A Participatory Rapid Appraisal (PRA) of Farming Systems in Western Kenya

Transcription

1 SDP Collaborative Research Report A Participatory Rapid Appraisal (PRA) of Farming Systems in Western Kenya Michael Waithaka, Angela Wokabi, Julius Nyangaga, Emily Ouma, Tineke de Wolff, Joseph Biwott, Steven Staal, Meshak Ojowi, Richard Ogidi, Ismael Njarro and Patrick Mudavadi Report of a PRA of Dairy and Crop Activities in Western Kenya September 2000 Smallholder Dairy (R&D) Project PO Box Nairobi, Kenya Ministry of Agriculture and Rural Development Kenya Agricultural Research Institute International Livestock Research Institute Department for International Development-UK (DFID) This Collaborative Research Report is circulated prior to full peer review to stimulate discussion and comment. Based on that process, its content may be revised i

2 Acknowledgements Gratitude is expressed to all farmers who participated and contributed to this study. Thanks go the Ministry of Agriculture and Rural Development, KARI and ILRI staff who participated or assisted in carrying out this study. We will always remember Dr. Richard Ogidi (RRC Kisii) who passed away two months after the studies were carried out. He was a considerate, sensitive and dedicated team player. May God rest his soul in peace. This publication is an output from a research project funded by the United Kingdom Department for International Development (DFID) for the benefit of developing countries. The views expressed here are not necessarily those of DFID. ii

3 Table of contents TABLE OF CONTENTS... I LIST OF FIGURES... II LIST OF TABLES... II LIST OF ACRONYMS... III EXECUTIVE SUMMARY... I 1 INTRODUCTION OBJECTIVES METHODOLOGY Site selection The participants Factors appraised Tools PRODUCTION SYSTEMS CURRENT PRODUCTION SYSTEMS CROP PRODUCTION LIVESTOCK PRODUCTION Livestock feeding calendar Livestock breeding Livestock diseases LABOUR PROFILE FARM AND COMMUNITY RESOURCE MAPS CONSTRAINTS TO CROP AND LIVESTOCK PRODUCTION MILK MARKETING FORMAL MILK MARKETING INFORMAL MILK MARKETING Milk bars Mobile milk traders Hawkers Milk wholesalers/distributors KENYA DAIRY BOARD SERVICE PROVISION VETERINARIANS, ANIMAL HEALTH ASSISTANTS AND STOCKISTS AI SERVICES CONCLUSIONS REFERENCES ANNEX 1: MAPPING OF SPATIAL VARIATION IN WESTERN KENYA Cattle densities in Western and Nyanza Tick borne disease distribution i

4 3. Spatial distribution of ethnic groups ANNEX 2 FARM AND COMMUNITY LEVEL RESOURCE MAPS ANNEX 3 MILK FLOWS TO TOWNS ANNEX 4. TYPICAL FARM LAYOUTS - CURRENT AND PAST List of Figures Figure 1 Clustering of similar sub-locations in western and Nyanza provinces... 5 Figure 2 Purchase and sales prices (KSh per litre) by site Figure 3 Grade and total cattle densities Figure 4 Distribution of ticks and reported tick-borne disease cases Figure 5 Distribution of ethnic groups in western province Figure 6 Distribution of ethnic groups in Nyanza province Figure 7 Farm and community resource maps for Kakelo Dudi, Rachuonyo district Figure 8 Farm and community resource maps for Kojwang /Katunde, Rachuonyo district Figure 9 Farm and community resource maps for Bokiamburi, Nyamira district Figure 10 Farm and community resource maps for Ekerenyo, Nyamira district Figure 11 Farm and community resource maps for Marani, Kisii district Figure 12 Farm and community resource maps for Suneka, Kisii district Figure 13 Community resource map for Mbihi, Vihiga District Figure 14 Community resource map for Kilibwoni, Nandi District Figure 15 Milk flow and prices to Nyamira town and Oyugis town Figure 16 Milk flow and prices in Bungoma district Figure 17 Typical current farm layout in Suneka, Kisii district Figure 18 Typical farm layout in 1980s in Suneka, Kisii district Figure 19 Typical current farm layout in Bokiamburi, Nyamira district Figure 20 Typical farm layout in 1980s in Bokiamburi, Nyamira district Figure 21 Typical current farm layout in Kojwang/ Katunde, Rachuonyo district Figure 22 Typical farm layout in 1980s in Kojwang/Katunde, Rachuonyo district List of Tables Table 1 Cluster means... 4 Table 2 Cluster description... 4 Table 3 Historical profile of the livestock systems... 9 Table 4 Historical profile of crop farming Table 5 Characteristics of the production system for a typical household Table 6 Ranking of constraints to crop and livestock production in declining order of importance Table 7 Cooperatives, year of establishment and current milk intake Table 8 Ranking of constraints cited by milk traders/farmers Table 9 Prices of livestock inputs ii

5 List of Acronyms ABS ADC AFC CAIS DFCS ECF FMD GIS GMR HPI ICRAF ILRI JSC KARI KCC KDB LDP MOARD NDDP PPE PRA SDP SSC TBD American Breeders Service Agricultural development Corporation Agricultural Finance Corporation Central artificial insemination Station dairy farmers cooperative societies East Coast Fever foot and mouth disease Geographical Information Systems Guaranteed Minimum Returns Heifer Project International International Centre for Research in Agroforestry International Livestock Research Institute juice (chewing) sugar cane Kenya Agricultural Research Institute Kenya Co-operative Creameries Kenya Dairy Board Livestock development Project Ministry of Agriculture and Rural Development National Dairy Development Programme precipitation over potential evapo-transpiration Participatory Rural Appraisal Smallholder Dairy Research and Development Project sugar, sugar cane tick-borne diseases iii

6 Executive summary Introduction In the second phase of the Smallholder Dairy Research and Development Project (SDP), as well as continuing work in central Kenya, attention will be given to Western Kenya where lessons learnt from the Central and Coast regions of Kenya will be applied. Western Kenya shares a number of features that present an opportunity for smallholder dairy research and development. The climate is favourable for dairy production and average farm sizes are declining rapidly due to increasing population pressure. This study follows a sequential process Geographic Information Systems (GIS) analysis to identify potential sites, participatory rapid appraisals (PRAs) in the selected sites and characterization surveys. Each step informs the design and analysis of next study, building each time a better knowledge and understanding of smallholder agriculture and dairy systems and the constraints and refining the recommendation domains for the pilot interventions to be selected with farmers, market agents, regulators and policy makers. Objectives The PRAs were organised in the districts at the sub-location level by the project management together with the farmers and various stakeholders to: Appraise the history and development of agricultural production systems Identify stakeholders and their interests Learn various farming activities from spatial and temporal points of view Identify local needs and constraints to the development of dairy farming Identify milk market outlets, flow patterns and constraints faced in this area Identify service providers and the constraints faced in their trade Methodology The PRA studies in Western Kenya were designed to gain information on broad agricultural activities. PRA sites were selected based on features described by the spatial mapping of factors crucial to dairy farming. The main factors were spread of people, cattle, towns and roads. Factors that describe natural dairy potential: rainfall and humidity, altitude, soils and disease risk were also used. To cover as much of the variation in a district, one sub-location was picked from each of the two most dominant clusters. The research team included staff from ILRI, KARI and the Ministry of Agriculture and Rural Development. At each site the PRAs were held with groups of farmers of different ages and gender. Farmer turnout ranged from 20 to 130 (on average 50). A variety of PRA tools were i

7 used in a flexible and sequential manner, cross-checking at each stage and improving on reliability and participation of the farmers. Evolution of production systems The earliest and most distinctive development was adjudication and demarcation of land from communal ownership to individual holdings in the 1960s and 70s resulting in smaller land sizes per household that could not allow the keeping of large numbers of Zebu cattle. With title deeds, credit became accessible and was mostly used for the establishment of the new high yielding maize hybrids, which were also encouraged by effective marketing through the National Cereals Board. During this same period, government supported services were strengthened through the provision of agricultural field extension workers, establishment of veterinary clinics, and construction of communal dips. Coffee was established as a cash crop in the 1940s, tea from the 1960s and 80s and sugar in the late 70s and only to the Northern districts. Dairy cattle were introduced only in the early 70s through the national AI and purchase of cows from large dairy farms. In many places Kenya Co-operative creameries (KCC) had established dairy co-operatives in the1960s and milk was marketed through these societies well through the 70s. The 1980s were the beginning of a period of poor fortunes for both crops and livestock. Poor quality seeds, unreliable markets, high production costs and further decline in land sizes reduced crop production. In the 90s the government withdrew support from essential services and initiated their provision through private enterprise. Withdrawal of veterinary services impacted negatively on the growth of grade cattle numbers. Current production systems Despite the conspicuous presence of Zebu cattle, crop husbandry dominates agricultural activities in all areas studied except Nandi where grade cattle are more prominent. A typical household has about three to five Zebu cattle and very few cross or grade cattle. The low number of grade animals kept per household and their low productivity is a result of low feed supply, and poor husbandry methods. The prevalence of the Zebus is associated with the cultural practices (dowry payment and prestige) where the number of cattle per household is more valuable than the quantity of milk produced. Tick borne diseases, especially East Coast Fever, are endemic throughout the area. Prevalence of foot and mouth disease (FMD) and lumpy skin disease is associated with uncontrolled movement of livestock into and out of the districts and the presence of Trypanosomosis to wildlife in the forests. ii

8 There are some indications of intensification in all areas with livestock playing an important role as a source of food, income and manure. Most areas produce their own food with much diversification. All areas indicate high market activities with sale of food and cash crops, milk and also for sourcing inputs such as fertilizer and food. Communities have close relationships between farmers with the ministry of agriculture, churches, self-help groups, tea and coffee factories, and the veterinary department. The communities did not have close relationships with stockists, KARI and the Agricultural Finance Corporation. Constraints to crop and livestock production The constraints that were mentioned by most of the 14 farmers groups are: animal diseases, extension services, animal breeding, roads, crop diseases, cost of inputs, water, land size, capital, marketing, quality of inputs, output prices, soil fertility, human diseases, veterinary services, credit, management of dips and livestock feeds. Milk marketing Most marketing is through the informal sector to nearest urban centre. A few farmers cooperative societies are still operational though beset by management problems and competition from milk hawkers. Constraints to milk marketing include seasonal fluctuations in supply, high cost of capital items, lack of organised marketing, high transportation costs, lack of storage facilities and knowledge on milk handling. Conclusions Over time, farmers have been responsive to incentives as shown by rapid increases (decrease) in production of crops and milk when conditions are conducive (not conducive). Most areas in the region have the potential for intensive livestock production due to favourable climatic conditions. Intensification of dairy production would increase the household labour demand and offer alternative sources of income. Western and Nyanza are generally deficient in milk and flows tend towards the major towns. There are opportunities for growth in milk marketing if there s improvement in the organization of traders and encouragement by regulatory bodies to trade in larger volumes under hygienic conditions. For SDP, the issues that will have to be addressed will revolve around the huge milk deficit while there is still potential to increase milk production. Immediate issues would be to iii

9 improve our understanding on why the production gap exists and why farmers do not adopt technologies that would increase milk output. iv

10 1 INTRODUCTION In the second phase of the Smallholder Dairy Research and Development Project (SDP), as well as continuing work in central Kenya, attention will be given to Western Kenya where lessons learnt from the Central and Coast regions of Kenya will be applied. Western Kenya shares a number of features that present an opportunity for smallholder dairy research and development using results of studies that have been done in the other regions. The climate is favourable for dairy production and average farm sizes are declining rapidly due to increasing population pressure. The sequential process to be followed is the national rapid appraisal with its broad description and diagnosis of western Kenya. This will include subsequent sub-regional reviews for each of the mandate areas under Regional Research Centers (RRC) of the Kenya Agricultural Research Institute (KARI), Geographic Information Systems (GIS) analysis to identify potential sites, participatory rapid appraisals (PRAs) in the selected sites and finally the characterization surveys. Each step informs the design and analysis of next study, building each time a better knowledge and understanding of smallholder agriculture and dairy systems and the constraints to, and opportunities for, their improvement, and refining the recommendation domains for the pilot interventions to be selected with our clients: the producers, the market agents, the regulators and the policy makers. These studies of a new area attempt to get a clear picture of the prevailing production systems and in particular the dairy industry and how they have evolved over time. Milk marketing structure is also appraised since it has been learnt from the previous studies that the development of commercial small-scale dairy industry is a function of milk demand and the product delivery systems. Moreover, the very recent but fast changes in milk marketing as a consequence of a liberalised economy have created opportunities for growth in dairy production and milk outlets that have not been adequately studied in these parts of the country. The initial diagnostic surveys of Western Kenya are expected to provide an avenue through which the current status of the dairy industry can be observed and provide a guide for project entry and implementation. These surveys include spatial analysis of secondary data to target site selection, rapid appraisals and farm characterisation studies that will form a major part of the first months of the project's second phase. 1

11 1.1 Objectives The PRAs were organised in the districts at the sub-locational level by the project management together with the farmers and various stakeholders to: Appraise the history and development of agricultural production systems (and in particularly the dairy industry) to the state in which they are today Identify stakeholders and their interests Learn various farming activities from spatial and temporal points of view Identify local needs and constraints to the development of dairy farming Identify the various milk market outlets and flow patterns while studying constraints faced in providing this essential link between producers and consumers Identify various other service providers and learn the constraints faced in their product delivery 1.2 Methodology The PRA studies in Western Kenya were designed to gain information on broad agricultural activities and dairy farming set-up. A team of researchers and extension workers met with farmers to discuss the evolution of agricultural production systems, movement of resources, services and general constraints. PRA sites were selected based on features described by the spatial mapping of factors crucial to dairy farming. Gaining preliminary insight into both Western and Nyanza Provinces was initiated by secondary data analysis, which served two goals. In the first place, mapping the spread of people, cattle, towns and roads (Annex 1) provides a very general overview of the kind of areas the study will be dealing with. It also provides an initial attempt at targeting research areas with certain characteristics. To identify specific and promising cattle or dairy areas, the relationship between cattle and spatial indicators were used to map differentiation in dairy potential to focus site selection Site selection Earlier studies show that a number of factors strongly influence the production and marketing of milk by smallholders. Apart from the individual household characteristics these are human population growth and densities, climate and rainfall and access to urban centers and services (Kiambu Pilot Survey 1997). Population growth and densities retain a dual relationship with dairy, presenting a market and price incentive for intensified production when numbers are high. At the same time, however, pressure on land and resources may leave farmers with little other choice but to intensify, if they have the means. Closely related to population density is dairy market access. Both the Kiambu study and the Other District Survey (1998) show that more intensified systems are 2

12 mainly found in highly populated areas and close to urban centers, which provide market outlets and good milk prices that act as an incentive to produce. Since the timing of milk delivery and collection is critical, particularly in a smallholder African setting where cooling systems are rarely available, distance to markets and available infrastructure are of prime importance to smallholder farmers. Of course, infrastructure and distances do not only influence market access, they also affect the availability of veterinary and artificial and insemination services. Hence, general accessibility is an important factor where dairy is concerned. Apart from factors related to market access and infrastructure, factors that describe natural dairy potential are rainfall and humidity, altitude, soils and disease risk. Given the use of crossbred animals and associated susceptibility of imported breeds, disease challenge plays an important role in farmer choice of production systems. Tick-borne diseases present a serious threat to the whole dairy system as a number of them cause heavy mortality losses (Staal et.al. 1999). However, due to data limitations, only those factors for which GIS coverages could be made available within a reasonable time span were used in the initial stratification procedure. These turned out to be mean household density and climatic potential per sub location and mean access to the nearest urban center: o Population data were derived from the 1989 census. Since the survey will focus on households and their practices, household densities have been given preference over population densities as an input for patterns of spatial differentiation. o Annual PPE (precipitation over potential evapo-transpiration) proved to be a useful indicator for climate related factors. It combines elevation, rainfall and temperature data into one measure of overall humidity. A PPE value of 1 indicates that the amount of rain falling is similar to the amount lost through evapo-transpiration (crop production can only be carried out when PPE is greater than 0.5) o The layer for access to urban centers was created by ICRAF and estimates the travel time to the nearest urban center (with population density greater than 2,500 persons per square kilometre) in hours. A major setback was that this layer for access to urban centers proved to be insignificant in almost all analyses tried. The layer is quite crude and does not take into account that different areas are serviced by different road types (and thus show tremendous variations in travel 3

13 time). However it proved to be quite useful for a general insight into distance to markets and spatial spread of densely populated areas. Mapping spatial differentiation To facilitate sampling in the seven selected districts (Bungoma, Kakamega, Vihiga, Nandi, Rachuonyo, Kisii and Nyamira) the challenge was to come up with groups of sub-locations that ranked similarly on combinations of these dairy related factors. According to Carter 1996, there are two ways in which to create these spatial stratifications. In the first, unique combinations of specific characteristics or value ranges are identified by mapping each factor involved and physically overlaying each map. In the second, statistical models of the relations between variables are constructed which might involve simple correlations, principle component analysis and cluster analysis (Carter 1996). For this characterisation effort we chose cluster analysis. With only three significant data layers available it seemed the most logical thing to do, since both principle component analysis and simple correlations seek to reduce the initial number of variables to serve as input for clustering. SAS software generated six clusters which portray quite some difference between the various sub-locations. The clusters all range somewhere between the highly urbanized and the remote and sparsely populated (Table 1 and Table 2) Based on these clusters it is possible to voice expectations on the presence and level of dairying, e.g., in sub-locations with high market access, high population densities and resulting low land holdings and a high PPE, chances are considerable that dairy will be present: farmers will be either lured or forced into intensification under these circumstances. Table 1 Cluster means Cluster number. Mean household density Mean access Mean PPE 1 (x1) 0 (low) 5 hrs (poor) 1.07 (high) 2 (x15) 497 (high) 0.2 hrs (high) 0.91 (reasonable) 3 (x155) 53 (low) 2 hrs (poor) 0.95 (reasonable) 4 (x106) 174 (medium) 0.6 hrs (good) 1.07 (high) 5 (x238) 61 (low) 0.8 hrs (reasonable) 0.85 (lower) 6 (x154) 75 (low) 1 hr (reasonable) 1.15 (high) Table 2 Cluster description Cluster Description 1 Mt Elgon 2 Highly urbanised sub-locations (municipalities) 3 Remote, sparsely populated areas 4 High potential areas (not too many households, good access and climatic potential) 5 Lower density areas with good access but lower climatic potential 6 Similar to 5 but with very high climatic potential 4

14 To cover as much of the variation in a districts as possible, one sublocation was picked from each of the two most dominant clusters. Thus the PRAs were carried out in Navakholo and Central Kabras in Kakamega, Kilibwoni and Mutwot in Nandi, Nangili and Temberera in Bungoma, Mbihi and Gisambai in Vihiga, Kakelo Dudi and Kojwang Katunde in Rachuonyo, Marani and Suneka in Kisii and Bokiamburi and Ekerenyo in Nyamira. Figure 1 Clustering of similar sub-locations in western and Nyanza provinces The participants The research team included staff from ILRI, KARI and the Ministry of Agriculture. At each site the PRAs were held with farmers from within the area. The farmers were mixed in age, gender even farmers who did not keep livestock. For each site there were 6 to 5 researchers and the farmer turnout ranged from 20 to 130, but on average 50. The farmers gathered at each pre-arranged location from around 10 am and the discussions went on till around 5 pm. 5

15 The market PRAs were conducted alongside the farmer PRA. In many places it was not very possible to have the milk marketers gather as a group for an effective PRA and the research team conducted semi-structured interviews. This method was also used for the service providers (veterinary, feed stockists, extension personnel from the Ministry, AI service providers both private and Government employees). Although the objectives of the study were covered as arranged it was felt that the time allocated was not adequate for an in-depth PRA using all the appropriate tools Factors appraised Diverse PRA tools were to use to gain more insight in: 1. Production systems 2. Condition and circumstances 3. Prioritisation of constraints and opportunities from activities 1 and 2 4. Milk marketing 5. Service provision For a start, the research team split into two groups and carried out trial PRAs in the two sites in Kakamega districts and compared notes on the outcomes. Later, one group handled Vihiga, Bungoma and Nandi districts and the other Kisii, Nyamira and Rachuonyo districts. The two groups met at the end of each week to review progress and prepare a report on information gathered Tools The PRAs sought to actively involve farmers in appraisal, analysis and prioritisation of local issues. This was done through listening, capturing local knowledge and encouraging partnership. The interviews with farmers were conducted in groups. A variety of different PRA tools were used in a flexible and sequential manner crosschecking with different methods at each stage to improve on reliability and to involve more people in the planning process. The actual sequence of methods used in each PRA site broadly followed a similar pattern in each site: Time line and trend analysis, to visualise the historical changes in the area: farmers drew on a manila sheet the typical farm in their areas in the 1970s, 80s, 90s and currently. Emphasis was placed in identifying land size, crops grown and their relative areas as well as livestock types and their numbers (Annex 4, Table 3 and Table 4). Resource and social map, to identify resources, occupations and means of livelihood: here farmers depicted on manila sheets the flow of inputs into the farm and outputs from the 6

16 farm. They would also show movement within the farm between crops and livestock (Annex 2). Seasonal analysis, focussing on temporal information pertaining to cropping, land and labour use: using manila paper, farmers would draw the months of a year at the bottom and then indicate with lines or symbols the frequency of occurrence of events for a given crop, e.g., planting, weeding to harvesting. After all crops and livestock were covered a labour profile was drawn depicting monthly requirements through the year (Table 5 and Annex 3). Scoring and ranking of constraints and opportunities: based on a list of say constraints that the farmers would give, they were asked to compare two at a time and state which was more severe compared to the other. The number of times a constraint would be repeated formed the scored and these were ranked to show order of importance (Table 6 and Table 8). Visualisation was used to encourage active participation of farmers in the group interviews. 7

17 2 Production Systems The farmers gave an account of various events that have shaped the current scenarios in the areas (Table 3 and Table 4). The earliest most notable event that has impacted greatly on agricultural systems in this area is the adjudication and demarcation of land from communal ownership to individual holdings with title. This took place in the 60s in most places though it was reported as early as 1957 in Mutwot and as late as 1970 in Kilibwoni, both in Nandi district. This resulted in smaller land sizes per household that could not allow the keeping of large numbers of the local Zebu cattle reported as high as 100 per household in the 60s in Kakamega (west Bunyala), though the average appears to have been 30 to 40, especially for the Southern districts. During this same period, government supported services were strengthened through the provision of agricultural field extension workers, establishment of Veterinary clinics, and construction of communal dips. Credit facilities such as guaranteed minimum returns (GMR) loans were made available through Agricultural Finance Corporation (AFC). These were made possible by the use of land title deeds as collateral. It seems that the credit was mostly used in the establishment of crops, especially the newly introduced high yielding maize hybrids which was encouraged by effective marketing through the National Cereals Board. The high grain yield, taste preference and ease with which maize could be produced and processed resulted in diminishing attention in the production of indigenous sorghum and finger millet especially in the 70s and 80s. However, in Rachuonyo, better hybrids of sorghum (Serena and Seredo) were introduced in 1982 and their production has been on the increase. Although coffee as a cash crop appears to have been established much earlier than 1960 (in Marani, farmers mention the existence of their co-operative society in 1940), tea was only introduced in Nandi (Kilibwoni) in 1961 and in Rachuonyo as late as Sugar, sugarcane (SSC) only came in the late 1970s and only to the Northern districts of Kakamega and Bungoma. These crops competed for land other crops and with livestock putting pressure on large local zebu cattle sheep and goat populations and indirectly encouraging the build up of grade cows. Dairy cattle were introduced only in the early 70s through the national AI and purchase of cows from larger Agricultural development Corporation (ADC) dairy farms in Kitale. Farmers in Kakamega (west Bunyala) and Rachuonyo obtained Napier grass from Kakamega research station at about the same time. The availability of the Government supported services for livestock helped sustain a relatively high grade cattle population during the 1970s (an average three to four per household then as compared to one to two now). However, grade cattle population in Kakamega and Rachuonyo, suddenly dwindled due to what farmers described as diseases outbreaks and inability to feed them. In a few places Kenya Cooperative creameries (KCC) had established dairy co-operatives in the1960s 8

18 and milk was marketed through these societies well through the 70s. However, administrative problems associated with non-payment of loans saw their collapse at the end of this period. The 1980s were the beginning of a period of poor fortunes for both crops and livestock. In Nandi, armyworm infestations and maize smut seriously affected crop yields. Growing ineffectiveness of government services resulted in dairy cattle populations dropping to almost nil in Rachuonyo. Maize production was also affected by poor quality seeds, unreliable market (cereals board was not purchasing farmers' produce), high production costs and further decline in land sizes. In the 90s the government withdrew support from essential services and initiated their provision through private enterprise. Dips were left to farmer committees, Veterinary and AI services were left in the hands of private providers. The floundering efficiency of this new approach has negatively affected agricultural production and farmers have not yet readjusted to these developments. These changes and the continuing subdivision of farmland have impacted negatively on agricultural. Table 3 Historical profile of the livestock systems PERIOD Event; activity 1960s Land adjudication: Nandi (Mutwot,Kilibwoni), Bungoma, Kakamega (west Bunyala, central Kabras) Sahiwals and crosses introduced: Nandi (Kilibwoni) Grade cattle introduced and increased in number till the 80s: Nandi (Kilibwoni) AI started: Kakamega (central Kabras) GMR loans provided for dairy production: Kakamega (central Kabras) Government supported services availed: meat inspection, hides and skins bandas introduced. Dips started: Vihiga (Gisambai) Kakamega and Malava dairy Coops started: Kakamega (central Kabras) Zero-grazing started: Kakamega (central Kabras) 1970s Friesian cattle obtained from Kitale ADC: Kakamega (west Bunyala) Napier grass from Kakamega KARI: Kakamega (west Bunyala) Co-operatives increase: Vihiga (Gisambai) AI started: Nandi (Kilibwoni) Grade cattle introduced but many died from TBDs: Rachuonyo (west Karachuonyo) Dairy production started to drop to the lowest level it is today: Rachuonyo (west Karachuonyo) Kakamega and Malava dairy co-operatives collapsed from unpaid loans Grade cattle introduced in Kakamega (central Kabras) and Bungoma Introduction of Ayrshire bulls after construction of cattle dips: Nandi (Kilibwoni) GMR loans stopped: Nandi (Mutwot) 1980s Grade cattle introduced but many died from diseases. Number of grade cattle increased due to availability of government supported AI services and organised milk marketing through KCC: Nyamira (Nyamira) Tea introduced: Rachuonyo (W/Karachuonyo) ECF outbreak affects cattle populations: Kakamega (west Bunyala) Dairy production encouraged through Finland supported Rural Dairy Project: Rachuonyo (west Karachuonyo) AI started: Bungoma Seasonal credit introduced: Nandi (Mutwot) 9

19 1990s Populations of local cattle declined because of their low milk production, low milk prices and small land sizes: Kakamega (central Kabras) Government AI and veterinary services declined: Rachuonyo (west Karachuonyo) Dips handed over to farmer management committees: Kakamega (west Bunyala) Government withdraws from provision of veterinary, AI and extension services. Dips cease to operate: Kakamega (central Kabras) Table 4 Historical profile of crop farming PERIOD Event; activity 1960s Land adjudication: Nandi (Mutwot), Bungoma, Kakamega (west Bunyala, central Kabras) Tea planting started: Nandi (Kilibwoni) Coffee 'forced' on farmers, production increases: Rachuonyo (west Karachuonyo) Sorghum varieties grown: local brown ('Andino) and red (Ochuti); production high: Rachuonyo (west Karachuonyo) Credit facilities availed: Nandi (Kilibwoni) Agricultural extension established, readily available: Nandi (Mutwot) Maize production increased with introduction of better hybrids: Nandi (Kilibwoni) Co-operatives increased: Vihiga (Gisambai) 1970s Land adjudication: Nandi (Kilibwoni), Rachuonyo (west Karachuonyo) Improved sugarcane (SSC) introduced: Kakamega (west Bunyala) Tea introduced: Kakamega (central Kabras) GMR loans availed, boosted maize production: Kakamega (central Kabras), Nandi (Mutwot) GMR loans stopped in 1978: Nandi (Mutwot) Sugarcane planting, overproduction and marketing problems: Kakamega (central Kabras) 1980s West Sugar Company starts buying cane: Kakamega (central Kabras) Seasonal credit introduced: Nandi (Mutwot) Army worm infestation and maize smut: Kakamega (west Bunyala) Sorghum hybrid varieties introduced: Rachuonyo (west Karachuonyo) Number of co-operatives decreased: Vihiga (Gisambai) Slump in coffee prices, crop uprooted and replaced by maize and bananas: Rachuonyo (west Karachuonyo) 1990s Cassava and maize diseases increase: Kakamega (west Bunyala) 2.1 Current production systems Despite the obvious presence of indigenous livestock, crop husbandry dominates agricultural activities in all areas studied except Nandi district where dairy is relatively more prominent. Table 5 shows the basic comparisons across the districts studied. Land size per household is low ranging from one to four acres in Kisii, Nyamira, Vihiga and Kakamega. Households in Bungoma, Rachuonyo and Nandi have larger acreage (over 10 acres). Leasing of land is common, especially in Kakamega where it is rented at Kshs 1,000 per acre per year. Typical farms practice mixed farming, combining cash and food crops with livestock production. The main cash crops are coffee, tea, sugar cane for sugar production (SSC) and juice sugar cane for chewing (JSC), and pineapples. The main food crops are maize, beans and sweet potatoes. Others food crops include bananas, cassava, finger - millet, sorghum, groundnuts, vegetables and arrowroots. Surplus food crops are also sold for cash. 10

20 The main livestock types are cattle (pure breeds or grade, local Zebu and crosses), poultry (exotic and local chicken), sheep, goats and donkeys. The livestock are fed on natural pastures, Napier grass, banana stems, sweet potato vines, vegetable waste, sugar cane leaves and tops, maize stover and shrubs. Cattle are mainly grazed though there is increasing confinement by tethering, zero-grazing or semi-grazing systems. Use of concentrates was not high and was said to be common in the dry season when pastures and fodders are in short supply. 2.2 Crop production In Kakamega, Bungoma, Vihiga and Rachuonyo, maize, beans, bananas, and sorghum are the main food crops while sugar sugar cane (SSC) and coffee are the main cash crops. Others are cassava, bambara nuts, ground nuts, simsim, though their production has been on the decline over the years. In Kisii and Nyamira and Vihiga maize, beans, bananas, millet and sorghum are the main food crops while coffee and tea are the main cash crops. In all the places beans are intercropped in maize. On a very low scale in the regions there are sweet potatoes, pumpkin, indigenous vegetables and arrowroots. Juice sugar cane (JSC) is grown in Kisii and Nyamira as a food and as a cash crop. Wheat was mentioned in Nyamira, but on small plots. However, it appears to be a very recent addition in the farms and is gaining popularity. Sorghum and finger millet in the region is gradually being replaced by maize which is less laborious to grow, process and is preferred increasing for consumption. Most farming activities in the region follow the rainfall pattern that is bimodal with the long rains occurring from April to June and the short from August to October. The dry season spreads from November to February. Since the short rains are not heavy and not much happens in this season in many parts. Kakamega has one long rainy season from mid March to August while in Kisii and Nyamira, there is rainfall every month of the year, the least being in January and July, the highest in April and May. In those areas and Vihiga a second crop of maize and beans is grown during the short rains. 11

21 Table 5 Characteristics of the production system for a typical household Sites and Districts Average land size (acres) Livestock Kakamega Nandi Bungoma Vihiga Rachuonyo Kisii Nyamira Navakholo C. Kabras Kilibwoni Mutwot Nangili Tembelela Mbihi Gisambi Kakelo Dudi Kojwang Katunde Marani Suneka Bokiamburi Ekerenyo Zebu cattle < Crosses Grade cattle Indigenous poultry Exotic poultry Shoats < Crops Cassava Beans Ground nuts Arrow roots Maize Sugar cane Sugar cane Maize Sweet potatoes Beans Bananas Millet Sorghum Ground nuts Vegetables Maize Beans Sweet potatoes Maize Beans Sweet potatoes Vegetab les Banana s Sugar cane Coffee Maize Banana Sweet Potato Beans Sorghum Millet Sugar cane Maize Beans Bananas Maize Beans Fruits Coffee Tea Millet Sorghum Napier Maize Beans Cotton Sorghum Cassava Sweet potato Groundnuts Napier Maize Beans Sugar cane Cotton Ground nuts Napier Fingermillet Sorghum Sweet potatoes Cassava Maize Beans Coffee Tea Bananas Napier Vegetables Maize Coffee Tea Banana Sugar cane Finger millet Napier Vegetables Groundnuts Sweet potatoes Maize Coffee Tea Banana Finger millet Napier Vegetables Maize Coffee Tea Bananas Napier Vegetables Pineapple Sweet potatoes 12

22 The early part of the year before the long rains is dominated by land preparation and planting the main maize/beans crop. Weeding follows just as the rains set in April and May. Beans are harvested between May and August, while maize is harvested between September and November. In Kisii, Nyamira and Vihiga the second crop of maize and beans planted during the harvesting of the first crop is harvested early the following year. In Kakamega the short rains are used for planting beans, sweet potatoes vegetables and groundnuts. Most of the biannual food crops (bananas, vegetables, etc.) are established during the long rains. Coffee and tea are perennial but have peak-harvesting periods from September to November and April to October respectively. There is no distinct periodical activity for sugarcane. Once sugar cane (SSC) has been planted weeding takes place six times in a 22 months cropping season. It is then harvested and all the stumps burned down to give way to a new crop. Often the buyer company does the planting and cutting. 2.3 Livestock production All livestock types are kept in all the areas although in Nandi, dairy production is the predominant agricultural enterprise with tea as the main cash crop. In Kakamega, Kisii, Rachuonyo, and Nyamira most of the farm holds have about three to five Zebu cattle per farm producing one litre of milk per cow per day kept by grazing mainly on unimproved and some improved pastures as well as at the roadsides. The few farmers with grade cattle keep on average about two cows producing an average of seven litres of milk per cow per day (ditto) or three crosses producing an average of five litres per cow per day per under zero or semi zero grazing in most areas. The low number of grade animals kept per household and their low productivity is a result of low feed supply, particularly during the dry season, and poor husbandry methods and their interactions that result in deaths of these types of cattle. The prevalence of the Zebus in Kakamega is associated with the cultural practices (dowry payment and prestige) where the number of cattle per household is more valuable than the quantity and quality of their produce (milk). Some farmers reported problems with grass and fodder legume seed acquisition, e.g., Rhodes, Desmodium, Lucerne, fodder trees for supplementation forcing them to rely on Napier grass and natural pastures. The main planted forages are Napier grass, sweet potatoes and crop residues are used routinely. There are on average 10 to 50 chickens and three to five heads of goats and sheep per farm, all of them indigenous and kept for egg and meat consumption as well as a form of capital. Donkeys are used for traction and means of transport. Ducks are not popular among the farmers and the market outlet is minimal. Pigs are not popular because they feed a lot, they are destructive and are at sometimes associated with witchcraft. 13

23 There is high milk production during the peak rainfall period and demand and prices tend to be low. Towards December the demand increases and prices may improve as supply declines. The sale of cattle is high from January to March when animals are culled as a result of inadequate forage and to raise school fees. Sales are lowest in June increasing up to October before declining since there is then a low supply of the animals Livestock feeding calendar The quantities of Napier and natural pasture grass follow the rainfall patterns while banana pseudo-stems, sweet potato vines and sugar cane tops are available uniformly throughout the year though the latter are rarely used. Utilisation of Napier grass and by-products stover and bean hauls is highest in the dry season between December and February, and concentrate use increases when Napier grass is not readily available. Concentrates used in livestock feeding include bran, dairy meal, local brewers waste, molasses and spoilt maize. In most parts dry maize stover and bean haulms are available for use after harvesting the grain. In some areas, Napier grass is left to overgrow and the canes are used to make baskets Livestock breeding The main breeding system is by natural service either using improved bulls introduced by previous development projects or the local zebus. At the moment AI services are provided by the government at the district veterinary offices and some very few private providers. In the Kakamega and Rachuounyo attempts to set up private AI services have not fared well because the farmers' demand cannot sustain them. The farmers, moreover, find all the services expensive at the range of KSh to depending on the quality and source of the semen and transport charges. Natural service dominates using locally owned bulls and bull schemes set up by Livestock development Project (LDP), Heifer Project International (HPI) and a few other NGO projects. Their impact at improving the dairy genotype is, however, so far very low. In Rachuonyo there are only 3,000 grade cows to the 150,000 zebus despite more than ten years of LDP and National Dairy Development Programme (NDDP) activity Livestock diseases Tick borne diseases (TBDs), especially East Coast Fever (ECF), are endemic throughout the area. There are very few communal and privately owned dips and these are only occasionally functional. Most farmers control ticks by spraying occasionally their animals at home or dip where dips are functional. The incidence of TBDs and worm infestation are high during the rainy season from April to November, while Foot and mouth, lumpy skin and Newcastle 14

24 disease mostly appear in the dry season. In Kisii and Rachuonyo the prevalence of notifiable diseases especially foot and mouth disease (FMD) and lumpy Skin was associated with uncontrolled movement of livestock into and out of the districts while in Kakamega the presence of Trypanosomosis was blamed on tsetse infestation and proximity to forests. 2.4 Labour profile Women are busy throughout the year as they perform more duties on the farm from cultivation and harvesting to household chores. Men are more busy in the planting season when demand for labour is high. January to May is the peak labour period due to the many farming activities (land preparation, planting and weeding the year's main crop) while July to October have the least labour demands. Tea picking is less in the months of January to March then increases steady to peak in October. The cost of hired labour in the whole of western Kenya was uniform at KSh 50 (and a meal) per working day (7 am to mid-day) regardless of gender. Permanently employed labour, however varied from KSh 700 to 1500 per month including food and accommodation but depending on the farmer's resource capability and /or length of employment. There is no child labour, i.e. no use of school-age children, except when the farmers kids were on holiday from school. Malaria cases in humans increase just after the rains between July and August. 2.5 Farm and community resource maps Farm resource maps indicated some degree of intensification in all areas with livestock playing an important role as a source of food, income and manure (Annex 2). Most areas were predominantly producing own food and show much diversification in production. There was widespread use of crop by-products (bananas and sugar cane tops) to feed livestock. Al areas indicate high market activities with sale of food and cash crops, milk and also for sourcing inputs such as fertilizer and at times food. Many areas indicated communal use of labour on a rotational basis to help out with farm activities. In Kisii, there is sale of trees for firewood and charcoal as well as sale and purchase of manure. Issues arising include increasing intensification, concern with market access and policies on inputs and output prices and marketing. Community maps (Annex 2) show close relationships between communities and the ministry of agriculture, churches, schools, self help groups, markets, provincial administration, health centres, tea and coffee factories and the veterinary department. The communities did not have close relationships with many service providers including; KARI, water development agencies, electricity providers, stockists, Agricultural Finance Corporation (AFC) for credit. 15

25 Issues arising include weak core support services and identification of probable dissemination avenues namely provincial administration, ministry of agriculture, schools and churches. There is a need to strengthen weak linkages between farmers and KARI, AFC and stockists. 2.6 Constraints to crop and livestock production The constraints to farming in general and livestock in particular were discussed with the farmers in the various sites. There were 25 constraints mentioned between the 14 farmers groups in declining order of how many times they were mentioned (Table 6). Out of the 25 constraints, three (animal diseases, extension services and animal breeding) were mentioned by more than 11 groups, eight (roads, crop diseases, cost of inputs, water, land size, capital, marketing and quality of inputs) by between seven to ten groups, seven (output prices, soil fertility, human diseases, veterinary services, credit, management of dips and livestock feeds) by between four and six groups and seven (crop pests, gender imbalance, theft of animals, other sources of income, corruption, poverty and management of cooperatives) by fewer than three groups. The constraints are discussed in order of declining importance as defined by the number of times they were mentioned by the groups. Animal diseases came up as farmers argued that the presence of diseases has clearly contributed to the low presence of dairy animals in most of Western Kenya. They said that local indigenous animals are easier to keep and their dominant numbers could be attributed to their ability to withstand the prevalent diseases. Diseases are linked to poor disease prevention measures such as low number of dips, poor management of the few dips available, poor home spraying methods. Some service providers felt that farmers may be finding the cost of acaricides to be too high and using low strength washes or spraying their animals irregularly. This was also linked to doubtful quality of inputs. Linked to this were lack of veterinary services and the high costs of the ones available. Many farmers have yet to get over the former heavily subsidised government services. Lack of farmer training and extension services was a common constraint and mainly referred to inadequacy in training farmer on how to keep and maintain dairy animals. It was described in various ways in different places but mostly pointed out to inadequate extension staff. During a discussion with the field staff themselves, they pointed out the very low worker to farmer ratio, though no figures were given. Then there is the issue of inaccessibility linked to either the terrain that cannot be covered fast enough by whatever transport means available (on foot or by bicycle) or the lack of such transport. The increase in number of dairy animals in the early 70s after introduction through AI services or the dairy projects that existed then was not maintained. Many of the animals died because the farmers could not keep up with the nutritional and health management demands the animals required. This can 16

26 be directly linked to the lack of education. There was very little evidence of farmers using available technology to sufficiently utilise the available feed resources. For example, even if some farmers kept poultry there was no one utilising the birds' dropping as a feed supplement. Inadequate breeding services was reported variously as lack of good dairy animals, lack of breeding bulls, ineffective AI services, and expensive AI services from those who provided them. Just like clinical veterinary services the privatisation of AI provision has resulted in total lack of private providers in some places (Kakamega), and a very high cost for the ones existing. Though this is the constraint projects such as HPI and LDP have chosen to facilitate their impact in increasing the number of cattle of having dairy genotypes has not been very obvious. Farmers expressed an interest in the establishment of more bull schemes and many farmer groups have been set up primarily for this purpose but they were not successful. Again there was a possibility of farmers pointing out this constraint hoping to be supported by the project but the very low numbers of grade cattle in these areas makes this an important starting point to the development of the region's dairy industry. Infrastructure (roads) as reflected in the rundown condition of main and access roads that have affected all sector including agriculture. Services and markets have become inaccessible in some places and the cost of transport has risen as providers charge the higher vehicle costs to users. Thus, the cost of veterinary clinical and AI services have become correspondingly higher. In Kisii, Nyamira and Rachuonyo where farmers sell their own milk (i.e. with no middlemen or small traders) this has meant higher costs for each litre that finally reaches the consumer. There is also the element of time spent as farmers walk or cycle long and difficult distances to deliver milk leaving little time for attention to management of their cows. Crop diseases show the prominence of crop production in the region and the dependence the farmers have on these crops for food and income this constraint was inevitable and was mentioned in almost every site visited. Both cash crops (tea, sugarcane) and food crops (maize, pineapples, and sorghum) were reported to be under some recent attack by diseases or pests that the farmers have not yet overcome. Lack of water was mostly referred to inaccessibility to clean water or to mains supply which meant reliance on rain catchment and rivers. It was a common constraint, but did not feature high in ranking. 17

27 Table 6 Ranking of constraints to crop and livestock production in declining order of importance Kakamega Nandi Bungoma Vihiga Rachuonyo Kisii Nyamira Navakholo Kabras Kilibwoni Mutwot Nangili Tembelela Mbihi Gisambai Kakelo Dudi Kojwang Katunde Marani Suneka Bokiamburi Ekerenyo Animal diseases Extension services Animal breeding Roads Crop diseases Cost of inputs Water Land size Capital Marketing Quality of inputs Output prices Soil fertility Human diseases Veterinary services Credit Management of dips Livestock feeds Crop pests Gender imbalance 11 4 Theft of animal 5 11 Other income sources 6 9 Corruption 1 Poverty 1 Management of coops 1 18

28 Declining land sizes was a constraint in the entire region given the high population increases and continuous sub-division of land among children to units that could barely support sustainable agricultural enterprises. This meant that farmers have to adopt intensification technologies that are again hindered by their lack of capital and education/knowledge on what can be done. Lack of capital was also a common constraint, and in areas where farmers identified it as such it featured as the top most hindrance. It was also reported as 'high cost of inputs' and in one place (Nandi) as 'the high cost of transporting produce to market' relative to the price of milk. There was a possibility of the farmers pointing out this constraint in the hope that the project was set to provide material support and in some areas this could have been the reason why it was stated. Previous projects in the area have had a strong feature of material provision but this approach has not had any sustainable impact because the achievements made by these projects were not obvious. In Kakamega and Bungoma lack of capital could be addressed by the incomes the farmers got from sugarcane. However, but this has not been the case because they said the cane took too long to be harvested due to crop type or there were delays in harvesting the cane (a management problem). The prices of the cane delivered were also reported to be low. This was pointed out in few places but is related to the above constraint. There were very few institutions offering some credit arrangement to smallholder dairy farmers. In Nyamira it was the stringent loan terms that were mentioned while in another it was the very high interest rates. All in all very few farmers were ready to go for credit because they were afraid of not being able to repay, especially from dairy production and milk sales. The only credit facilities mentioned were those associated with the cash crops where farmers obtained material inputs (seeds, fertilisers, implements, etc.) payments were deducted from the produce delivered. As a result these facilities were only available for use in the cash crops but not to dairy. This was closely related to low output prices which were mentioned too. Poor marketing systems were mentioned for all farm products but more so for milk. This was exacerbated by lack of co-operatives and poor management and organization where they exist. This leaves farmers to market their small outputs on their own and covering long distances. Soil fertility could have contributed to the feeding constraint. It was mentioned everywhere as having compounded the problem of small land sizes. The farmers felt that their soils were not as good as they used to be previously when they could plant crops with very little or no fertilisers. As a result the production of, not just the foods but forage for the animals, had 19

29 declined over the years. In Kakamega, Rachuonyo and Nandi poor soil quality was also associated with soil erosion in the long rains. Human diseases were mentioned especially malaria in Kisii and Nyamira, and AIDS/HIV in Rachuonyo. This was as a result of declining health care services in public hospitals. Lack of feeds was mentioned in only five of the seventeen sites this constraint has contributed to other problems mentioned by the farmers. This refers to the low milk yield and, as a consequence, low incomes from dairy farming. The main feed resources for cattle are the rain-dependent grass pastures and fodders. Though the area is well endowed with rainfall, especially Vihiga, Nandi, Kisii and Nyamira the land sizes are too small to provide adequate amounts for the stocks kept and especially for grade cows. The farmers also complained of not being able to conserve feeds when there was plenty particularly during the rainy season due to lack of know how. The extensive use of land for food crops and cash crops implies that the utilisation of residues and by-products can be used to boost feed supply and this was not evident. For example sugarcane tops were not being used and in some places the farmers said tops actually reduced milk yield. Gender imbalance was mentioned where women talked of being left to do all the work. One of the reasons why breed development projects, e.g., LDP and HPI have not done well in Kakamega was felt (by the extension workers) as being the use of women groups without consulting or involving the men. Corruption was mentioned in Kakamega and associated with general poor governance and administration. This can be associated with mismanagement of co-operatives mentioned in Kisii and quite highly ranked there. 20

30 3 MILK MARKETING Milk marketing in Western Kenya is mainly informal with liberalisation of milk marketing in 1992 and the lifting of urban milk market monopoly previously enjoyed by the Kenya Cooperative Creameries (KCC). There are a few dairy farmers co-operative societies that came up to market members milk, after most collapsed or are on the verge of collapse. By December 1999, there were 16 active milk intake societies in Nyanza and seven in western. Other milk market outlets available for farmers in some areas include private processors such as Kitinda, Nasyanda and Magharibi dairies in Bungoma district. They source milk from farmers for processing. KCC is still operational in certain areas like in Nandi where there are two cooling plants in Kapsabet and Lessos. 3.1 Formal milk marketing There are few functional Dairy Co-operative Societies (DFCS), beset by management problems and operating below capacity. The main service rendered to members is bulking and transporting milk to processors. Others like the Vihiga Dairy Farmer s Co-op Society offer AI services and award soft loans to members. The milk price to members from the Coops varies between KSh 17 to 25 per litre depending on seasonal variation of milk supply. Mode of payment is mainly fortnightly or monthly. The members prefer this because they receive lump sum payments as opposed to cash payments, which are hard to manage. The milk disposal outlets include household consumers, institutions and hotels with the price ranging between KSh 25 to 30 per litre for fresh milk and KSh 32 for fermented milk. The table below gives a list of the few dairy co - operatives existing in the region: Table 7 Cooperatives, year of establishment and current milk intake Name of Farmers Co operative Society Year of establishment Vihiga/Sabatia Awach Multipurpose Current milk intake (litres) Quality control measures undertaken by the DFCS prior to milk intake include use of a lactometer and match test. The main constraints hindering the progress of the DFCS are management problems and competition from informal market agents. 3.2 Informal milk marketing This is the most common channel of milk marketing in the region. Majorities of the traders interviewed rely on this type of business as a sole income generating activity. There are direct sales from the producers to the consumers and as well from traders to consumers. Some of the traders sell the milk from dairy bars (milk bars), mainly located in the urban areas and 21

31 retailing raw, boiled or fermented milk. Other traders are not stationed anywhere and move around, hawking milk. There are also milk distributors/wholesalers who source milk from farmers and sell to traders who in turn sell to the consumers Milk bars Most milk bars have licenses from the Kenya Dairy Board (KDB) and the Municipal Councils. As long as the licenses are up to date, they face no harassment from the KDB or the Council. The amounts paid comprise KSh 2,000 per year to KDB, cess KSh 0.20 per litre, Municipal Council annual fees of KSh 200, annual food and hygiene license at KSh 300, and public health annual license of KSh 200. Amounts handled by these traders are as low as 40 litres and as high as 200 litres. The milk deficit areas are Kakamega, Vihiga and Rachuonyo, where milk has to be sourced from very far to meet local demand, which has a bearing on the high sales price (Case 1 and 2). Case 1: Everlyne Vugutsa s milk bar is situated in Vihiga town. She sources milk from Danger forest market in Nandi (about 70 km) (40 to 60 litres) at KSh 16 to 17 per litre. She pays a further KSh 100 for her own transport and KSh 30 per 20 litre jerrican. The final milk sale price is KSh 20 per 750ml treetop bottle. The quality control measure she undertakes prior to milk purchase includes pouring a little milk on the ground and observes the flow. The faster the flow, the higher the adulteration levels with water. Case 2: Emmah Adolwa runs two milk bars in Kakamega. She sources milk from two suppliers Malanga dairies and Turbo (75 km away). She handles about 200 litres per day and sells at KSh 28 per litre. She tests for milk quality using a lactometer and wants to start undertaking the acidity test. She has a freezer for preserving milk. The main mode of transporting milk is by public means (matatus), bicycles ( boda-boda ) and a few of the traders use private vehicles. Milk purchase price ranges from KSh 13 to 16 per litre after the rainy season (May to October) and peaks to KSh 17 during the dry season (January to March). The sales price range from KSh 20 to 28 per litre for fresh milk and KSh 32 to 40 for Lala (fermented). The main market outlets include households, hotels and institutions. Payment is commonly cash on daily basis for households and monthly or weekly for hotels and institutions. Occurrence of milk spoilage is not high, as almost all milk is disposed off the same day. In addition some of the traders have freezers to cool the milk Mobile milk traders Hawkers The KDB officials, council officers and the public health officials restrict hawking on the grounds of non-payment of licenses and concern for hygiene. The KDB offices in Kakamega indicate that they had licensed 10 hawkers in the district. On the other hand, in areas like Nyamira and Oyugis, the traders have never experienced or heard about the activities of the 22

32 KDB. There is still no proper regulation of the informal milk markets, because there are many traders still not paying the required licenses, while others have to pay both at the point of collection and sale (Case 3). Case 3: Milk Hawkers at Danger in Nandi The traders source their milk from two organised collection centres at Danger in Nandi. While at the purchase point, they pay the County Council levy. In addition at the point of sale, such as in Kakamega or Kisumu they pay the Municipal County charges. In Nandi, traders source milk as early as 4 a.m. as they have to travel long distances (Annex 3). They handle small amounts about 40 to 60 litres per day as required by milk bars, using plastic containers (jerricans) which are less costly and easier to transport using matatus and bicycles. The source of milk is mainly directly from individual farmers, organised collection centres as in Case 3, or purchases from other traders. Sales are made in the morning, and by 9 a.m. the sales are completed. The main market outlet are the individual households, with the purchase price ranging from KSh 13 to 22 per litre and the sales price KSh 22 to 25 per litre for fresh milk. Payment is made on daily basis Milk wholesalers/distributors There exist a few milk distributors in the region. One interviewed in Kakamega indicated that he sources milk from Turbo where he has employed five milk agents (three full time and two part time), collecting milk from collection points. The suppliers are 31 farmers all within a radius of five kilometers. Milk is brought at a central point and dispatched on matatus for Kakamega, the sales area. He pays a transport charge of KSh 50 per can and a further KSh 30 per can per milk agent. He sells to private traders, mainly the hawkers at KSh 25 per litre. Sales prices are higher in milk deficit areas because of the associated transaction costs as the milk source is far. The highest milk sale price recorded was in Oyugis (KSh 30 per Litre. fresh milk and KSh 32 per litre. fermented) (Figure 2). In some areas like Suneka (Kisii) and Ekerenyo (Nyamira), farmers sell their own milk, as there are few traders. The quality control measures used prior to milk intake is the use of a lactometer, matchstick and organoleptic tests such as odour, smell and milk flow, (Case 1 and 2). Left over milk is soured and sold to consumers the next day at a price of about KSh 32 per litre. There are few occurrences of milk spoilage as they have informal contractual arrangements whereby the suppliers replace spoilt milk. 23

33 Ranking of marketing constraints was carried out in Suneka (Kisii), Ekerenyo (Nyamira), Kojwang Katunde (Rachuonyo), Kilibwoni (Nandi) and Nangili (Bungoma) (Table 8). Traders suggestions on opportunities for overcoming constraints Figure 2 Purchase and sales prices (KSh per litre) by site Kakamega Vihiga Nandi Ekerenyo Oyugis Suneka Bungoma Purchase Sale 1. Organised milk marketing, with clear cut rules and regulatory authorities to ensure fair play on all sides 2. Improvement of milk quality (unadulterated milk) and consumer awareness 3. Government /NGOs should create a credit scheme to provide farmers and traders with soft loans to purchase capital equipment Table 8 Ranking of constraints cited by milk traders/farmers Kakamega Vihiga Kisii Nyamira Rachuonyo Nandi Bungoma Harassment by the KDB Fluctuations in supply/prices (season) High transportation costs Lack of knowledge on milk 7 handling Lack of storage facilities 3 8 Competition from other traders 2 Adulteration of milk by suppliers High cost of capital items High default rates 4 Lack of AI services/bull schemes 4 3 Lack of organised marketing Transportation problems (foot)

34 3.3 Kenya Dairy Board The Kenya Dairy Board (KDB) has provincial offices in each province. The Western province office is situated in Kakamega. One of the objectives of liberalisation of milk marketing was to allow for more processors. However, according to KDB in Kakamega, it opened markets for milk vendors (hawkers) which pose potential public health hazards. It also caused the collapse of the Kakamega Dairy Farmers Co-operative Society and has led to the few cases of Tuberculosis and Brucellosis reported in district hospital. In Kakamega, they have licensed 50 farmers, 20 milk bars, 1 mini dairy (Kitinda) and 10 hawkers. They have also licensed KCC, Premier and Ilara processors. The requirements for a license are that one should have a premise, use aluminum milk cans, knowledge on milk handling and eventually have a cooler. KDB's role in the liberalised market is to carry out quality control checks but they are constrained by lack of testing kits. The way forward according to the KDB is to educate farmers and milk market agents on milk handling hygiene practices and to encourage more processing of milk. 25

35 4 Service Provision 4.1 Veterinarians, Animal Health Assistants and Stockists Incidences of cattle diseases are high in the region as a result of poor animal management, poor spraying methods and feeding practices. The main diseases include tick - borne diseases (East Coast Fever, Heartwater and Anaplasmosis) Brucellosis, scours in calves, mastitis, nutritional disorders and foot and mouth disease. The drugs are unaffordable to many farmers who therefore settle for cheap, ineffective drugs or herb using quacks. The drugs stocked are from Coopers, Cosmos and Bayer. There are very few stockists serving vast areas. The items stocked include animal feeds, acaricides and minerals (Table 9). Table 9 Prices of livestock inputs Item (KSh per 100ml) Livestock feeds (KSh per bag) Dewormer Wormicid 70 Dairy Meal per 70Kg bag Nilzan plus 180 Mineral Lick 160 per 2Kg bag Stock spray Triatix 240 Almatix 180 Stelladon 170 The sources of livestock feeds include; Unga Feeds, UniFeeds and Milling Corporation of Kenya. Constraints cited by the service providers include; 1. Lack of means of transport 2. No record keeping by farmers 3. Reliance on quacks by farmers until animal condition is critical 4. Unfair competition from quacks 5. High default rates by farmers 4.2 AI services Procurement of semen is from CAIS and ABS. Government services through the veterinary department do not serve well because of lack of transportation in most areas. In Rachuonyo district, AI is currently provided by the Government run Oyugis scheme. The charges are KSh 300 per series of four services. Private AI providers such as the Amani Centre (Oyugis) and PUSU Women s group have collapsed. The Amani centre started providing services to farmers benefiting from the Heifer Project International (HPI), but was extended to other farmers. The scheme collapsed due to high charges caused by high cost of liquid Nitrogen from CAIS. Currently, there are bull schemes set up by HPI and private farmers. The constraints cited include transport, low conception rates due to poor heat detection and deterioration of semen quality due to poor storage and handling facilities. 26

36 5 Conclusions The conclusions being presented here are derived from interactions and deliberations with the 14 farmers groups within the seven districts. They are not precise and they were not intended to be, but are general and indicative. They present an initial attempt to understand the region and set the scene for further interventions by research and development organizations such as SDP. Most regions except Rachuonyo have the potential for intensive livestock production due to favourable climatic conditions. Extensive dairying is restricted to Nandi, Bungoma and Kakamega where land sizes are relatively larger. Intensive dairying is to be found in Vihiga, Kisii and Nyamira. Rachuonyo has both extensive and semi-intensive dairying. In the region farmers also have a relatively long experience with farming crops and livestock. From the historical development farmers have been responsive to incentives as shown by rapid increases in production of crops and milk when conditions are conducive and then quickly reducing production when conditions change. In Bungoma and Kakamega, there exists a bias on crops enterprises especially sugar cane. Zebu are preferred to grade cattle for traditional cultural reasons. These cultural factors may be the greatest constraint to dairy development along side lack of forage, technical know how and diseases. The few grade animals cannot sustain a private AI scheme. Intensification of dairy would increase the household labour demand and offer alternative sources of income. Rachuonyo is unique from the other areas in that most parts are drier and not densely populated. However, the potential for intensification exists in a few areas such as Oyugis In Vihiga, there is potential for increased intensified dairy production. The farmers have some experience in livestock production, though they still have a strong leaning towards crop enterprises. The milk market is good whilst the limiting factors are unavailability of breeding stock and cost of animal health services. In Kisii and Nyamira there is still heavy reliance on cash and food crops whose prices are attractive, and the majority of the farmers have local cows. Main problems are animal diseases, availability of grade animals, feeds and input prices. As with Rachuonyo, human diseases especially malaria are of major concern. There are concerns with management of societies. Western and Nyanza are generally deficient in milk and all flows tend towards the major towns. A few Dairy Farmers Co-operative Societies are still operational though beset by management problems and competition from milk hawkers. Trade in milk is mainly informal 27

37 in the region, with participants involved in transportation and distribution. The informal traders serve both farmers and consumers more efficiently, a situation that is unlikely to change until milk production rises significantly to the point where farmers have difficulties selling milk. The role of regulatory bodies is unclear, in that where they exist, they are seen to be too restrictive and where they have no offices, they are ineffective. There are opportunities for growth in milk marketing if there s improvement in the organization of traders and encouragement from regulatory bodies to trade in larger volumes under hygienic conditions. For SDP, the issues that will have to be addressed will revolve around the huge milk deficit while there is still potential to increase milk production. Immediate issues would be to improve our understanding on why the production gap exists and why farmers do not adopt technologies that would increase milk output. 28

38 6 References Environmental Systems Research Institute (ESRI) Arcview Version 3.1, 380 New York Street, Redlands, CA, USA. Jaetzold, R. and Schmidt, H Natural conditions and farm management information. In: Farm Management Handbook of Kenya, Volume II, Ministry of Agriculture, Nairobi, Kenya. Owango, M., S. Staal, M. Kenyanjui, B. Lukuyu, D. Njubi and W. Thorpe Dairy Cooperatives and Policy Reform in Kenya: Effects of Livestock Service and Milk Market Liberalisation. Food Policy, 23(2), April 1998, pp Smallholder Dairy Project Phase 1 achievements seminar. Proceedings of the seminar held at KARI headquarters, Nairobi, March 9, Staal, S. J., L. Chege, M. Kenyanjui, A. Kimari, B. Lukuyu, D. Njubi, M. Owango, J. Tanner, W. Thorpe and M. Wambugu Characterisation of dairy systems supplying the Nairobi Milk market: A pilot survey in Kiambu district for the identification of target groups of producers. MoA/KARI/ILRI Collaborative Research Report, Smallholder Dairy (R&D) Project, Nairobi, Kenya. Staal, S. J., M. Owango, H. Muriuki, B. Lukuyu, F. Musembi, O. Bwana, K. Muriuki, G. Gichungu, A. Omore, M. Kenyanjui, D. Njubi, I. Baltenweck And W. Thorpe Dairy systems characterisation of the Nairobi milk shed: Application of spatial and household analysis. MoA/KARI/ILRI Collaborative Research Report, Smallholder Dairy (R&D) Project, Nairobi, Kenya. 29

39 Annex 1: Mapping of spatial variation in Western Kenya 1. Cattle densities in Western and Nyanza Just released by the ILRI-GIS section is a dataset on cattle numbers and densities for each division in Kenya. Data for this coverage were obtained from the Livestock Production Department which provided the latest district level report on livestock numbers ( ). Since the western regions of Kenya tend to be high potential livestock regions in terms of agro-climatic conditions, total cattle densities were mapped as opposed to the densities of grade cows in the area. Although cattle rearing in general is widely prevalent, only a few areas tend to show intensive numbers of dairy animals. 2. Tick borne disease distribution Data on tick borne diseases were provided by the ILRI-GIS section and supplemented by household data on the occurrence of tick related illnesses from the Other District Survey. The available layers cover brown ear tick distribution data and expert opinions on the spread of ECF and other tick borne diseases. To check the consistency of all these very different tick and disease data, an overlay was made of the distribution layers and point data. The resulting overlay of all presents a consistent pattern of expected and reported distribution of tick borne diseases: the whole of Western Kenya is affected. However, several important parameters were not available when assembling data, such as different breeds and management practices, which affect infestation. Hence, the forthcoming characterization study in Western might contribute to a more sensitive layer on actual tick related risks. 3. Spatial distribution of ethnic groups The obvious but indistinct relationship between ethnicity and dairy practices was reason to add basic data on the spread of ethnic groups throughout the districts under study. The Kenya Central Bureau of Statistics provided data the presence of different ethnic groups for all districts at sub-location level. For sub-locations with one predominant ethnic group (being >95% of the total inhabitants) no secondary ethnic group was recorded. Most areas proved to be quite consistent especially Kisii and Nyamira. In all other areas, secondary ethnic groups were recorded, with Nandi harboring the largest number of other ethnic groups apart from the predominant Kalenjin. 30

40 Figure 3 Grade and total cattle densities 31

41 Figure 4 Distribution of ticks and reported tick-borne disease cases 32

42 Figure 5 Distribution of ethnic groups in western province 33

43 Figure 6 Distribution of ethnic groups in Nyanza province 34