pO \ a 1 a mk(i,t) e‘rt dt 32 Milk stream a + pma ma(i,t) e'rt at O Manure stream + pmk + Social Value 76 Where: Pr(a) 8 Present discounted profit from an animal at age a; p a Price of meat, which is a function of i and a; i = Bundle of daily inputs; a = Age of the animal; w a Weight, which is a function of i and a; r = Discount rate; t = Time; C1 = Cost of input 1 (feed,shelter,fences,machinery, veterinary care,herding); Cv = Expected value of calf born in year t, assuming the cow has received input stream i; po = Unit value of d; d = Amount of draft power that is a function of i and t; a1 = Age at which draft services begin; pmk = Price of milk or per unit value; a2 = Age at first lactation; mk = Amount of milk used for sale or consumption and produced by a cow aged t and receiving input 1; pma = Price of manure or per unit value; and ma = Manure produced by an animal aged t and receiving the input stream 1. 77 Taking ther first order conditions of equation 4.1 provides the optimal age of sale and input stream for an animal, given known growth curves. Complexity can be added by considering price expectations, by letting p = p(i,a,t), whereby the price of meat is not only a function of the inputs and the age or quality of the animal but also seasonal time trends. Also, more interestingly, such factors as climatic variations and the impact of diseases can be included in the growth function with w = w(i,a,b), where the weight is a function of the inputs, the age of the animal and a coefficient measuring biological efficiency. By adjusting weights for the fluctuations occurring seasonally, this coefficient enables us to obtain a curve fitting more precisely the real world age-weight relationship and the production conditions. By satisfying the first order conditions for an optimization of Pr(a) or Pr(i), the optimal economic lifetime of the "durables" and the optimal input stream can be analysed and the replacement or liquidation rule determined on the basis of net present values. Further- more, inferences can be made about the directions of change in the optimal age of sale and the herd structure that will result from changes in policy and technical variables. The solutions to such a model are theoretically accessible but have many mathematical complications along 78 the way. As stated by Naylor and Vernon (1979), it is possible to conceptually formulate a mathematical model describing the behavior of a dynamic, multiprocess firm operating under uncertainty, but present-day mathematical techniques are simply incapable of yielding solutions to a problem of this magnitude in an efficient manner. The time involved in solving such a problem through calculus and the various sources of computational errors make them inappropriate for practical use. Decision-makers are interested in faster solutions to more complex problems. Therefore, for a numerical analysis based on decision rules, a simulation by computer is of great help. After this rather abstract but necessary introduction on modeling using capital theory,the following chapter presents a more operational approach with a practical application to a specific livestock production system, the Ndama cattle production system in the Upper Casamance. CHAPTER V PRESENTATION OF THE MODEL AND THE SIMULATION METHODOLOGY Characteristics of the Model A basically classical nomenclature and approach are used in this study for the model and simulation methodology (Dent and Blackie, 1979). The simulation technique is taken as a numerical representation of the conceptual model of the herders' behavior over time as embodied in the capital theory approach. The model is more mechanistic than predictive. Its main goal is to assist in the understanding of the problem under study rather than making projections on the behavior of the system. As with any model, it has limitations in representing the complex interrelationships and dynamics of the phenomena it seeks to represent. For the purpose of our study, it was not possible to collect detailed primary data to use in the model. Thus, the available information on the Ndama production system, reviewed in Chapter II, is used for the qualitative and quantitative description of the model. Also, a stochastic specification could have been useful since it takes into account the uncertainty implied by 79 80 unexplained events, truly random elements and the environment's unknown influence. Stochasticity can be introduced by generating random variables in the modeling exercise. However, as shown by Dent and Blackie (1979), stochastic specification is important in bioeconomic simulation but it should be introduced with caution due to the complexities of the processes being modeled and the lack of knowledge about the probability distributions of data collected. Because of the lack of information on probability distributions of the major variables in the model, it is not possible to incorporate stochastic features in this study. The model used is deterministic. It is assumed that the most likely situations are the ones where all random factors take on their average values. This applies to prices, weather, human behavior and physiological processes such as growth, disease frequencies and intensities. In addition, one of the major problems in simulation is the inclusion of time since computers are sequential processors. In this work, a time-stepping procedure is used. The dynamic characteristic of the real world situation is introduced by a time-varying discrete procedure used to approximate the movements of factors that vary continuously. Time is advanced by single fixed units of one year. Finally, the model does not include explicitly 81 resource constraints. It is assumed that the system is not yet bound by constraints on the main resources, such as pastures and managerial capacity. However the levels of calving and mortality rates create limits against the infinite expansion of the animal population. Description of the Model Optimal Age of Sale Component Egpption The profit function used is (5-1) Pr(a) = p(a).vmv(at)e"ra ://3t(p.t)e‘rtdt - E//3e’rtdt O O In this equation, Pr(a) is the net present worth of the stream of benefits from the animal from birth to age a, that is, it is the discounted total profit a herder would receive from an animal that was sold at age a. Pr(a) is the amount that, if invested at the rate r, would have the same value at time t = a as the animal sold on the market at time t. P(a)-W(a)e-ra expresses the market sale value of an animal whose weight and price are both functions of age a. a ’//’f(P.t)e-rtdt expresses the flow of benefits 0 other than meat production from the animal from birth until age a. These benefits are functions of their prices 82 (or equivalent unit values) and time that is the age of the animal at the moment of their production. séfhe-rtdt is the flow of costs over time.They are taken as fixed here, i. e., costs are assumed to be the same for both sexes and for all age categories. Only sufficient data would allow us to consider cost as a function of age and sex, especially for the cows, because of the nutrition and veterinary requirements related to their reproductive function. So, cost would be expressed as ‘4{3f(c,t)dt, that is, a function of their unit value and the time where they are paid. The sensitivity analysis in Chapter 6 examines the case of reproductive cows presenting a changing cost structure over time. e-rt allows the discounting procedure, as shown in Appendix B. r is the discount factor used. It permits us to consider the one-period losses caused by not selling the animal and investing its value in another better alternative opportunity. Decision Rule In the conventional marginal analysis, the decision rule is Do not sell if Expected gains in one extra year > Expected costs for the corresponding year. 83 Algebraically, the decision-maker is faced with the comparison of: (5.2) wdp/da + pdw/da + f 1% rpw + c This relation means that the change at each age in sale value due to quality and weight plus the flow of benefits from holding the animal one extra year, are compared with the sum of the interest foregone by not investing value of the animal (pw) at the interest rate r and the costs incurred by keeping the animal during this same period. This relation is valid only if there is a declining rate of gain in the animal growth over time and also a declining level in the price per kg of the animal as it ages (second order conditions). In our study, the decision rule that has been described is translated to more simple terms with the comparisons of net expected present values. The decision maker reaches the goal of profit maximization with the sale of animals whose net expected present values are at their maximum. This approach is based on the assumption of an optimizing pastoralist. It also refers to animals presenting average performances in terms of meat, milk and calf production. The variations due to performance differentials are examined in the sensitivity analysis in 84 Chapter 6. With this model, the optimal age to sell for the producer can be derived as well as the differential impact of parameter changes upon the capital value of animals of different ages and sexes. In this study, a generalized approach is used. There is just a consideration of males and females instead of the different categories of cattle. A more subtle differentiation could, however, be done if data were available. Finally, optimal age of sale are used as the key variables in the herd structure component of the model which derives the herd size, age/sex composition, offtake and herd productivity. The main objective of the model is to suggest the most likely effects of changes in cattle management and the nature of the relationships among technical and economic parameters in the context of southern Senegal. Firsthand data are not available and also could not be collected. Therefore, all the data inputs are based on secondary information , experience and assumptions on the semi-extensive and traditional Ndama cattle production system that was described in Chapter II. 85 Liveweights Due to the unavailability of weight—age relation- ship data from birth to adult age, a growth model is used in this study to generate age-specific weights for both sexes. Various growth models are available in the literature (Brody,1964; Brown, Fitzhugh and Cartwright, 1976; Aguilar, Friedli and Canas,1983; Landais,1983; Abassa,1984). Despite the variation from breed to breed, the overall shape of the weight curve is sigmoidal and related to the maturation patterns under given environmental conditions. Thus, it is assumed that, within each breed, the variations due to seasonal short-term fluctuations in body composition average out when a year-stepping model is used. In this work, the intent is not to test the growth models, although a realistic and accurate one is needed. The Gompertz curve has been chosen. Its limitations stem from the assumptions of a fixed inflection point and the tendency to overestimate early weights (Brown, Fitzhugh and Cartwright, 1976). In addition, it does not provide us with a good fitness of the curve at very late age for the female. In effect, the assymptotic adult weight used in the computation does not represent well the decreasing trend in cow weights after 10 years. Nevertheless, the Gompertz curve provides a good fit at the ages of impor- 86 tance in the valuation of the animals and requires only three parameters for its computation. A slight over- estimation of female weights around 15 years has only a minor influence on the total expected present values of the cows. The bias would have been more serious in the case of the males, for whom weight plays an essential role in the determination of the optimal age of sale. After manipulation, the general form of the Gompertz curve is transformed to (see footnote at the end of the chapter): Wt = Woexp.b(1 - rt) Wt is the liveweight at age t W0 is the birth weight b and r are coefficients derived from the mathematics. The initial data regarding weights and the age at the inflection point have been chosen as the average values for these parameters in the Upper Casamance (Table 8). The resulting equations are For the male. Wt = 18 exp.2.966(1 — .580t) For the female. Wt = 17 exp.2.605(l - .szat) Figure 2 presents the two curves generated with the Gompertz methodology. 87 WEIGHT CURVES OF NDAMA (Male and Female) LEGEND 34534734834! 349349 ___Male 350+ 33: 342 Female “4- 23m: 0 x '5 m 4:0 m 'n 0" .c ,: .9 I " 2 N7; 150" / 12: I/ INN ;' 50? y I p / a 1 ‘ :w L I L a : + Birth12 3 4 5 6 7 8 9101112131415 Age in year Nfizweights from Gompertz formula; Birth weightsmaleurIBfiemaleafl Figure 2. Weight Curves of NDAMA 88 Table 8. GROWTH DATA FOR THE GOMPERTZ CURVES (Ndama cattle; male and female; Senegal) Sex Data Male Female Birth weight(KG)§-/ 1a 17 Adult weight(KG)3/ 350 230 Age at inflection pointQ/ (year) 2 1.5 SOURCE :gfiLCA, 1979 -age at puberty (assumption) Mortality Rates ILCA (1979) refers to an overall mortality rate of 12% in the region. Shapiro (1984) are used in this study. mortality occurring in a breeding herd (Therefore, The data provided by Ariza-Nino and They describe the slaugh— tered animals are not acounted in the number of deaths occurring in each age class and sex category in the calculation of these mortality rates). The pattern of age— specific values reflects very high mortality rates at early ages followed by a stabilization at adult ages (Table 9 and Figure 3). 7. Rate in 89 MORTALITY RATES (Age specific — Males and Females) ‘° LEGEND __ Mortality rates (BA) . Mortality rates(SA) 207 151’ i ....... L l 1 Age in year BAzrates in Base AnalysisfiAxates in Sensitivity Analysis Figure 3. Mortality Rates 90 Calving Rates The available data refer only to aggregate birth rates, which are given as 56% (ILCA, 1979) and 46% (Starkey, 1984). In Upper Casamance, estimates ranging from 50 to 55% are reasonable (ILCA,1981). Considerations of the productive life of cows permit us to draw age- specific calving rates. The data used are based on a conservative overall calving rate (Table 10 and Figure 4). TABLE 9. MORTALITY RATES (Ndama cattle; male and female; Senegal) Death rate (%) Age (Year) (1) (2) 0 - 1 .35 .30 1 - 2 .10 .15 2 — 3 .05 .05 3 - 4 .03 .02 4 — 5 .03 — 5 - 6 .03 - 6 - 7 .05 - 7 - 8 .07 - 9 - 1O .09 - 10 - 11 .13 - 11 - 12 .15 - 12 - 13 .17 — 13 - 14 .19 — 14 - 15 .21 - Overall rate .12 .12 SOURCE : (1) Ariza-Nino and Shapiro, 1984 (2) ILCA, 1979 Rates In 2 91 CALVING RATES (Age specific) LEGEND w. : .................. ‘. __ CaIving rates (BA) Calving rates (SA) 7m- RU 10‘ 4 5 6 7 8 9101112131415 Age in year Bkrctee in Base MalysimSkrotes In Sensitivity Analysis Figure 4. Calving Rates 92 Table 10. CALVING RATES (Ndama ; Senegal) Calving rates (%) Age (Year) (1) (2) O — 1 0 0 1 - 2 0 O 2 - 3 O O 3 - 4 .30 .30 4 - 5 .60 .60 5 - 6 .70 .70 6 - 7 .70 .70 7 - 8 .70 .70 8 - 9 .70 .60 9 - 10 .65 .60 10 — 11 .60 .50 11 - 12 .55 .40 12 - 13 .50 .30 13 - 14 .45 .30 14 — 15 .40 .30 Overall rate .56 .50 SOURCE : (1) Ariza-Nino and Shapiro (1984) (2) Data used with a conservative assumption Milk Yield and Value Various lactation results recorded on stations with Ndama cows are available in the literature (ILCA, 1979; Landais, 1984). However, few investigations have been done on farms. Starkey (1984) refers to a full lactation averaging 500 to 600 kg. The lactation yield of 500 kg can be used as a conservative assumption which 93 takes into account the general decreasing productivity in Upper Casamance in the last decade. These data do not consider the lactation fluctuations over the cow's life. Different quantities are produced at each lactation according to specific trends. Such trends are a function of not only the environment (water, health), but also the number of past lactations. These variations definitively have a strong impact on the economic returns from each cow. Research is still very much needed to provide data on traditional dairy production. Meat Value Producer prices vary considerably depending on the area of production, the place of sale and the age of the animal (SOMIVAC, 1982). It is, therefore, inaccurate to consider a single price based only on weight in the estimation of per kg liveweight prices (Shapiro, 1979). Prices can also vary according to the quality of the animal. In Upper Casamance, variations from 120 to 160 FCFA per kg liveweight are usual. The price structure used in this study is based on the information available from the public agencies (SODESP, 1984; SOMIVAC, 1982). Average prices that have been collected from 1978 to 1984 are presented in Table 11 and Figure 5. The derived market values of the animals for their meat values are 94 PRICES OE LIVEWEIGHT KILOGRAIVIS (Males and Females) 1 Price In FCFA I? INT SH“ I LEGEND : __ Male (BA) 1 .. Male (SA) I Female (BA) \ . _ Female (SA) i I I I a 1 n l L 1 L l L J I' T' I' r I r T r r 1' I' r r ‘l fi— fi 12 3 4 5 6 7 8 9101112131415 Age in year BA : Price: in Base Analysis; SA : Prices in Sensitivity Analysis Figure 5. Prices of Liveweight Kilograms 95 correlated to their growth pattern as shown in Figure 6. Table 11. PRICES ( Liveweight; Casamance; Averages from 1978-84) Age (year) Prices (FCFA/KG) 0 - 1 150 1 - 2 150 2 - 3 150 3 - 4 150 4 - 5 150 5 - 6 150 6 - 7 150 7 - 8 150 9 - 10 140 10 - 11 140 11 - 12 130 12 - 13 120 13 - 14 120 14 - 15 120 SOURCE : SODESP, 1984; SOMIVAC, 1982 Costs of Production The opportunity costs of land, pastures and agricultural residues are assumed to be nil. In reality, the land may have some income-earning potential without livestock production, and incorporating that opportunity costs into the model would reduce the net benefits from livestock production somewhat. Furthermore,the pastoralist uses them as communal property without paying any direct costs. Labor is used for supervision, watering and 96 MARKET VALUES (Meat values) ...... LEGEND __ Male MT ‘ . Female O. .... .0 OOOOOOOOO oooooooo Value In FCFA 10M" A l 1 L L L I l L l L 1 L l _ j. 123456789101112131415 Age in year Figure 6.. Market Values 97 milking. The labor need is very important and is solved by relying on members of the family (children or un- employed persons). The valuation of labor is a major area of research and there is no objective estimate available for crop production. In the case of cattle rearing in the area of Upper Casamance, where the stocking rate is not critical and the competition with crop production low, it is reasonable to consider that the labor costs are very low in terms of financial analysis. Thus, for the base run no value is given to the labor cost. There is a minimum cost of 375 FCFA for vaccinations and eventual veterinary care (SODESP, 1984). Conceptually, death losses are an element of cost and the cost inputs for the modeling should explicitly incorporate them as well as their effects on past income (Shapiro, 1979; Panayotou and Tokrisna, 1982, Ariza-Nino and Shapiro, 1984). In this study, death losses are included in the computation indirectly by using the probabilities of survival for each age and sex category. The survival rates are used to adjust expected present values for death losses. Therefore, the cost figures of the input data do not contain the death component, but all expected present values have been obtained with the consideration of death losses and other monetary costs. 98 Interest Rate The Senegalese farmers had, until the §g§_(Chapter II) access to substantial amounts of credit for crop production through the Programme agricole and various credit schemes offered by the development agencies. The credit programs were extensively subsidized and created significant inflationary pressures. The efficiency of the formal credit system was also plagued by the high fungibility of the loans and bad repayments from the lenders. As shown in Chapter II, the livestock subsector was not directly affected by the formal credit institutions. Therefore, estimates of the real interest rates are based on data from the informal credit market. In a study of the formal and informal markets in Senegal, Tuck (1983) shows that 67% of the households surveyed borrowed at least once during a year. That figure expresses the intense activity of the informal market, which has been described as a primary source of financial services for farmers. Loans are usually short-term. They are not for agricultural production but rather for consumption and recurrent annual expenditures. Tuck concludes that these loans are use to solve resource budgeting problems caused by a lack of savings or a mismanagement of cash flows. Even though Moslem tradition forbids charging interest, very high average annual interest rates are used 99 (Table 12). For our modeling, the forgone interest on the current value of the animal is estimated at 20%. Such a high value is in the range of the simple interest rate found by Tuck (1983) in Casamance and is also confirmed by Shapiro (1979), who estimates the opportunity cost of capital between 20% and 25% in West Africa. It is considered a reasonable estimate reflecting, in particular, the time preference for money, the risk component for the lender and his/her costs for providing such services. Table 12. AVERAGE ANNUAL INTEREST RATES (Informal markets; 1982-83; Senegal) Regions ‘ Rates(%) Senegal Casamance Simple interest ratea 34 16 Average interest rateb 79 4O /3 including loans without interest /b 44% of loans being interest bearing SOURCE: Tuck (1983) 100 Herd Structure Component The starting assumption in the herd structure component of the model is that any change in the use of an animal of a given age and sex has repercussions on the use of the other animals and, thereby, on the age and sex structures of the herd (Jarvis, 1981). Studying the herd structure, on the one hand, and its transformations under various parameter changes, on the other hand, help us understand producer destocking decisions. Potentially, the model could be elaborated to study the sale response differentials for various categories of animals (feeder calves, fattened steers, etc.). This would allow us to derive a ranking of the sale elasticities of the different categories. In this study, the herd structure component is a simple demographic model. The aim is to evaluate the trends in offtake, growth and productivity correlated to the management decision taken. The model is ,therefore, limited to a simulation of the demographic processes generated under given initial technical and economic conditions and their evolution under the influence of alternative management decisions. The initial herd structure is generated by applying a cumulative survival rate of the different age classes on 1000 animals (males and females). The ideal initial starting population is in the range of 70 to 90 animals, which is that of the average herd size in the 101 area of study (ILCA,1979). However, to reduce the effects of rounding errors, the simulation was done on 1000 animals. The aggregation of 10 to 12 herds should not create serious aggregation biases. Once the initial herd structure has been generated, then in each succeeding year, calves are added according to a sex ratio of .5, the age-specific calving rate and the number of productive cows. After that, each year, the surviving animals in each age class are transfered to the following age class animals reaching the optimal age of offtake are removed from the herd. Finally, the growth rate and productivity of the herd are calculated for each year as well as the average herd growth rate for every 4 years. These production parameters allow an overall appreciation of the herd evolution over the simulated period of 25 years. All the data on technical and economic parameters are the same as that for the age of sale component. The optimum age of sale is incorporated as an input in the determination of the offtake in the herd evolution. Block Diagrams The block diagrams presented are sets of boxes indicating the major components of the model and linking arrows indicating the interrelations and flows of material or information. Figure 7 describes the cattle production component, the overall mixed production system and the 102 various measures of performance that are used for analyses at the microeconomic and macroeconomic levels. Figure 8 presents the determinants of the herd demography. Figures 9 and 10 show the dynamics of the two components of the model, following the nomenclature chosen by Dent and Blackie (1979).They specify the interrelation- ships among the exogenous variables , the rates of changes and the state variables. Computer program The computer program is written in FORTRAN. It constitutes an approximation of the mathematical models used to represent the real world with restrictions related to the inclusion of time and stochasticity and the data availability. The program with comments and variable directory is presented in Appendix C. 103 Marketing_ Price policies PRODUCTION TRADITIONAL Animal imports Income and population growth Animal output Milk output Other services Animal output CONSUMPTION Income (meat,milk,others) Expenditures TDN TRADITIONAL GRAZING TDN Crop residues Land CROP PRODUCTION Land allocated MODERN to cash,food PRODUCTION and animal feed crops Population and income growth ACCOUNTING MECHANISMS Modernization costs MEASURES OF SYSTEM PERFORMANCE Discount rate Discounted return Foreign exchange Farm income Nutrient output Income from beef and milk Figure 7. Functional flow diagram of cattle production (Adapted from Manetsch and al., 1971) 104 Herd management Price Deaths policies of females Females sold Births of females Females transferred NUMBER TDN OF FEMALES X females lactating TDN/animal TOTAL POPULATION yield per animal Birth of uales other service output ‘ NUMBER OF MALES Deaths Males Males of males transferred sold Herd management Price policies Figure 8. Causal map of cattle production (Adapted from Manetsch and al.,1971) 105 .Life expectancy Death rate Age - Liveweight 3+___Meat value ' TCEEEEE> . Progeny value k icalving rate ) (Milk value ' (Quantity ) nit value Quantity .Other services VALUE 9‘.__ Costs a__. Discount rate NET PRESENT VALUE Alternative opportunites AGE OF SALE m Exogenous O variable State variable E3 Variation _ 8 Figure 9. Diagram of age of sale dynamics 106 [:1000 HEADS 3 Survival rate ge specific eath rat Sex specific Death rate MALES FEMALES Age specific Age specific Death rate Death rate Offtake r‘g #— Age specific Calving rate 2H Offtake MALES I FEMALES ___4 HERD YEAR t+1 t MAX NO SINK Exogenous variable (:) State variable ::3 Variation -——8 Figure 10. Diagram of herd dynamics 107 Footnote to Chapter V Gompertz curve computation p. 86 Wt = W0 exp.[b(1 - exp.-mt)] wt w wo A exp. IP 3 ll weight at age t birth weight adult weight or asymptotic mean weight a base of natural logarithm age at the inflection of the curve ln(b).m'1 (when dW /dt is maximum) 1n (W / w t A a) rate of maturation; refers to growth rate relative to mature weight and is the rate of decline in the growth rate. exp.-m = r W0 exp.b(1 - rt) CHAPTER VI RESULTS OF THE SIMULATION AND SENSITIVITY ANALYSIS Optimal Age of Sale and Herd Structure Base Analysis Results Tables 12 and 13 present the year step results for both males and females based on the data in Chapter V. Figure 11 shows the graphical evolution of the net expected present values (EPV). The maximum EPV determines the optimum age of sale for a rational decision maker at 4 years for a male 12 years for a female The final herd equilibrium structure is after 25 years of simulation on the hypothetical population is the following: Males : 34% ; Females : 66% Male calves : 10% ; Female calves : 10% Young Males (1 to 4 years) : 24% Young females (1 to 4 years): 24% Adult females (up to 4 years): 32% These results are relative percentages from the numbers generated and presented in the computer output in Appendix C. 108 109 NET EXPECTED PRESENT VALUES .... j LEGEND l ............. E ... Male Net EPV M? Female Net EPV ..-: -0“- ‘A f -—--._-.—vm— ~m-~‘-.— .0. Net present value In FCFA «all a: e+keee e++ea¥=1~| 12 3 4 5 6 7 8 9101112131415 Age in year flax Net EPV at 4 year: for male: and 12 year: for female: Figure 11. Net Expected Present Values 110 Table 13 . WEIGHTS AND EPV OF A MALE (Base analysis results from 1 to 15 years) Age Weighta Meat Value Survival Lifeb NET EPVc (year) (KG) (current) rate(%) expectancy(%) (FCFA) 1 63 9384 .65 .65 4994 2 129 19327 .90 .59 7432 3 196 29387 .95 .56 8702 4 250 37473 .97 .54 8725 5 288 43146 .97 .52 7875 6 312 46822 .97 .51 6671 7 327 49096 .95 .48 5308 8 336 50465 .93 .45 4006 9 342 47858 .91 .41 2641 10 345 48303 .89 .37 1769 11 347 48563 .87 .32 1082 12 348 45235 .85 .27 475 13 349 41831 .83 .22 60 14 349 41875 .81 .18 -177 15 349 41900 .79 .14 -342 :. weights generated with a Gomperz curve . cumulative product of survival rates used to adjust costs and returns for mortality Net EPV = PV at birth when animal sold at age i minus costs adjusted for time and mortality NB: - Discount factor is 20% - Current costs = 375 FCFA after year 1 - Epv = Expected returns minus costs — Expected returns = current meat value'life expectancy *discount factor — Costs = Current costs*life expectancy*discount factor 111 WEIGHTS AND EPV OF A FEMALE (Base analysis results from 1 to 15 years) TABLE 14. Year Weight EPV Components (Age) (KG) Meata Milkb Malec Femaled NETe progeny progeny EPV 1 58 4641 O O 0 4641 2 111 6545 0 0 0 6398 3 157 7172 0 O 0 6911 4 188 6826 2271 317 342 9403 5 207 5965 5878 820 1270 13509 6 217 4982 9220 1287 2586 17604 7 223 3980 11819 1650 3945 20868 8 226 3073 13798 1926 5176 23404 9 228 2153 15062 2103 5947 24679 10 229 1575 15983 2231 6591 25777 11 230 1124 16529 2307 6956 26299 12 230 727 16834 2350 7109 26394 13 230 456 16989 2372 7161 26346 14 230 303 17092 2386 7205 26350 15 230 196 17158 2385 7242 26342 a .PV in FCFA of meat when cow slaughtered at age 1 .PV of milk produced = Quantity*va1ue*discount factor *probability of lactation Probability of lactation = calving rate at age 1 *.5 0.5 represents the probability for a female of age 1 to have a normal lactation considering the high mortality rates from 0 to 2 years and the various post-partum problems that can occur - PV of male progeny = sum of the maximum EPV of a male'cumulative probability of a male birth ‘discount factor to discount back to birth of the cow. - Cumulative probability of a male birth = female life expectancy*calving rate at age 1 *sex ratio of .5 d. - PV of female progeny = sum of female calves present values over the reproductive life of the cow weighted by their corresponding mortality, calving rates, prices and discount factors. - Sum of female calves PV = Total expected present value of a cow at birth - PV of meat, milk and male progeny. e. - Net EPV 3 PV at birth - Costs discounted and adjusted for mortality. 112 Validation According to Dent and Blackie (1979), the adequacy of a model should be seen in relation to its purpose and not from an absolute reference to the controversial notion of truth. Thus, the validation of a model should be based on an assessment of its prescribed use and a comparison of its performance either against recorded data with statistical testing or a subjective determination of what the output should be, given a broad understanding of the type of system represented by the model. As shown in Chapter V, this work is entirely based on secondary information. Therefore, the validation is based solely on subjective assessments grounded in a critical appraisal of the secondary information reviewed. It is assumed that such information provides us with average values and realistic trends. Among the data available, as much as possible has been drawn from the on-farm data collected by ISRA in a study of about 1000 head of cattle, grouped in 13 herds in the area of the Kolda research station. The conditions of the data collection and the lack of systematic statistical analysis require from us cautiousness in their use (Gueye, Ly and Diallo, 1981). Nevertheless, the data are considered to be valid in terms of a general description of the structure and composition of the herds. The validation of the model is presented in two parts each related to one of the model's components. 113 Age of Sale Component of the Model For the validation of the age of sale component, little data are available on the specific ages of cattle at their time of sale in the traditional livestock production systems in West Africa. In the literature reviewed, Shapiro (1979) and Ariza-Nino and Shapiro (1984) give a range of values that are the same than those produced with our model. In its 1978-1980 study of 500 cases at the Kolda slaughterhouse, ISRA (1981) showed that 76% of the females slaughtered were above 8 years. Among them, 22% were from 8 to 10 years old, 18% from 10 to 12 years old and 36% above 12 years. The comparable data for males are 80% of males above 4 years. Among them, 50% were above 6 years. 60% were also castrates. The data for the females are representative of management decisions leading to a high age of sale whose range is similar to the output of our model. In the case of the males, the data are quite different, with a higher age than that obtained with our model. The position of Kolda in the marketing channel explains such results for the males. In effect, Kolda is not a terminal market for live animals. Therefore, young animals coming from the local herds are directed instead toward Ziguinchor, Tambacounda and even Dakar (SOMIVAC, 1982). In addition, there is an increasing use of animal 114 traction in Upper and Middle Casamance (Landais, 1984) and consequently, a relatively important supply of castrates and old males replaced after service. Thus, the ages of males slaughtered at Kolda are not representative of the age when cattle owners in the area destock usually. More accuracy could be reached with age data from the collection markets or from the arrivals at the markets supplied by the region of Kolda. Finally, the validation of the age of sale component of the model is directly related with that of the herd structure component. Relatively more accurate data are available on the herd structure. Thus, since the age of sale as a management decision has direct effect on the herd structure, the validation of the herd structure component contributes to the validation of the age of sale component. Herd Structure Component of the Model The time path of the animal's use and the transaction strategy of the livestock owner determine the time path of the total herd structure and composition, but the reproduction and mortality parameters are also decisive variables. Therefore, the validation of the herd structure component verifies the reliability of the age output but also allows to check the validity of the assumptions made to describe the production system and some biological parameters, in particular, the age and sex 115 specific parameters. Tables 15,16 and 17 present the relative values derived from the output, on the one hand, and the secondary information available on the herds in the area under investigation on the other hand. The sex composition and age distribution computed from the model output are very close to the comparative data present in the literature. However, a significant discrepancy appears for the production parameters. The results of the model are lower. Further research is needed in this domain since we did not find in the literature current levels for those parameters. In fact, the studies found refer to predrought 1970 conditions (SEDES, 1975) and "normal times"(Shapiro, 1979). The description of the Senegalese livestock subsector in Chapter II shows, however, that the trends have changed since 1970 with more perturbations and a generally lower productivity. Thus, it is very probable that the offtake suffers from conservative attitudes of the herders. Furthermore, the growth of the herds remains low as the rebuilding processes have been constrained by the successive contractions in numbers in a decade of unfavorable ecological and climatic conditions. Besides the comparisons of the model output with data from the real world, the model is also judged in the following sensitivity analysis. 116 Table 15. HERD STRUCTURE SEX COMPOSITION (Model output in year 25 and comparative data) Sex Model output ISRA SEDES (96) (96) (96) Males 34 32 30 Females 66 68 70 Total 100 100 100 SOURCE ISRA (1980) ; SEDES (1975) Table 16. HERD STRUCTURE AGE COMPOSITION (Model output in year 25 and comparative data) Age groups Model output ISRA SEDES (96) (96) (96) Male calves 1O 12 12 Female calves 10 13 12 Males (up to 1 year) 24 19 18 Females (1 to 3 years) 24 15 16 Females (4 to 10 years) 27 34 )42 Females (above 10 years) 5 7 } SOURCE ISRA (1980) ; SEDES (1975) Table 17. PRODUCTION PARAMETERS (Model output in year 25 and comparative data) Parameters Model output SEDES Shapiro (96) (96) (96) Offtake 8 11.5 11 - 13 Growth 1.3 1.5 2 - 3 Productivity 9.3 13 13 - 16 SOURCE SEDES (1975) ; Shapiro (1979) 117 Sensitivity Analysis The sensitivity analysis is a systematic evaluation of how the optimal solutions change with changes in the input data. It permits us to check the internal coherence of the model and to determine the effects of technical and economic parameter changes, especially those that are crucial to the model. In addition, the sensitivity analysis allows an evaluation of investment returns and parameter improvements by comparing the resulting net EPV with the base analysis considered as a 'without project intervention' situation. In the following section, parameters are studied according to two categories : technical parameters and economic parameters. The technical parameters are the mortality and calving rates, the birth and adult weights, the age at the inflection of the growth curve, the quantity of milk produced and the probability of lactation. Various interventions can change their levels. Research in animal health, nutrition and reproduction can lower the mortality rate and heighten the calving rate. Genetic selection can also influence the growth pattern and the lactation characteristics of the breed, although with a longer time frame. The economic parameters are the prices of live animals and milk and the costs of maintenance and production set by the markets or fixed by the government. 118 The discount factor is studied in this category. The animals' EPV are always adjusted for mortality and other costs. Internal Consistency of the Model In this part, different levels of relative values are generated from the base analysis input and output data. Then they are compared with the initial values. Such an approach allows us to verify the internal logic of our model by checking the results with what may be expected theoretically in terms of directions of changes. In addition, this leads also to a better understanding of the decision-making process, the herd structure and the herd growth pattern, with the capital theory acting as the supporting economic framework. It is, however, necessary to indicate that the analysis is made in terms of relative changes from the base analysis even though that leads to some absolute values that are unrealistic for the area of study. Nevertheless, a comparison of each relative change with respect to the base analysis helps in a better understand- ing of the relationships and a determination of the parameters with the greatest impact on the dependent variables. 119 Mortality Rate Chagggg, Table 18 and Figure 12 present the results from relative changes in the mortality rate for each age class and sex. As the mortality rate decreases for each age class and sex, the maximum EPV and optimal age of sale increase, as predicted theoretically. The important result is, however, that higher relative changes occur with the female because there are cumulative effects of the mortality rates in the own age class of the cow and those of all its offspring. Calving rate changes. Table 19 and Figure 13 present the results of the relative changes in calving rates. The maximum EPV and the optimal age of sale change in the same direction as the age specific calving rates. Furthermore, very rapid relative increases in EPV occur at high calving rates since more offspring enter in the valuation of the cow. Identical trends would happen if the quantity of milk produced and the probability of lactation were varied. Epigegand Cost Level Changes. Table 20 and Figure 14 present the results of the relative changes in the levels of liveweight prices. Prices, EPV and age move in the same direction. It is also very interesting to observe that the relative changes of male EPV are propor- tional to the price changes and the age remains at 4 years. Such a trend suggests that liveweight prices are Effects of mortality on EPV (Male and Female) 120 A_ g # LEGEND Mole -.-- Female -503 -258 -158 BA +158 +258 +508+1OOX Mortality levels Effects of mortality on age of (Male and Female) sale ill? “T 211- 4 Figure 12. \ A— A ....... .s a .. oooooooo A_ Mortality levels -503 -258 -133 BA +15% +25% +50% +1002 Effects of Mortality Rates LEGEND .... Female ....-. ”-... -.— 121 Effects of calving rates on EPV asses LEGEND vases-u / / “T ’ “V ”T, / 1“" -503 -153 +1 53 +503 -253 BA +25! +1 008 Fertlllty levels Effects of calving rates on age of sale " LEGEND 1e" _ Optimal age for a cow IO‘l' ‘40 :21- in.» :1- A r -so:: -1'sx +1'sx +503: -25:: BA +251: +1 00: F ertlllty levels Figure 13. Effects of Calving Rates 122 Effects of prices on the EPV (Male and Female) “‘“ LEGEND _.__. Mal. . ... Panels . . ... a a... o‘.. a... 0‘. 0.. a... .- ....... 0.. .0 sees-1r EWIIIFCFA llIIO‘P -508 +1 58 8508 + 1 008 +1 508 BA +258 +758 +1258 Price levels A f Effects of prices on the age of sale (Male and Female) " LEGEND . . . . Female .... ‘. 180 i ”T «l ...... ... .0 ooooooooooooooooooo a 9“ ....... A 41 a . . A A ' w—r f -508 +1 58 +508 + 1 008 + 1 508 BA +258 +758 +1258 Price I els Figure 14. Effects 3% Prices #— 123 Table 18. EFFECTS OF MORTALITY (Relative changes in optimal age of sale and maximum EPV ) (Males and Females) Mortalitya Male Female levels _ Age EPV Relative Age EPV Relative (year) (FCFA) change (year) (FCFA) change -50% 4 12192 +40% 15 52449 +99% -25% 4 10392 +19% 15 36873 +39% -15%b 4 9709 +11% 15 32193 +22% 0% 4 8725 0% 12 26394 0% +15% 3 7802 -11% 12 21707 -18% +25% 3 7223 -17% 11 19062 -27% +50% 3 5845 -33% 11 13663 -48% +100% 3 3379 -61% 10 6429 -75% 3. Mortality rates for the base analysis are in Table 9. . Base analysis level from which the relative changes are computed. Table 19. EFFECTS OF CALVING RATES (Relative changes in optimal age of sale and maximum EPV) Calving rate levelsa Age(year) EPV(FCFA) Change -50x 8 11662 -56% -25x 11 18256 -31x -15x 11 21309 -19% 0%” 12 26394 0% +15x 15 32213 +18% +25% 15 36537 '+38% +50% 15 49111 +86% a Calving rates for the base analysis are in Table 10. . Base analysis level from which the relative changes are computed. 124 not the major determinant in the optimal age of sale for the males, given the data used. In the case of the females, the relative changes in EPV are much lower than the price level changes, and the optimal age has an opposite trend with respect to prices. This suggests that, as the meat value increases, the producer might consider more sale options of younger cows, although their EPV is less sensitive to liveweight prices. This can lead to important policy considerations with respect to the role of prices on the nature of the herd (cow-calf or fattening). Tables 21 and 22 show percentages that result from comparisons of relative values with the original level of cost and price of milk. From them, the effects of wrong assumptions on those two parameters can be checked. The movements of costs are in the opposite direction of those for the EPV, as expected. But, with the structure of costs used, it appears that they are so low that, after adjustment for time and mortality, they have a minor effect on age of sale and EPV. As was true for the liveweight prices, the milk price also varies in the same direction as the EPV. Changes in the EPV are roughly proportional to changes in the milk price. In addition, the higher the milk price gets, the later is the cow culled. 125 Table 20. EFFECTS OF LIVEWEIGHT PRICES (Relative changes of optimal age of sale and maximum EPV) Pricea Male Female levels _ Age EPV Relative Age EPV Relative (year) (FCFA) change (year) (FCFA) change -50% 3 4220 -52% 15 24502 -7% oxb 4 8725 0% 12 26394 0% +15% 4 10086 +16% 12 27045 +2% +25% 4 10994 +26% 11 27497 +4% +50% 4 13263 +52% 11 28695 +9% +75% 4 15532 +78% 11 29893 +13% +100% 4 17801 +104% 11 31091 +18% +150% 4 22340 +156% 10 33629 +27% :. Liveweight prices are presented in Table 11. . Base analysis results from which the relative values are computed. Table 21. EFFECTS OF COSTS (Relative changes of optimal age of sale and maximum EPV) Cost level Male Female levela _ Age EPV Relative Age EPV Relative (year) (FCFA) change (year) (FCFA) change -1ooxb 4 9077 +4% 12 27380 +4% 03° 4 8725 0% 12 26394 0% +50% 3 8571 -2% 12 25909 -2% +100% 3 8440 -3% 12 25432 -4% a. Initial costs are set at 375 FCFA (Chapter V) Costs are equalised to zero to see the results of such an assumption c. Base analysis 126 Table 22. EFFECTS OF MILK PRICE (Relative change of optimal age of sale and maximum EPV) Price per litre Age EPV Relative absolute relative (year) (FCFA) change 50 FCFA -60% 10 12892 -51% 100 FCFA -20% 11 21805 —17% 125 FCFAa 0% 12 26394 0% 150 FCFA +20% 15 31072 +18% 200 FCFA +60% 15 40533 +54% a . . .Base analy51s price Rate of interest changes. In the data section, the assumption of .20 was made for the discount rate. With the study of the effects of different levels of discount rate, it is possible to check for biases introduced by over- or underestimation of the discount rate. Table 23 and Figure 15 show the results with different discount rates. A noticeable sensitivity is found for the model output with respect to changes in the discount rate. In fact, the discounting is fundamental in the capital theory framework where cattle are considered as capital goods. This is particularly true for the females since their offspring themselves also produce capital goods. 127 Effects of the Discount rate on the EPV (Male and Female) ““" LEGEND seesaw ——“°‘0 .. i ...- Female asses-41- .1 sees-sq» .‘ O O " v—-_—.. ~— ” -——.. 1 1“? 4; v -758 -508 -258 IA +258 +508 +758 Discount rote levels Effects of the Discount rate on the age of sale (Male and Female) " LEGEND 1.1L Male «4- ..-- Female “4,. ............... x ‘ 12+ ‘. ...... gal. ............. “l L -758 -508 -258 BA +258 +508 +758 Discount rote levels Effects of Discount Rate Figure 15. 128 Table 23. EFFECTS OF THE DISCOUNT RATE (Relative changes of optimal age of sale and maximum EPV) (Males and Females) Discount ratea Male Female level Abs Rel Age EPV Relative Age EPV Relative (year) (FCFA) change (year) (FCFA) change 10% -50%b 5 13096 +50% 15 93343 +254% 20% 0% 4 8725 0% 12 26394 0% 30% +50% 3 6435 ~26% 11 11205 -58% a. absolute value and relative value . base analysis The study of the internal consistency of the model provides us with indications on the directions of changes and the relative effects of various parameters when compared with the base analysis. It is, however, necessary to evaluate some changes under more practical circumstances and simulate more realistic situations with the available information on prices and costs. 129 Evaluation of Livestock Development Interventions It is crucial to appraise the impact of investments that can affect the technical parameters and of changes in the economic parameters. The determination of how these investments and changes affect profitability enable us to identify livestock development interventions that can lead livestock producers to make decisions that are consistent with the policy goals set by policy makers. In this study, the data do not permit us to procede with a systematic cost-benefit analysis, i.e, to compare, over time and with respect to known circumstances, the costs of investments and policy changes with their associated returns. Such an analysis would require more precise data, especially for the productivity parameters in the herd and their relation with each age and sex category, the costs without and with the new production conditions and the new production or input-output coefficients. Therefore, in terms of an operational use of the model in the conditions of the Ndama production system of Upper Casamance, the following evaluation is limited to a comparison with the base analysis of the relative changes in net expected present values generated by changes in selected economic and technical parameters compared. This provides us with lower- and upper-bound estimates about the results of major livestock development actions. 130 Changes in Mortalityiand Costs. One of the primary development actions is to reduce the mortality rates by providing better health care and an easier access to veterinary inputs and services. The overall mortality rate of 12% can be reduced by 50% with a systematic intervention on the calves which currently have a mortality rate around 35%. Such a goal would require an annual cost evaluated at 1000 FCFA per animal. This cost is considered as an average cost for a normal preventive and curative maintenance. In addition, the maximum cost that a livestock owner could incur in reducing herd mortality and still break even is calculated. The procedure used to determine the maximum affordable costs involves finding the difference between the Net EPVs in the base analysis and after the change in mortality, and then expressing this difference as an annual benefit. This is done by dividing the difference between the net EPVs by the present worth of an annuity factor corresponding to the number of years equal to the optimal age of sale after the change and a discount rate of 20%. The new age—specific mortality rates used are from 1 to 15 years, respectively : .15, .04, .03, 5*.02, .03, 0.05, .07, .09, .07, .09, 3*.10. They are represented in Figure 3. The results are in Table 24. Their graphical representation is in Figure 16. 131 Changes in mortality and costs (Male) ”” LEGEND ans-V _ EPV (Bass analysis) ween» .... E’V (Changes in M and C) Hausat cm: ;fi:;; - tracers; cafe 1 2 3 4 5 0 8 9101112131415 Age Changes in mortality and costs (Female) ‘°°°° LEGEND _ EPV (lass analysis) .... EPV (Change in M and C) L L a a L TE 6 5 D 5131715131175 Age Figure 16. Changes in Mortality and Costs 125 132 Table 24. CHANGES IN MORTALITY AND COSTS (Male and female) Resultsa Male Female Optimal age of sale 3 15 Base analysis age 4 12 Maximum EPVb 11784 52368 Relative change in EPV +35% +98% Maximum affordable costsc 1452 5562 . Age in year and EPV in FCFA . After changes in costs and mortality rates . Maximum current costs in FCFA that a livestock owner could afford to pay for this reduction in mortality and still break even. 00"” There is a significant increase in net EPV for both sexes, with almost a doubling of the net EPV of females. Such results suggest that a provision of 1000 FCFA per animal for health expenditures is profitable if the overall mortality rate is decreased by 50% and ceteris paribus. For the age of sale, there is an increase for the female as expected but for the male the optimal age decreases relative to the base analysis. Such a result should not be considered as an inconsistency in the model since the net EPV in year 3 (11784 FCFA) is very close to that in year 4 (11731 FCFA). Therefore, it can be assumed that there is a same optimal age of sale for the male. 133 Changes in Calving7Rates and costs. The calving rate is also a critical technical parameter. Its improvement is at the core of development efforts in the livestock subsector. Such an improvement requires, however, substantial investments in animal health, feed for the cows and equipment for good calving conditions. The cost structure presented by SODESP in its interven— tions in the Sylvopastoral region can be used as good estimates of costs in an extensive production system with some actions for its development. However, an adjustment is made when analyzing costs in southern Senegal, where costs related to emergency feedstuffs are not likely to be faced. For the sensitivity analysis, we use the following annual cost structure: Groundnut cake 75 FCFA/KG * 75KG ........ 5625 FCFA Minerals and vitamins 250 FCFA/KG * 10KG ........ 2500 FCFA Veterinary care ........ 375 FCFA Equipment ........ 1000 FCFA TOTAL ....... 9500 FCFA (per animal and per year) First, 9500 FCFA are used to follow the SODESP estimations of complement feed needed by a productive cow. Then, the figure is lowered to 3500 FCFA by reducing the minerals and vitamins by half and the groundnut cake from 75 KG to 30 KG. This is done to obtain much lower estimates for the feed needed to complement the natural 134 pastures. Better health and adequate nutrition lead to an immediate improvement in the 'luxury function' that is the reproductive capacity of the cows. In the simulation run, the overall calving rate is increased from 50% to 70%. The new age—specific calving rates from 4 to 15 years:.50, 7‘.80, .70, .60, .50, .40. They are represented in Figure 4. The analysis is made for two probabilities of lactation (50% and 90%) so that, indirectly,a decrease in calf mortality be introduced. Following the procedure explained earlier, the maximum current break-even costs for each interventions are also calculated. The results are presented in Table 25. Their graphical representation is in Figure 17. In each case, there is an important relative increase in the net EPV of the cow,with a quasi explosive change when the probability of lactation is at 90%. The maximum affordable costs are also very high and suggest that expensive investments for changes in calving rates could be profitable. Also, the results suggest that substantial returns are possible when the calving rate is improved only by 20%, even with high costs. Important gains can also be expected when, at the same time, the mortality of calves is reduced. All results are, however, ceteris paribus. 135 Changes in calving rates and costs (Lactation probability at 907.) LEGEND 'm" ...... EPV (Bees analysis) 1000001} ..... EPV (Change In CR) .....- EPV1 (Changes in CR and C) «seas» -.-- EPV: (Changes in CR and C) IPv1 meets-5500 FCFA : Masts-05°C FCFA Changes in calving rates and costs (Lactation probability at 50%) ""' LEGEND --------- _._.. EPV (Bass analysis) ...... EPV (Change in CR) .....- EPV1 (Changes in CR and C) -..- EM (Changes in CR and C) menu)- mucous-seas FCFA : Nassau-0500 FCFA) Figure 17. Changes in Calving Rates and Costs 136 Changes in Price and Cost Structures. Among the economic parameters, prices are considered as decisive in the determination of farmer decision—making. The following analysis is based on the price and cost structures used by SODESP for the cow—calf herds in the Sylvopastoral region. This allows us to examine the likely effects of a development agency aiming at a reorganization of the structure of the herd in the area of study, according to the stratification framework. For such a framework, males should be sold around one year and directed to growing-out areas. Also, females should be culled when their productivity goes down, around 10 years. These goals are expected to be realised thanks a differential pricing of animals, according to age and sex. The prices used for this section are: -263 FCFA/KG liveweight for one-year-old male calves. This price is the average price for the four different categories of calves set by SODESP: -200 FCFA/KG liveweight for two—year—old steers; -18O FCFA/KG liveweight for three—year-old steers; -175 FCFA/KG liveweight for four-year-old bulls; and, ~125 FCFA/KG liveweight for cows culled after 10 years. For the categories other than those noted above, the estimated market prices listed in Table 11 are used. The results are presented in Table 26 and graphically in Figure 18. 137 Changes in price and cost structures (Male) T: LEGEND uses" __ EPV (Bass analysis) ...-.. EPV (Changes in P and C) 1s.“- _.. EPV (Change in P) uses" messy lash-1r ,/..\ FCFA \\ J:r;er;¢:raertfi 123455759101112151415 Age Changes in price and cost structures (Female) '"" LEGEND ...... EPV (Bass analysis) ...- EPV (Changes in P and C) ...- E’V (Change in P) P L L A A 4“ A L A A A 0 N 4 l 1 4 J 1 1 4 12 3 4 5 G 7 11 9101112131413 490 . Figure 18. Changes in Price and Cost Structures 138 Table 25. CHANGES IN CALVING RATES AND COSTS Resultsa 50% lactation 90% lactation cost1 cost2 cost3 cost1 cost2 cost3 Optimal age 15 15 14 15 15 15 Maximum EPVb 62683 52607 33311 107275 97199 77854 Relative change in EPV +137% +99% +26% +306% +268% +195% Maximum affordable current costsc 7762 5607 1500 22383 15147 11007 3 Age in year and EPV in FCFA c After change in calving rates Maximum affordable costs in FCFA N.B :- cost1 in Base analysis= 375 FCFA; cost2 = 3500 FCFA; cost3 - 9500 FCFA - Cost changes only for females - Calving rate runs are made from 4 to 15 years with no change for the age at first gestation Table 26. CHANGES IN PRICE AND COST STRUCTURES (Male and Female) Resultsa Male Female cost1 cost1 cost2 Optimal age 3 12 10 Maximum EPVb 10494 27046 13041 Relative change in EPV +20% +2% -50% 2 Age in year and EPV in FCFA After changes N.B: cost1 in base analysis = 375 FCFA; cost2 = 9500 FCFA 139 In the case of the male, the price differential is not sufficient to reduce the optimal age of sale to one year. In contrast, for the old cows, the price offered leads to an optimal age at 10 years. However, this happens to lower herders' returns since the base analysis EPV is greater. Therefore, an increase in price alone, with the given cost structure and ceteris paribus. would not be profitable for the herders. In conclusion,the results of the sensitivity analysis in its two parts suggest important research implications that are examined in the following chapter. CHAPTER VII RESEARCH IMPLICATIONS AND SUMMARY Introduction The microeconomics of livestock owners' behavior suggests important research needs. The results provided by the sensitivity analysis indicate particular relation- ships among technical and economic parameters that have different effects on the optimal age of sale and the net expected present values of both males and females. Further research is needed using more reliable data to verify the directions of changes and relative results indicated by our initial analysis. Moreover, the study is limited to cattle production because not enough data are available for the other types of livestock. There is a great need to complete the analysis by extending it to small ruminants, especially goats, which are considered as having the greatest potential in small mixed farming systems. Therefore, it is crucial to get more specific information on the different uses, costs and returns from livestock in mixed enterprises with crops. This will allow a better representation of the production context and trade-offs for an optimal allocation of resources. In addition, the trends in the Senegalese livestock 140 141 subsector, as described in Chapter II, create an urgency for sound policy actions. The role of the Senegalese research institutions is fundamental in guiding the policy process. With regard to such a situation, it is important to allocate research resources in a productive and efficient way. By providing a framework that includes economic and technical parameters, the capital theory approach contributes to a better understanding of the development needs in the livestock subsector and of the data needed to help formulate sound livestock policy. In the following section, the Senegalese research context is described, and then the problems that might be faced in the research process are examined. In addition, a synopsis of research priorities according to the farming system research methodology is presented. Finally, a summary of the study is provided. The Senegalese Research Context Reviewing technology development and adoption rates in various developing countries, Eklund (1983) concludes that the adoption of new technology in peasant agriculture remains lower than expected due to biases in the conduct of research and technology prescriptions by the national research systems. Furthermore, in many instances, insufficiencies in the conduct of research have played a greater role than external and economic factors such as price structures or lack of efficient agricultural 142 support services. In particular, socioeconomic research is plagued with problems related to data availability and reliability (Hardaker, 1979). To solve these problems, greater emphasis is placed on on-farm adaptative research and the farming systems research methodology (FSR) (Eicher and Baker, 1982; Byerlee and al., 1982; Eklund, 1983). In Senegal, since 1982, there has been a general reorganization and reorientation of the research activities in ISRA, with the application of the FSR methodology to the complex array of the Casamance agricultural production systems. However, despite the multidisciplinary approach of FSR, the livestock component of the farming systems studied has not been investigated, except ox-cultivation aspects and inventories of productive animals. Also, the Bureau d'Analyses Macroeconomiques of ISRA (BAME) has pointed out the lack of studies on live— stock production and the need to find a basis for an overall program by doing preliminary diagnostic studies of the organization, functioning and performance of the livestock marketing sectors and the incentive systems faced by livestock producers. This is particularly crucial currently because,as shown in Chapter II,research- based policy recommendations are needed to implement the Nouvelle Politigpe Agricole with respect to the live- stock subsector (ISRA, 1984). 143 By mid-1985, a new FSR team will be created to study the production systems in Upper Casamance and Senegal Oriental. The focus of its work will be the pastoral systems and the livestock component of the farming systems that are predominant in those areas (Landais, 1984). FSR and Livestock Production Systems Despite its holistic philosophy, a major limitation of the FSR approach is its chief focus on cropping systems ( Deans, 1981; FSSP, 1984; Bernsten and al., 1983; Bernsten, 1982). Also, in Senegal, after a long history of on-station and commodity research, there is a great need to encompass the Unites experimentales' conceptual approach for an integrated on—farm research that recognizes the true importance of livestock production in the farming systems. According to Eicher and Baker (1982), in 1982 socioeconomic research on livestock production in Africa was at the same point where research on the economics of crop production was 20 years ago. Namely, many assertions were being made but they were backed by few facts. This is particularly true with research on the behavior of livestock holders. "The causes for [such] oversight are numerous and subject to conjecture " say Mc Dowell and Hildebrand (1980, p.58 ). In terms of policy choices, they are 144 reflected in the overall emphasis on cash- and food—crop production to the detriment of livestock production. Also, more technical problems have been related to the nature of research on animals. Particular attention is placed, in the next part of the paper, on the constraints related to on-farm research on animals and the means of tackling them for greater efficiency. Specific Probleg§_in Rgsearch on Aging; Production Animal-related research has higher costs than other forms of research. In addition, its economic benefits are slow to be realized due to reproduction and physiology constraints. Heady and Bhide (1983) show the difficulties in estimating livestock production functions in the context of the US livestock industries and the numerous questions remaining for the model specifications, algebraic forms and appropriate statistical procedures. With the existence of dynamic processes, including continuous flows of inputs and outputs, there are high autocorrelation risks under the classical experimental designs. Furthermore, for crop research, there is a single or finite harvest time and all selections are made by the researchers, whereas, in animal research, it is the animal rather than the reseacher that controls the decisive variables of feed inputs (Heady and Bhide,1983). In the context of small mixed systems or pastoral 145 systems, identical problems arise with added complexity (Fitzhugh, 1982; Bernsten, 1982; De Boer, 1983; Behnke, 1982). The considerable variations among small farms, herds and physiological status of animals; the lack of experimental control on inputs and outputs under farm conditions; and the very long time frame, with its high risk of experimental breakdown create usually unbalanced and non-orthogonal data sets. Such data sets must, then, be treated with generalized least—squares analyses (Fitzhugh, 1982). Very high coefficients of variation are usually obtained despite the efforts to standardize (De Boer, 1983). One other major problem, which is found in cropping systems also, is the farmer's interference due to his/her own priorities and perception of the benefits from the experiment. In addition, because of the social role of cattle, there is often an "emotional relationship" that is another source of biases in the treatments. "Experience from on-farm cropping research provides some basis for resolving these research problems. However, much more experience is needed with the animal component [of mixed production systems]. While we can theorize as to what types of data or experimental data or analysis are appropriate, the best answers will come from experience." (Fitzhugh, 1982, p.16) Thus, for greater realism, this study concludes with a synthesis and synopsis of the research steps and priority data that have been defined in the literature as necessary for successful research on livestock production 146 systems. In practice, the scheme will be applied differently in different areas, according to the research resources available and capabilities of the researchers. Sypopsis of Research on Livestock Production The FSR basic framework used for cropping systems is generally accepted, even though its history is brief and little empirical evidence exists to show that it has been as successful as expected. Under the name of pastoral systems research or livestock systems research, ILCA is trying, in various African locations, the FSR approach (De Haan, 1983; Bernsten et al.,1983). The objective is to develop methodologies, for specific ecological zones, that would guide and assist national research capabilities. Table 27 is a synthetic summary of the stages in a livestock systems research project, inspired by the general FSR philosophy. In addition, an attempt is made to present an extensive list of data necessary for understanding a livestock production system in the context of the capital theory framework. Among the data, those describing the social and economic roles of women have been often overlooked and need to be highlighted (Safilios-Rothschild, 1983; Broch- Due et al., 1981). This is important because policies, Table 147 27. STAGES IN LIVESTOCK PRODUCTION RESEARCH STAGES OBJECTS OF OBSERVATION COMPLEMENTARY MEANS OF OBSERVATION 1. Descriptive Stage Stratification of the target system Natural,livestock and human resources Non Ecological situation Biomass species and composition Water,vegetal,feed availability Herd size and structure, sex and species ratios, stocking rate,productivity and trends Population parameters, cultural background livestock production (on and off farms) Livestock production Enterprise goals and economic uses Physical and/or financial returns and/or costs Grazing systems Pasture-arable land interaction within production units and at aggregate level Health status and management Watering management Crop residue uses Herd size and demand for labor and other inputs Livestock transactions Market Social relations Home consumption Livestock—related household income and expenditures Secondary information 5011 survey (pedological map) Land system survey (remote sensing) (low altitude aerial surveys) On-farm surveys Herd inventories Progeny history Laboratory analyses (nutrient values, disease diagnostic) 148 Table 27 cont'd. STAGES OBJECTS OF OBSERVATION COMPLEMENTARY MEANS OF OBSERVATION Institutional environment - Social organization of labor - Typology of herd owners and gradient of wealth - Distribution of rights to animals and products in the unit of production and consumption -Marketing and terms of trade for products in circulation 2. Diagnostic Stage Constraints in the production system Priority research requirements Constraints to investigate 3. Design Stage (Researcher-Managed and Executed) On-station experimentation (forage, artificial pastures) Body of knowledge (Health,nutrition management) 4. Testing Stage a. Researcher-Managed and Farmer-Executed b. Farmer-Managed and Executed Identification and Quantitative models of acceptability of output relationships improvements and decision making- process Producer management of improvements Experimental designs Statistical analysis 149 Table 27 cont' d. STAGES OBJECTS OF OBSERVATION COMPLEMENTARY MEANS OF OBSERVATION and inferences (Randomized block designs, split—plot designs) 5. Extension Stage Evaluation of technical Multilocational trials and socioeconomic impacts planned jointly by of improvements research and extension, executed by extension and analysed by research and extension SOURCE : De Haan (1983); Grandin (1983); Eklund (1983); Fitzhugh (1983); De Leeuw and Milligan (1983); Milligan and De Leeuw (1983). projects and even ecological disasters have different effects on the two sexes due to the sexual differential in access, control and ownership of productive resources. Therefore, certain conceptual adjustments are necessary with respect to labor input analysis,economic opportunities or trade-offs and their resulting effects on income and status, management of the herd, and veterinary medicine. 150 Summary The description of the Senegalese livestock subsector shows that its development is hindered by numerous constraints. In particular, insufficiencies in the government policies have been noted. Thus, the study suggests a possible global perspective on the issues that need to be considered in implementing successful live- stock development program. Such a perspective is completed by a review of the competing frameworks of analysis of the Sub-Saharan livestock production systems. Among them, the 'production rationales' framework is retained as a synthetic approach that could be the basis for a structured research program and policy recommendations. Then, an analytical model, based on capital theory, is developed to address some technical and economic issues faced by the livestock owners in the Ndama cattle production system of south Senegal. The model is used to investigate the livestock owners' decisions on the ages of sale for their cattle (male and female) and to examine the impact of different management practices on the optimal ages of sale and the herd demography. The results suggest that the decisions farmers currently make on ages of sale for Ndama males and females and the subsequent herd management are close to that of the optimizing livestock producers represented in the 151 model. Furthermore, technical parameters, such as calving and mortality rates, have more important effects on the farmers' decisions than economic parameters such as liveweight prices and costs. Finally, the modeling exercise provides suggestions on research data priorities that could contribute to the development of policy recommendations for the Institut Senegalais de Recherches Agricoles (ISRA). In particular, these suggestions would help ISRA consolidate its efforts to follow the farming system research methodology in the study of the livestock component of farming systems in Senegal. APPENDIX A A-152 APPENDIX A ANIMAL STATISTICS 1 —Livestock¥population 1947 - 1984 , Senegal Year Cattle Sheep and goats Pigs (1,000 heads) (1,000 heads) (1,000 heads) 1947-52 720 1,253 35 1960-65 1,760 1,790 40 1966 2,425 2,375 85 1967 2,477 2,448 88 1968 2,527 2,521 90 1969 2,600 2,597 93 1970 2,615 2,700 97 1971 2,674 2,804 93 1972 2,508 2,698 154 1973 2,200 2,500 180 1974 2,318 2,533 190 1975 2,380 2,619 196 1976 2,440 2,660 160 1977 2,514 2,811 170 1978 2,533 2,821 184 1979* 2.500 2,920 136 1980* 2.238 3,170 150 1981* 2,361 NA NA 1982* 2,329 NA NA 1983 2,220 NA NA NA Not available SOURCE FAO Production Yearbooks (1965 - 1980) IBRD (1974) Societe Africaine d'Edition (1983) * DSPA (1984) 2 - Meat production (indigeneous)*: 1960 — 1982, Senegal Year Beef Mutton—goat Pigmeat (1,000 MT) (1,000 MT) (1,000 MT) 1960-65 21 6 2 1966 29 NA NA 1967 33 8 3 1968 35 8 3 1969 36 9 4 A-153 Table cont'd 1970 36 8 4 1971 30 10 3 1972 27 5 8 1973 25 4 6 1974 26 5 5 1975 33 8 6 1976 34 8 7 1977 35 8 7 1978 39 10 7 1979 39 9 8 1980 29 8 5 1981 33 9 5 1982 35 9 6 Production of meat from indigenous animals plus meat equivallent expoted live minus meat equivalent imported live. NA : Not Available SOURCE : FAO Production Yearbooks 1965 - 1982 3 -Milk production and imports of milk : 1960 - 1982, Senegal Year Milk animals Yield Production Imports' (1,000 heads) (KG/year) (1,000 MT) (1,000 MT) 1960-65 531 370 107 NA 1966 700 NA NA NA 1967 770 NA 118 NA 1968 820 NA 115 NA 1969 830 NA 113 NA 1970 860 NA 110 15.853 1971 313 NA 116 16.722 1972 314 300 94 14.170 1973 275 300 79 8.853 1974 283 320 66 8.013 1975 260 370 91 8.503 1976 270 350 97 8.410 1977 270 360 95 17.058 1978 271 350 95 14.564 1979 271 350 95 17.058 1980 250 350 95 14.564 1981 225 360 81 NA 1982 230 370 85 NA SOURCE : FAO Production Year Books 1965 - 1982 -Societe Africaine d'Edition (1983) APPENDIX B B-154 APPENDIX B MATHEMATICS OF THE CAPITAL THEORY 1.Discrete stream of values 70 ll 1 R0 (1 + r) 70 II n R0 (1 + r)n RV = Present Value of the discrete stream PV = R0 '0 < :1 ll Rt (1 + r)‘t 2.Continuous stream of values R = 30 (1 + r/f)fn Rn = R0 ((1 + r/f)f/r]rn (NB : r/r . 1) :0 l -RO[A]rn If f goes to infinity, the system is countinuous and A goes to 2.7183 or e. If f smaller than infinity, the system is discrete. n PVn Rit) e‘rt dt 3.Compptation of the gecision rule Given Pr(a) and Pr is profit, a function of age, its maximum is reached if and only if the following conditions are satisfied ,at the same time. - the first-order condition, id est, the first derivative of Pr with respect to age a is equal to 0 ; and - the second-order condition, id est, the second derivative of Pr with repect to age a is negative. APPENDIX C «Jam OCCOOCCCOOO~*OC CDC-'VHW'DLOFOCIO-Nfl @hhnhhhhnhhhooo 30000000000 nohem-Nfivmw --———«nnnmn anA: «f ..z...n a as. u...>_._nzsm :_ 2. sum: «- eo—oa w:p o ‘0 a.u «seem a so o1 xx>im 0 u:.— 02¢ s<>_>¢:m use 0.. n:saeh u . < 3.35.:a:¢.w sea a sea. as a m<>im o .zsaozi u.1 m>__<.:x:3 o.>in o .zmcoa; ..41u. so u Asa: a sz.>__«.:::o x>>am o usam zcza ..asaa xa eso.>a o . pa slow a .az ox>i op>a a _ sea» 2. a 0m .e!.2< 25:) :~a_a be >a ax>a ¢>>a u assm a: .04 ass—pea x_«330 ~.a~za:c as. nuasa> ego sxo 0:0 as as as u u. m~..x..x La uo_aa x._six o Asa¢s 5:05.555.» zo__<—o<. _u<.a a «pee .muau_z_ .03 u :o_s<. .zzccm s a u .Aa_. s24 .¢>_>zon so. sa.nswo< o emu—n < 30 pm 0 0»uo u :30 a so .mso u>_.<.:1:o “a :00 u :04 s we nmoo .zueaau . «so u .u_~_oaim use s_a¢ »__...auu ex 0 oz.«.«« Lo .mou 0x a. 0 «Au. .2 e330 2. u:.<> paw: >x >> o .01 _. 1A 3» u .w _A_w an ua<~ «be: .a>_>:om mx m» u so an sea a 4: »._s<—zca xx x» u 321:: a s o o uxaz «.xoo >.¢uu_m u U000000.00.........OOOOOOOOOOO00.000.00.000eO...OOOOOOOOOOOOOOOOO'OOOOU o ,ao_amz_o Lama—ca) o u usem ‘0 was . a «a . Nos 3 .ca a «a . «co ax -. a x m»;a e «o \o on..n o_onm o cow.n. o om.so .a» a so . . . . . oem. Goa.n oo.s.«uu so as a «a \.a, e., 5.. n., a...._. a .o o no..n no o.. an“ a> a «a 0 ma. 0 coax 25.:ax . «ix 4 «a . can moe.« co n.m.¢ac u«o 5x5 4 «a \a o o.+\ .xaa <3o+>a xx>a .m4>am x >am « «a .\o.on..a. a on.. o oqm.n 0.0m..a ax A «a . .:o« .a. .0“ on. .oe..«,o.,.n as. am o.o.n \ ax a «a \.~ 0. 50 .o no no .a as o_. ma. \ 4x a «o \ooou n A x «Eaa mm see. we ~0aw. 0. m memo Laue awn 5x0 ..uxm .0». ... x same .~.3 5: aa. a. aza.“n ma.. . a; same a >> a. u». 3» a. a. a. a» saw: u u auhm>m 2<0_sa< sezo_~_oA : 3:. A .3 .:.¢ ...: .....z<: : ...:«a...aa..fi«. .cAo- A.Jx n.. ._ a co n...3..) 2... ..a.x‘-o....A..nA.A1~mu o m..«.a v on A. :~.0 ..Wn ... UUIDUU 7UU r A¢AUuLx A.rau ( 33A u.(n A: ~00. xA.vww<8¢Ou v Ava u..:) USU Uoooooooo.oooooooooooooooo-oopoooooooooooo.ooooocoo-cocooooooooooooooou U ICU d «On wAdm no add U90000.000000090000009...00.000000.ono.oo.o0000QOOOOOOOOOOOOOOOoooooooU U an. 0 AAA“) U <13»); >-‘<1 ma3A<> AdxAAAO .2 .An 0.A oA Ao.. x. «Sac ...xA. AAAQ. A..mA. ...>A. ...Aon=.>;o:AuumzA>m. ...9. WA...:) Cum—m 4 33A mAAanuz AZASA ..A.‘¢.A:oA> an a. a: A; A. =A>a .. CU GZA~A1—1 GA A0uzxu :uA2304mA0 A AJ< zuA3334Am “IA . uADJ ZOAmAUA: 0 ¢A>¢ 0A); . A uA>u A 0. A. :— UUUfinUU UUUUU . 2. AAAQ.“- Mum-wu UUn3>UJ A n uADnDJ o.w.A ms ....11 ....xA..3 .Ai >:..AM_.aAuu.. w..:.»a.fl AA .Aa. ..o.w «sue. on 3mm .MAA.¢: aA>z .: 3..<. axau O O O>UU A: 23 O J» A. -Q «AAA:— 300 JO ...—03A..no.p «AA AAa>.AA ::.u<. .z330n a 324 :. A....:a.o‘...au...na.A ammo. m. E: A A........«..An,....._..ww,,..... 2%.... ... "...: 25:, 2,: A.5.1A-a,.....AmA.“wwmA ..a.1 o. a .mu_.>::m .: .u2333. .>..<.:::u. wuz<_u.mxu. A. .. A Am «A ; b). -U O>O A a 3 d ‘ > 3 00 3 "UUU 'JUU non.A a Z.» w: .3 \1 \n \< .32233.AnA;3 nhA\vA AOAwaax ICQOOIA o-Nfl'mthaC-NfivmQFOOO-NHVW¢FGmo-N --.“m--‘“W--‘HVHOKVfiOHNflCHanNWHPHQHPHW'WWW'VVNWYVHWNOW ao-«nvmmhom 0 pm chemo-«nvmwhcmO-«nvmm ~~~~oacooooncommmmamm «31; ——---—--—finnnmmnnn OOOOOOOCOOOOCCGOO —~nvnohaoO-~P-JJ~ 0 r. -~fiwxha OOFNNMN~~~~HWMSQQ¢MMDGOQ 000000 8 -NH'mQh-nm ------—-_--_-_--_‘--__--_ 308000000000 30" an 0' «A A.«!. u zaocx.: m.>._>a:m .3 A....m«mo¢¢.Amn.Anm .~mx o u A.m :3¢. :3. ..«x .<>.>33n .>..<.:133-A . ma w u u A33.u.x.3 ..3¢.3¢> w u U...oOOOOOOOOOOOOOOOOOO.I...o0.000.000.0000.0.000000000000000000000000U u u u «>33< 3.A<.:u...A.A«x.ao.o....¢a m .nn.o.u..zz m .n 3....3 3....303..~.3 o .3” .on «sec. .3 . . . . . AA Ax. :4 A 1x . ...)A ...mA.:A>.n xx>;n.3A>;n nA>zm 3A>. ...Auu :A>; o o A .2) 3w m..3m.a .z.:; w ..A.Aa.:13A> on 3. 33 .<13.>¢ A. =A>mm .. ..3: 23.. 3.3 u U ...AJu.oA>..c<>. :.3.3 .¢ >3 w .AA>;n 3 ..x.._.Auu.zA>...3A>. AzAaoag u . n_. .3 .344) .z. :z_::.az. :.z.3 .¢ :33 ¢ .3 A. w mAAAm.xx>¢n.oA>¢n.z<>m . .«ut 32¢ Az.:3:. ..«x.x..x .3 :.:.a .« A. w x13A>¢.AA>am.3A>am , AZAJQIA «AC! .3 >8 u>AA¢m.xx>¢m 0..33:z. x..: .0 >3 0 U u.w>x.._.c. . m..mx>cm 3.2..:33<.w .. 3. < .3 .3 A ...: >a w A...A.o...3¢u A. x < .3 .uz.zzu o .3 A....3«ooaa 3 A_a.3a .....A. “ mx.xA>;n.xA>am Azuuoas u.x «m ....u...xo..AA mm. «m .x: mm..xm.. A.mm. An . . A>.m. 1n .0A>a. x9 ....a. (a .A. An .0.w «:30. an .33 ...31 u 333. .4.2 2. 3..3zxau .m..3n.: ...: :3. .3... 3.2..z.u. w v :<.A .< a_a«.w :3.A0333¢a.¢ u Azaaaz. .0 m.:.«> w .1.3..3.mA..3.uA.A..xuu..n.uuw an m..n.3 on 33 :33 « 3:32.. ..3. . 8 .2233 A... «:13... 3 AA 3.AAAA¢.AAAU‘a AAiUU OI\nJ\v3 fianvA a 2AA A: .3 \1 \A \1 .32338.AaA10 ahA\Vh AOA4331 8(3323; N0! V\V\V\‘\V|V\V\V O-Nfl'mmhdflc‘fifivmWFQO 8 nvmohomo-finvmmhomO-«nvmwhoao-«nvmohomo-Nnvmchom nmmmmmnowwm000mowhnsnan~~~~¢oooooooooaamamammam x1.v.v.v\v.‘-v‘v‘wAv‘v‘v‘v‘vuA‘v‘v‘v‘v‘wuv‘v‘vsv‘1“\V\V|‘\v 03’. 0000000008 —~rurncnwnmg-"Vfiwuum~oa 000000000 On 0000080000000 -~nwmo~ 00500 1001‘. .3 naxnaxva «14.3.0 .0 ZS.—<.3&100 2:..." w Zuni“: JO 3.. ..m «33.3 3... nn3.3.w23333.2 :33. 3..m233. 3 «\m.>.3.-3> «\m.>.33-33 3.3.33 23 3. . .3 3 .33 3.» u .3: 3 . .... e a 3 x. 3 . . . .3 3 3 ..3._ .3 3..3 3.....3 33 3: m. 3. n n33... 3 ... na3.3 3. . n 3.3 :33. 33.3233. 3 o 3. ..x32 . ...333 3.3.3 > o 3. 3.332 x 3.333 3-3.333 .n..33.. 323 n..33. we“ .33. 33 .3 .33...3 u ...3. . .3.nu>.33.n.>.3u 33 a. 3.. 33 33 3 o.m.>.33 n.>.33 .3 3.33:2 3 ..33..3.33.>x. mm ..x 0.. 33 «33.» an 33. 23.....3.3 w “3.“..3H“3~>~”3>H33u.3 «n 3* u..33 a. .-3 3 3 33 33.»: .p 3 ...33 ...33...3 . . . . . , . . - u - . m. 3. . mm3.3 . - -. . - - - . . - . . ...323333.2. 3...33... «.*.3333. .3 ... 3 ...33 3 3 3 3.3» 3,3.3x . 3.33.33 m3...33333 323 32333332 33.» .3 23..3w..3...2. w 000.00.00.00.30.0.0000000DOOOOOOOOOOOO0.0...OOOOOOOOOOOOO0.000.000.0000 3 33.: 3:. 2. m3.332.3 3 U...OOOOOOOOOOO00.00.000.000.oOOOOOOOOOOOOOO0.00.00.0000......OOOOOOOOU 3 .333..mm.A ..m«.ooo.... 3 3a. 33: ... On 3 .3nn..wn.\ ..m..333.. . 3 33. > 3 3 ... 3'. 33 ..w3.3mm.3mm an. 3.33! .-. On. 33 3 333mm ...m...mm..mm 3a. ..333 .-. 33. 33 3 3..nm 3 U.30030000003330.0000...-0033o3o0333333303303300030030333030.3330.ooooU u m332_23 033. 23 :33 .33—>335 .32—3.3350 u u u:- :._3 23.—33.3.».3 33.: .33...:.33.: 23 .3 23.—33.2.3 u UOOOOOOOOOOOOOO09000000000030...o.o0.09.0.000000000000000000000000.000U 3 name ...3: . an ...33 ..mu.332.3 33.:. non ... 3233. 93 .3n ...33 Us00.00.000.000...0.00.0.0...OOOOOOOOIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOU u u u x 33.» 2. >—.>..333333.3333 3 3 3 33.» 2. 3.33 2—Jo3w..3.3 3 3 3 33.» 2. .33. .3- x...3 3 3 x 33.. 2. 33.2 .3_3p. 3 33.2.3. 3 3 33.. . .3 33.3. 3).—3.33: . 29:3 3 3 3..3.33_m 33 3.3 ..3 3 u wzumpw .3 3 3 _ > u 3 m333 .3 3 . . 3 3 u .33 1:33:33 p3 3.3m curve 3 > o u .33 1:1..33 p3 3.3m aau.3.3m3 u 3 23333.2 3.3.33 3.33.3» 3 .33.. a 2.. n3 .3 \3 \n \3 .32333....33 a..\vn .3..33x 2333333 o-«nvnmhoao-«nvmoh000-«nvmwhom0¢~nvnwhamo—Nnvmohomo-Nflvmo~ooo'flfivmwha annnnnnnnavvvvwvvvvvmmmmmmmmmmoomoomocwwh~~~annmascmmmamaaocammawmomm nouaneuwaouunen~ncu~~cu~nou~nru~neuNncunaouwnouvncuwmenu«auxneuwaeuxncuwneuwneuwnouw«runnoun 43¢; - ‘A —~¢"-r:1‘a~ no: —"«fl—r Nflfiflfififlfiflflfififififl'fififlv-y-r-y-y-r-y HHHHHHHHHHHHHHHHHHHHHHHHH 300000000000000000000000 ‘CQ‘QntDBQO‘ man a? a. na\90\10 32. ..Q-m . .32.p233 333 .n 3.. 333*.3133. .. 33>3..b 3 .p.3J vx3333..33>3 232.233 333 Amt..x.v.=23.3293..oafio— v..~xo:vco 3 333333. a 3..x n 33 3.3. x 333 ...3333. 3.33.: .3 :.:333 .333.>3..33. . . 3233. a. .33 3 up.3J .33.. v .3.>. 2.3333 33.: .333.>3 .3 :3..3.33x33 3 .32._:au «xXS . .v‘n...o 33.3,...H33.3 3......3 ...”...n. xonw.3333. as 3333 .3.3 .3...3 .3.33.:.3. 33... 3.. ...33 an 3 ...33 3.3. 3.333 ax“..o..o.3-3333 .x.33.:.3.\..3.33.:.3.3...x 33.:—3....3.3 ..33....33.>. 33 ..x coon 33 3u33.». 3 0.3333 3 3._3.3 . . A3Auavziu.‘o.3—°-J~ 0 33 ...3..33..33.:.3...3. .33.. ..xo.w.3333. '5 v. ....31 ...w.3333. a. an ....33 ...)..333333 323 ..33 :.3333 .3 23..3.33233 w u . 233..233 3.w ..3.3.».. .. { ..3~.3‘ . ..33. .m...a .3 .3333. ~s .39.“....73.”..«23331;:.£..._33 . . . 3 3 . 3 . . o 3 ' wb— 3.3“33 “N sadan‘3 0 3. o a.— «m 3 m.m. xa w.m.. n. x. .o. "3343. .5 .3...3 3.333 n. ..3 .3 3 33 33..3 ».~ 3 up 33 m. 30.3 .2.33 w 3:.37Mx 31:3— «3:3.\.3.3m».3.3 3 u x”..3 .3: 233 333. ........3.Q..3..::3. a. ... 333. 33 3:373:3— hln.. a 2-5 n34.3.\x \n \3 12:33....33 3.13.. -0 daze! 1(10033 U) .4 .1 3c3——-—-—----~n«nmnmmnnnnnnanmnnnvvvvvvvvv«mommmm finnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn U! SAIE ‘0“ A Sltlfl AGE ccvu rvvu vat!) vvtll vstll DtII val!) PVVR momma-~flmm-—v-o -~vv30903 fi-m~3m ammvnfinn---OOOO OUOOHFNOQGOCHOH anfifiOOCOU’1M‘COQ ammooovvwnnan—- v~~flfl~cmonnm~n 'VVWVYVVVVV OCHDOONPONUSFOGGUI OfimflO‘NHVVWVfi'V --nannflflfiflfinnfi v~nmm-mo-mnmonn ,flOfiBRC woohohv $v~~omn 050' -n QNOOFOmVN-— - os-«.a¢¢a..¢aaa vmmnoflooo-NHHM -«nvvmmmoouoou cans nvoona-«ah $hoh mnmnanamo anach—cmnnmwovn 'PQOOFOQON¢- -«nvnoh¢GO-«nvm -_-“- 'CIA 8H5. IS 4 MAX IOIAl PV OF A Slttk Al AGE 0' SAIE fUR A LUU ccxll) vau srvxs svvvu spvux svvvx xslll xvll) lull) Anti) Attlb PVKN 383333333333333 fl mflfifimfifll ~07qu 3—3333333--—--~ 3‘~QF~HOOOOOOOC n-moo- “an --—nnnannmunmn o-n-~Nn~ovoommm na-a -3wnonouma som- 31330-3mmn OOHO'NHH-NNO!~¢~~ -fififlflfififlfiflflflflfl OWOGHhfiODHWQHOH anHHOQvOQ-Qmov anmnmovvvnnan—- §§§:v-o--mva«nu nav-avmoos~~ ”Nflflflflfififl 300::oooanovaa¢ wmwm Haan :0monoamam o- «mm-nmmmoopa -----‘ 300:0;OOS-~o«om afiomnmo ;; oaoo—«nnnn —-—«ananan —mflomddnnhvscnm V‘DF‘OW'HHN- ‘C'fia'c'fihm'QON vm-O000035mavmv Ono vmoovwhannnn vmomn:onvmmoocw «nannma Ohufl'flQom'hOflOO Hanan 05°-flflfi fl ~«nvvmmmooumco -mfionmvv00w 33- U Nflfiflfiflflflfl ~«nvuwhoao-finvu ----‘- OCIA 26384 W MAR lUlAt PV 0? A LUH A] "(H0 DVNAMICS UfilAKC l0 lb LIASS I VtAR NEHBUHNS DO 33 33 NO. I. 33 00 Old) 34. RSUID Off- 66‘ VSOlD 34. 33 33 33 SH OIIJ ASULU 00f“ V5010 33 33 aiai «a Ulbb lSUlD Off. 34 6|. VSOLD 33 33 33 so so an sh 33' an 0137 0 Off- 33 33 33 «'3' 55 ha Rh fiO ;_ " UHUJ 55010 055‘ 34. 75. V5010 33 33 33 on DH 59 BB GO 00 am Ul9b ASUID 0ft- VSOtO 34 75. 33 OO 33 on ~~ 5.; DR a“ 8. «a @O N" -- 90 Ola: ISU‘U UIF‘ 34. 75. V5010 3° 33 33 an hp 80 NR TH"! 8‘ v6 33 'V -— RF 0114 KSUIU Off‘ vvn' GD 33 33 33 an RR 30 50 vv 80 9'3 016? AbUID Off“ 34 76 VSOtD 3o 33 33 0’64 ASUID 0ft- VSOID 6: 33 33 00 33 DO 016! 4, 15010 Ult- 70, V50l0 3 34. 33 33 .0Ibfl KSUID Uti- VSOID 19. 37. I 33 33 ad -- NN UIBU ASUID 0f!- VSOID 3° 33' 33 s3 a F” -- an UIII ISUID Uffl VSOIO 37. II 33 33 33 66 so no no «a 33 an V? Ollb ASUIU Uff‘ 37. I2. VSOID 33 33 33 an so 68 an rid 33 TV Oil) ASUID 05f- VSOtD 3i. I3. 33 33 SIN OD U77| Oil' 33 33 33 .33 33 an ad 33 8:4 am ~R O¢ DIIU ASU|0 00}- V5010 30. .5. 33 3'3 90. 90. 90. 95 5h ’- ‘- .0710 55010 0”“ J9. 0‘. V5010 33 33 3'3' 3' an N“ am 30 $0 OIIU 150‘” 090' 33 33 33 33 33 ~66 no N“ Q— 30 «a .OIII $0!” 00}- 40. I I9. VSOID 3° 33 vv .- NH .UIII 40‘ lSUtD Olf' VSO‘D 90‘ 33 33 33 @— 1“!“ H“ OI]! ASOID 0ft- V5010 4‘. 90. 33 33' 33 33 .‘H'! N“ 33 33 33 n3 .1 N5~D 6.0w» . 3.3 3.3w‘ vu .av 30 33 an -- D!!! N“ 0030 . 0000 3000. 3090 3060. 3000 0000 0000. 0000. 0000 0000 6000. 0000 .000. 0000 0:30 0.00. 0.60. 3600 6.00. 3600. 6.00. 0000 .000 00!... 66.0 66.0 66.0. 66.0. 66.0 06.0. 06.0 .6.0 06.0 66.0 .n.0. 6~.0. 06.0. 03.0. 03.0. 63.0. 03.0. 03.0 On.0- 00.0. ..00 .0.0. ...0. O..0. .c.c .m130> 3.030: 50 2 «..0. 6..0. ...0. ...0. 0..0 0..0. 0..0. ...0, a..0. 0..0. ...0. 00.0. 00.0 .0.0. 30.0 .0.0 3..0. 00.0. 00.0. 6000. .0.0. 00.0. 6..0. fi0.0. ..0 .06.. .6... .000. .000. .030. .360. 600. ..00. .000. 060. .060. .600. .603. .603. .33. 0N3. .003. ..06. .606. .036. .066. 366. .006. 306. 0am:.0— 23 9.3 v.3 23 3.3 .33 3.. na an .n 3a m. 80 003 830» BIBLIOGRAPHY BIBLIOGRAPHY Abassa, K.P. 1984. " Systems Approach to Gobra Zebu Production in Dahra, Senegal." PH.D. dissertation. University of Florida. Aguilar, C., Friedli, C. and Canas, R. 1983." The Growth Curve of Animals."Agricultural Sytems Vol 10: 133-147. Anteneh, A. 1984.a. " Financing Animal Health Services in Some African Countries". O.D.I Pastoral Development Network.Paper 17b. 1984.b. "Trends in Sub-Saharan Africa's Livestock Industries." ILCAggulletin. 18. Ariza-Nino, E. and Shapiro, K.H. 1984. " Cattle as Capital, Consumables and Cash: Modeling Age of Sale Decisions in African Pastoral Production" in Simpson J.R. and Evangelou P., ed..Livestock Development in Subsaharan Africa. Boulder, Colorado: Westview Press, Inc. 1984. Atherton, J.S. 1984. " The Evolution of a Donor Strategy for Livestock Programs in Subsaharan Africa" in Simpson and Evangelou. ed., Livestock Development in Subsaharan Africa. Baker,P.R. 1981. " The Social Importance of Cattle in Africa and the Influence of Social Attitudes on Beef Production" in Smith A.J, ed.,Intensive Animal Production in Developing Countries. University of Edinburgh. Centre for Tropical Veterinary Medicine. 1981. Behnke, R.J. 1982. ed. "Communal Area Livestock and Range: Research Priorities." Communal Area Research Group, Ministry of Agriculture and ILCA. Gaborone. Botswana. . 1983." Production Rationales: The Commercialization of Subsistence Pastoralism." Nomadic Peoples. No. 14. 164 165 1984. " Fenced and Open Range Ranching: The Commercialization of Pastoral Land and Livestock in Africa. " in Simpson and Evangelou,ed., Livestock Development in Subsaharan Africa. Bernsten, R.H. 1982. " Analytical Approaches to Farming Systems Research with and Emphasis on Animal Production." in Fine, J.C. and Lattimore, R.G, ed., Livestock in Asia. International Development Research Centre (IDRC). 202e. 1982. ,Fitzhugh,Jr, H.A and Knipscheer, H.C. 1983. "Livestock in Farming Systems Research." Winrock International Livestock Research and Training Center. Berry, Sara. 1980. " Rural Class Formation in West Africa". in Bates, R.H. and Lofchie, M.F. Agricultural Development in Africa. Praeger. 1980. Broch-Due, V., Garfield, E., and Langton P. 1981. "Women and Pastoral Development." in Galaty, J.G, Aronson D., Salzman, P.C. and Chouinard, A., ed., The Fpture of Pastoral Peoples. Ottawa, Canada. International Development Research Centre (IDRC) 175e. Brody, S. 1964. Bioenergetics and Growth. Hafner Publishing Company , Inc. New York. Brown, J.E., Fitzhugh, Jr., H.A. and Cartwright T.C. 1976. " A comparison of Non—Linear Models for Describing Weight-Age Relationships in Cattle." Journal of Animal Science. 42, (4), 1976. Bonte, P. 1977. "Troupeaux et Familles chez les Eleveurs Saheliens." in Les Systeps Pastorapg Spheliens FAO. 1977. Burt, 0.R. 1965. "Optimal Replacement Under Risk" J. Farm Econ. Vol. 47. 1965. 324-346. Byerlee, D., Harrington, L. and Winkelmann, D. L. 1982. "Farming Systems Research : Issues in Research Strategy and Technology Design." Aper. J. Apr. Egon. Vol. 64. No 5. 1982. Chisholm, A. H. 1966. "Criteria for Determining the Optimal Replacement Pattern." J. Farm Egon. Vol.48. 1966. 107-112. 166 Christensen, C. and Witucki, L. 1982. "Food Problems and Emerging Policy Responses in Sub-Saharan Africa" Amer. J. Agr. Econ. Vol. 64. 1982. 889-908. Chudleigh, P. D. and Cezar, I. M. 1982. "A Review of Bio- economic Simulation Models of Beef Production Systems and Suggestions for Methodological Development." Agricultural Systems. No 8. 1982. Crawford, M. A. 1975. " A Reevaluation of The Nutrient Role of Animal Products." Proceedings of the Thirld World Conference on Animal Production. 1975. Crotty, R. 1980. Cattle, Economics and Developgent Commonwealth Agricultural Bureau. 1980. Dahl, G. 1981. "Production in Pastoral Societies." in Galaty et al. ed.,The Future of Pastoral Peoples and Hjort, A. R. 1976. Having Herds. Stockholm. Liber Tryck. 1976. De Boer, J. 1983. "Design of Alternative Crop-Livestock Systems for Asian Farmers." Working Paper No. 21. Research Institute for Animal Production. Bogor Indonesia. October 1983. De Haan, C. 1983. "Towards a Framework for Pastoral Systems Research in ILCA." in Pastoral Systems Research in Sub-Saharan Africa. ILCA. August 1983. De Janvry, A. 1984. "The Political Economy of Rural Development in Latin America : an Interpretation." in Eicher, C. K. and Staatz, J. M., ed., Agricultural Development in the Third Worlg, The Johns Hopkins University Press. 1984. De Leeuw, P. N. and Milligan, K. 1983. "Integration of Remote Sensing Techniques for Resource Evaluation." in ILCA. Pgstoral Systems Research in Sub-Saharan Africa. Deans, R. J. 1981. "Farming Systems as it Relates to the Animal Sciences." Working Paper No 5. FSR group. Michigan State University. 1981. Delgado, C. L. and Mellor. J. W. 1984. "A Structural View of Policy Issues in African Agricultural Development". International Food Policy Research Institute. Washington, D.C. July 1984. 167 Dent, J. B. and Blackie, M. J. 1979. Systems Simulation in Agriculture. Applied Science Publishers, LTD. 1979. Doppler, W. 1980. The:§conomics of Pasture Improvement and Beef Production in Semi-Humid West Africa. GTZ. Eschborn. 1980. Doran, M. H., Low, A. R. C. and Kemp, R. L. 1979. "Cattle as a Store of Wealth in Swaziland." Amer. J. Agr. Econ. Vol. 61. No 1. 1979. Doutresoule, G. 1947. L'Elevage en Afrique Occidentale Francaise. Paris. Editions Larose. 1947. Dumont, R. and Mottin, M. F. 1982. "Le Defi Senegalais." ENDA. Series Etudes et Recherches. No 74. 1982. Eicher, C. K. and Baker, D. C. 1982. Research on Agricultural Development in Spy-Saharan Afg;ca: a Critical Survey. Michigan State University. International Development Paper. No. 1. 1982. Einzig, P. 1949. Primitive Money. Eyre and Spottiswoode. London. 1949. Eklund, P. 1983. "Technology Development and Adoption Rates." Food Policy. Vol. 8. No. 2 . 1983. FAO. 1976. "Possibilities of Increasing World Food Production." FAO Freedom from Hunger Campaign. Basic Study. No 10. 1976. 1977. Les Systems Pastoraux Sgpeliens. Etude FAO Production Vegetale et Protection des Plantes. 5. 1977. 1982. "Crop Residues and Agro-Industrial By-Products in Animal Feeding." Proceedings of the FAO/ILCA Workshop held in Dakar, Senegal. 21-25 September 1982. FAO Animal Production and Health Paper. No 32. 1982. Ferguson, D.S. 1976. "A Conceptual Framework for the Evaluation of Livestock Production Development Projects and Programs in Sub-Saharan West-Africa. CRED. University of Michigan. 1976. Fine, J.C and Lattimore, R.G, ed., 1982. Livestock in Asia. Ottawa, Canada: International Development Research Centre (IDRC).202e. . 168 Fitzhugh,Jr, H. A. 1982. "On-farm Research Involving Animal Components of Farming Systems." Paper presented at the reunion for Grupos de Trabajo en Sistemas de Production Animal, Pucallpa , Peru. January 1982. Winrock International. Foxall, G. 1979. "A Note on the Management of ‘Commons'." J.A.E. Vol. 30. No 1. 1979. Galaty, J.G., Aronson D., Salzman,P.C. and Chouinard, A., ed., 1981 The Future of Pastoral Peoples. Ottawa, Canada: International Development Research Centre (IDRC) 175e. Gallais, J. 1972. "Les Societes Pastorales Ouest-Africaines Face au Developpement ". Cahiers d'Etudes Africaines. Vol. 12. No 3. 1972. Girardot-Berg, I. 1982. "Livestock : Policy Issues in Trade, Pricing, and Marketing." in Fine and Lattimore, ed., Liyestock in Asia. Grandin, B. E. 1980. "Small Cows, Big Money : Wealth and Dwarf Cattle in Southern Nigeria." PH. D. dissertation. Stanford University. 1980 .1983. "Livestock Transactions in Data Collection" in ILCA . Pastoral Systems Research In Sub-Saharan Africa. Gueye, E., Ly, C. and Diallo, A. Z. 1981. "Bilan de la Pre- vulgarisation au C.R.Z de Kolda." ISRA. 1980. Hardaker, J. B. 1979. "A Review of Some Farm Management Research Methods for Small-Farm Development in LDC's." J. Agr.p§con. Vol. XXX. No 3. 1979. Hardin, G. and Baden, J. 1977. Managing the Commons. W. H. Freeman and Company. 1977. Harsh, S.B, Connor, L.J and Schwab, 6.0. 1981. Managing the Farm Business. Englewood Cliffs,N.J.: Prentice-Hall. 1981. Heady, E. O. and Bhide, S. 1983. Livestock Response Functions. Ames: The Iowa State Press. 1983. Herman, L. A. 1981. "The Livestock and Meat Marketing System in Upper Volta : An Evaluation of Economic Efficiency." PH. D. dissertation. U. of Michigan. 1981. 169 Hopcraft, P. N. 1981. "Economic Institutions and Pastoral Resources Management : Considerations for a Development Strategy." in Galaty et al., ed., The Future of Pastoral Peoples. IBRD. 1974.8enegal: Tradition, Diversification and Economic Development. A World Bank Country Economic Report. 1974. 1983.Wor1d Development Report. World Bank 1983. ILCA.1979. Trypanotolerant Livestock in West and Central Africa. ILCA Monograph. 2 Vol. Addis Ababa.1979. 1981. "Trends and Prospects for Livestock and Crop Production in Tropical Africa." Working Document No. 5. Addis Ababa. 1981. .1982. "Cattle Herd Dynamics : An Integer and Stochastic Model for Evaluating Production Alternatives." Research Report No 2. 1982. ISRA. 1981. "Rapport d'Activites. Annee 1980. C.R.Z de Kolda." ISRA. 1981. . 1984. "Orientations et Programmes de Recherches Macroeconomiques sur 1e Systeme Agro-Alimentaire Senegalais." BAME. ISRA. 1984. Jahnke, H. E. 1982. Livestock Production Systems and Livestock Development in Tropical Africa. Kieler Wissenschaftsverlag Vauk. 1982. Jarvis, L. S. 1977. "Cattle as Capital Goods and Ranchers as Portfolio Managers : an Applicaqtion to the Argentine Cattle Sector." J. of Pol. Econ. Vol. 82. No: 3. 1977. . 1980. "Cattle as a Store of Wealth in Swaziland. Comment." Am. J. Agr. Econ. Vol. 62. No 3. 1980. .1982.a. "To Beef or not to Beef ? Portfolio Choices of Asian Small-Holder Cattle Producers." in Fine and Lattimore, ed.,;ivestock in Asip, .1982.b. "Discussion Summary." in ibid. Josserand, H. P. 1983. "Small stock, Large Dividends : Sheep and Goats in Sub-Saharan Africa." Discussion Paper No 105. CRED. U. of Michigan. 1983. 170 Konczacki, Z. A. 1978. The Economics of Pastoralism. Frank Cass. 1978. Landais, E. 1983. "Analyse des Systemes d'Elevage Bovin Sedentaire du Nord de la Cote d'Ivoire." IEMVT. 1983. 1984. "Elements pour la Preparation du Programme de Recherche sur les Systemes de Production et le Transfert de Technologies en Milieu Rural. Haute Casamance et Senegal Oriental."ISRA.Aout 1984 Lawry, S.W., Riddell, J.C. and Bennett, J.W. 1984."Land Tenure Policy in African Livestock Development." in Simpson and Evangelou, ed., Livestock Development in Subsaharan Africa. 1984. Levine, J.M. 1982."Modeling Livestock Sytems under Tropical Conditions with Special Reference to Indonesia." in Fine and Lattimore, ed., Livestock in Asia. 1982. Little, P.D. 1983."The Livestock-Grain Connection in Northern Kenya:an Analysis of Pastoral economics and Semi- Arid Land Developmewnt." Rural Africana. 1983. Livingstone, I. 1977. "Economic Irrationality Among Pastoral Peoples: Myth or Reality?" Development and Change. No. 8. 1977. 209-230. Low, A. 1980."The Estimation and Interpretation of Pastoralists' Price Responsiveness." ODI Pastoral Network. Paper 103. 1980. Manetsch, T.J and A1. 1971. A Generalized Simulation Approach to Agricultural Sector Analysis. East Lansing: Michigan State University. 1971. Maxwell, T.J. 1980."Economic Techniques for the Assessment of Sheep Systems." Hill Farming Research Organization. Bush Estate,Penicuik, Midlothian. 1980. McDowell, R.E. 1984."Livestock Nutrition in Sub-Saharan Africa: an overview." in Simpson and Evangelou, ed., Livestock Qevelppment in Subsaharan Africa. . and Hildebrand, P.E. 1980.1ntegrated Crop and Animal Production. The Rockefeller Foundation. 1980. 171 Mellor, J.W. 1969. "Production Economics and the Modernization of the Traditional Agricultures." Austr. J. Agr. Econ. June 1969. Milligan, K. and De Leeuw, P.N. 1983."Low Altitude Aerial Surveys in Pastoral Systems Research." in ILCA. Pastoral_ Systems Research in Sub-Saharan Africa.August 1983. Naylor, T.H. and Vernon, J.M. 1979. Microeconomics and Decision Models of the Firm. Harcourt, Brace and World, Inc. 1979. N'Diaye, A.L. et Ba, C. 1972. "Elevage et Cooperation en Afrique Tropicale. L'example du Senegal." Rev. Elev. Med. Vet. Pays Trop, Vol. 25. No 3. 433-443. Ngategize,P.K. 1982. "Herd Structure, Production Traits, Cattle Transactions and Economic Analysis of Alternative Production systems of Zebu Cattle in N. Tanzania." Plan B Paper. Michigan State University. 1982. Panayotou, T. and Tokrisna, R. 1982. "Microeconomics of Rural Livestock. The Case of Buffalo and Cattle in Thailand" in Fine and Lattimore, ed., Livestock in A513, 1982. Penning de Vries, F.W.T. et Djiteye, M.A, ed., 1982. La Productivite des Paturages Saheliens.Agricultural Research Reports 918. Centre for Agricultural Publishing and Documentation. Wageningen, 1982. Perrin, R.K. 1972. "Asset Replacement Principles." Amer. J. Agr.Econ. Vol 54. 1972. 60-67. Porter, R.C. 1979. "The Uses of Economic Models in Analysis of the Cattle Sector." CRED. U of Michigan. 1979. Powers, T.A. 1975. "Herdsism Simulation Model." Inter American Development Bank. Papers on project Analysis 2. February 1975. Pratt, D.J. 1984. "Ecology and Livestock." in Simpson and Evangelou, ed.,;ivestock Qevelgpment in Sub- saharan Africa. 1984. Raimbault, A.M. 1980. "Les Maladies Nutrionnelles." La Recherche. No. 115. 1980. 1096-1104. Rucker, R.R; Burt, 0.R. and LaFrance, J.T. "An Econometric Model of Cattle Inventories." Amer J. Agg. Econ. Vol. 66. No.2. 1984. 172 Runge, C.F. 1981. "Common Property Externalities: Isolation, Assurance and Resource Depletion in a Traditional Grazing Context." Amer. J. Agr. Econ. Vol. 63. No. 4. 1981. Ruttan, V.W. and Hayami Y. 1984. "Induced Innovation Model of Agricultural Development." in Eicher and Staatz, ed., Agricultural Development in the Third World. 1984. Safilios-Rothschild C. 1983. "Women in sheep and Goat Production and Marketing." FAO. Expert Consultation on Women in Food Production.1983. Sandford, S. 1981. "Organizing government's Role in the Pastoral Sector." in Galaty et al., ed., The Future of Pastorg; Peoples. 1981. Schneider,H.K. 1974.Economic Man. The Free Press. 1974. 1979._Aivestock and Eguality in East Africa. Indiana Unversity Press. 1979. 1981. "Livestock as Food and Money." in Galaty and al. ,ed., The Future of Pastoral Peoples.1981. 1984. "Livestock in African Culture and Society: a Historical Perspective." in Simpson and Evangelou, ed., Livestock Development:;p_ Subsaharan Africa. 1984. SEDES. 1975.Recueil Statistigpe de la Production Animale. Editions du Ministere de la Cooperation. Republique francaise. 1975. Shaffer, J.D. 1983. "Influencing the Design of Marketing Systems to Promote Development in Third World Countries." Paper presented at the International Workshop on Agricultural Markets in the Semi-Arid Tropics, sponsored by the International Crops Research Institute for the Semi-Arid Tropics, Hyderabad, India. October 24, 1983. Shapiro,K.H. 1976. "Efficiency Differentials in Peasant Agriculture and their Implications for Development Policies." CRED. U.of Michigan. Discussion Paper No.52. June 1976. 173 1979. "The Livestock Economies of West Africa: An Overview." in Shapiro, K. H., ed.,Livestock Production and Marketing in the Entente States of West Africa: A Summary Report. CRED. U of Michigan. Simpson, J.R and Evangelou, P., ed., 1984 Livestock Development in Subsaharan Africa. Boulder, Colorado: Westview Press, Inc. 1984. and Farris, R. 1982. The World's Beef Business. Iowa State Press. 1982. SO.D.E.S.P. 1984. " Des Prix du Betail et de la Viande". (1984-1985). Note Technique No. 5. (5e Edition). Societe Africaine d'Edition. 1983. Le Senegal en Chiffres. Edition 1982-83. Sollod, A.E., Wolfgang, K. and Knight, J.A. 1984. "Veterinary Anthropology : Interdisciplinary Methods in Pastoral Systems Research." in Simpson and Evangelou, ed., Livestock Development in Subsaharan Africa. SOMIVAC. 1982. "Commercialisation des Bovins et de la Viande Bovine de la Casamance". D.E.E.P. SOMIVAC. 1982. Starkey, P.H. 1984. "Ndama Cattle. A Productive Trypanotolerant Breed." World Animal Review. No 50. 1984. Stryker, J.D. 1984. "Land Use Development in the Pastoral Zone of West Africa." in Simpson and Evangelou, ed.,Livestock Development in Subsaharan Africa. Sullivan, G.M. (undat.) "A Socioeconomic Model for Analyzing Economic Risk and Social Reciprocity in Sedentary Livestock Systems in Tanzania." Auburn University Mimeogr. Undat. .1984. "Impact of Government Policies on the Performance of the Livestock-Meat Subsector." in Simpson and Evangelou, ed., Livestock and Levelgpment Chgpge in Sub-Saharan Africa, Swift, J. 1979. "West African Pastoral Production Systems." Working Paper No. 3. Livestock Production and Marketing in the Entente States of West Africa CRED. U of Michigan. 1979. Teele, T.T. 1984. "Development and Management of Livestock Projects in the Sahel Area of Africa." in Simpson and Evangelou, ed.,Livestock pevelopment in Sub- Saharan Africa. 174 Teitelbaum, J.M. 1975. "Human Versus Animal Nutrition: a Development Project Among Fulani Cattle Keepers of the Sahel of Senegal. Mimeogr. December 1975. Timmer, P.C., Falcon, W.P and Pearson, S.R. 1983. Food Policy Analysis. Johns Hopkins University Press. June 1983. Trail, J.C.M, Murray, M. and Wissocq Y. 1984. "The Trypano- tolerant Livestock Network in West and Central Africa." ILCApgulletin. No 18. 1984. Tuck, L. 1983. "Formal and Informal Markets in Rural Senegal." Princeton University. October 1983. Ware, T.A. 1979. "Wolof Farmers and Fulani Herders : a Case Study of Drought Adaptation in the Diourbel Region of Senegal." PH. D. dissertation. U. of Michigan. 1979. Widstrand, C.G. 1975. "The Rationale of Nomad Economy." Ambio Vol. 4. No. 4. 1975.