Dairy cattle farming in Kenya contributes about 14% of Agricultural Gross Domestic Product  with an estimated annual growth rate of 4.1% (Behnke and Muthami, 2011). In addition, it is the most rapidly expanding dairy sector within sub-Saharan Africa, and with over 85% of the dairy cattle  population   raised   within   Eastern   Africa  region (Odero-Waitituh, 2017). The current dairy cattle population in Kenya is estimated at 4.3 million herds and they produce about 3.43 billion liters of milk annually, which constitutes about 18% of the paltry 3% contribution of sub-Sahara African countries global milk production (Behnke  and  Muthami, 2011). Indeed, smallholder  dairy farming plays an important part in the dairy value chain within Sub-Saharan Africa, where three technological approaches have been used to improve productivity (Chagunda et al., 2016). These include applying agricultural ecological processes such as continuous housing of cows, practicing cut-and-carry feeding system, introduction of purpose bred forages, pastures and agro-forestry within the dairy systems, utilization of modern livestock breeding techniques including importation of dairy breeds and crossbreeding of indigenous cattle with exotic breeds, and other types of socio-economic intensification.

It is reported that smallholder dairy farmers own over 80% of the dairy cattle population in Kenya, and they produce over 56% of total milk output in the country (Kaitibie et al., 2010; Odero-Waitituh, 2017). These cattle are raised under both intensive and semi intensive systems with the predominant breeds kept comprising of Friesian, Guernsey, Ayrshire, Jersey and their crosses (Bebe et al., 2003). These production systems are not unique to Kenya, but have been reported to be practiced in different parts of the World where they play a significant role in production of milk which support family incomes and the economies (Devendra, 2001). In addition, milk is also produced from the numerous herds of indigenous cattle raised by many smallholder and pastoralist communities within the country (Onono et al., 2013). According to Bebe et al. (2003), the management decisions on breeding objectives by the Kenyan smallholder dairy cattle farmers are guided by a number of factors including level of milk yield, high butterfat yield, heavier body weight, unselective feeding behavior under zero grazing units, hardiness and disease resistance. In these farms, the Friesian, Ayrshire, and Jersey cattle breeds are preferred because of the high level of milk yield as compared to the zebu, which is preferred for their hardiness and disease resistance.

Since liberalization of the dairy industry in Kenya in early 1990s, marketing of fresh milk and other value-added products have been controlled by market forces. Accordingly, milk supplied to the formal sector declined due to unpredictable prices and delayed payment by some milk processors (Odero-Waitituh, 2017). This also led to the emergence of another market segment characterized by farmers who sell milk through informal marketing channels which handle raw milk and other dairy products and they often fail to conform to milk market regulations (Institute of Economic Affairs, 2001). It is estimated that about 80% of milk produced by smallholder farmers is neither processed nor packaged, but is purchased by consumer in raw form through informal milk outlets (Institute of Economic Affairs, 2001; Kaitibie et al., 2010). Marketing of milk through these informal channels is associated with numerous disadvantages including loss of benefits associated with value addition, reduced opportunities for employment creation, lack of quality standards to control  adulteration, poor controls in ensuring hygienic and clean healthy products that are produced and distributed, reduced safety standards to consumers, poor keeping quality for milk and reduction in contribution of export earnings from surplus milk that is produced.

Although the smallholder cattle production activities contribute enormously to the economies of the farming communities in Kenya, these enterprises are greatly hampered by a number of challenges including poor access to breeding services, inadequate and poor quality feeds for the dairy cattle, occurrence of livestock diseases, poor access to credit facilities and lack of markets for the fresh milk which is often characterized by inadequate processing of raw milk and the numerous informal milk markets characterized by hawking of raw fresh milk (Bebe et al., 2003b, 2008; Odero-Waitituh, 2017). Much of the available literature on dairy farming in Kenya are based on reports and peer reviewed publications from the highland agro ecosystems where dairying has been practiced for several years, however, there is a paucity of studies that have reported milk production and challenges facing dairy farming within the mixed production systems in the lowland areas of western Kenya, where milk is still obtained from farms predominated by indigenous cattle and crosses between indigenous and exotic breeds, with only a few farms keeping pure exotic breeds. This paper presents baseline information on the level of production in these systems; in addition, it reviews the prevailing challenges facing this production system and the opportunities that are available to be exploited for improved milk production. The findings will be useful for policy makers and organizations that support livelihoods of the rural communities through implementation of projects that target activities along the dairy value chains under similar production settings.

Study area and design

A cross sectional study design was used for collection of primary data from farms, while secondary data was obtained from published and grey literature sources. The study was conducted in Homa Bay County in western Kenya, located within the Lake Victoria basin, and its geographical coordinates are 0° 31’ 0’’ South and 34° 27’ 0’’ East. The primary data was only collected from Rangwe sub-County which is one of the eight sub Counties in Homa Bay, but secondary data and key informant interview focused on an area that covers two sub-Counties: Rangwe and Kasipul/Kabondo sub Counties. The cross sectional study design has been described to be useful when collecting data from systems where minimal information is available about the parameters that are being investigated (Kothari and Garg, 2014). These regions were selected because of the emerging dairy farming practices which are partly supported by the partnerships between County government of Homa Bay and its development partners to accelerate development along the dairy value chain.

Selection of study units and data collection

The farms  targeted  for  data  collection were keeping indigenous cattle breeds, crosses between indigenous and exotic breeds and farms keeping pure exotic breeds of cattle (Jahnke, 1982; KARI/ODA, 1995). Additionally, key informant interview was done with the County Director of Veterinary Services. The farmers who participated were selected by the field animal health assistants in Rangwe Sub County; with the farms drawn across all Wards within the Sub County. A total of 52 cattle farmers were interviewed of which 34 were raising indigenous breeds, 15 raising cross breeds and 3 raising exotic breeds of cattle. Data was collected from these farmers using a pre-tested semi structured questionnaire, while an interview guide was used to collect data from the County Director of Veterinary Services. The cattle farmers questionnaire included questions on the size of household land ownership in acres, the cattle herd sizes per household, the average milk yield per cow per day in liters, the average price of milk per liter in Kenya shillings, the average fees farmers were charged for veterinary services, the average cattle market prices and fees charged for artificial insemination services, and the challenges associated with cattle rearing in the region. In addition, the County Director of Veterinary Services was asked to list the diseases that impacted farmers’ livelihoods, and which of these diseases occurred in farms at a higher frequency. Similarly, they also listed the opportunities that existed for improvement of cattle production in the region based on the different types of projects the County government had initiated that targeted improvements on both crop and livestock sectors. Additional information on types of investment projects on crop and livestock sub-sectors were obtained from the official website of Homa Bay County government, and its development partner institutions like the International Livestock Research Institute (ILRI).

Data management and analysis

The quantitative data that was obtained from cattle farmers and grey literature sources were analyzed through computing the descriptive statistical measures including mean, range and percentages. Inferential statistical measures were also computed to test the level of significance between the mean differences for continuous quantitative variables between farms that were keeping exotic, crosses and indigenous breed of cattle. These quantitative variables included cattle herd sizes, daily milk yield, fees charged for veterinary services, market prices for mature cattle, fees charged for artificial insemination services, family land ownership in acres and market prices for milk. This analysis was achieved by performing One-way analysis of variance and for all statistical analysis which was done using Genstat statistical package (VSN International, 2011)at 95% level of confidence. For the qualitative variables that included description of different challenges facing cattle farming and existing opportunities for dairy cattle improvements through projects initiated by the County government, and its partners, the framework analytical approach was used to summarize the data using thematic analysis as described by Gale et al. (2009, 2013). The data collected from the County veterinary services on disease challenges facing the cattle farming communities were used to triangulate findings that were obtained from cattle farmers on various disease challenges that existed in the region.

Review of livestock population and milk output in Homa Bay County

Most of the dairy animals raised in Homa Bay County were of indigenous breed of cattle, estimated at 180,048 herds of cattle, constituting about 97% of animal population; however, there was a smaller population of exotic breeds of dairy cattle comprising about 4,848 herds of cattle which were raised under intensive farming systems    that   constitute    about   3%   of   the    animal population (Table 1). In addition, some farmers also kept dairy goats. These animals produced about 18 million liters of milk annually, of which the indigenous breed of cattle produced about 12 million liters comprising about 69% of total milk produced, with the exotic breed producing 5 million liters, which was 30% of the total milk produced. The dairy goats on the other hand produce 50,400 L of milk. This level of milk production created shortages since the estimated demand for milk in Homa Bay County was about 34 million liters annually, resulting in an estimated shortage of about 16 million liters of milk, which comprised about 45% of the level of milk production.

Description of dairy cattle demographics in Rangwe sub-County

Most of the farms included in the study owned about 4 acres of land (Table 2). However, the herd sizes were 7 (2 to 18) for those farms keeping indigenous breeds of cattle, 4 (2 to 6) for those keeping cross breeds and 3 (2 to 4) for those keeping exotic breeds of cattle. The farms keeping exotic breeds had cows producing averagely 18.33 L of milk daily, with farms keeping cross breeds producing about 7.2 L daily. Those with exotic breeds of cattle sold their milk at a premium price of ksh 75 L^-1, with the indigenous farms selling their milk at ksh 68 per liter. The farms keeping exotic breeds generally incurred higher cost on veterinary services than in other systems; similarly, the market prices of dairy cattle in the exotic farms averaged Ksh 51,667 compared with Ksh 13,383 for farms keeping indigenous breed. The artificial insemination services were only practiced on farms keeping exotic and cross breeds. The average land ownership by these cattle farmers and market prices for selling a liter of milk were not statistically different across the systems (P > 0.05) (Table 3). However, cattle herd sizes, daily milk yield per cow, average fee charged for veterinary services, market prices for purchasing cattle and charges levied for artificial insemination services were statistically different between the studied production systems (P < 0.05).

Description of challenges affecting cattle farming in Rangwe sub-County

The sources of income for households in the three cattle production systems (indigenous, crosses and exotic cattle breeds) included crop and livestock farming, small businesses and employment (Table 4). Milk which was the main output from the farms was sold to neighbours and used for household consumption. In addition, for the farms keeping crosses and exotic breeds of cattle, they would sell the excess milk to consumers within the local market   places.  These   farms   fed  cattle  on  extensive grazing and crop residues, but for the farms keeping cross and exotic breeds, they would also feed commercial feeds including hay, salt licks and planted fodder   trees.  The   challenges  associated   with  these sources of feeds included frequent and prolonged drought conditions, water shortages, high costs of feeds and inadequate advice on dairy cattle feeding and feed formulations  for   the farms keeping  exotic breeds.  The farms mostly obtained water from rivers and boreholes, except for farms keeping exotic breeds that relied on tap water. Disease control in these farms was often achieved through vaccinations which were conducted by the County government, regular spraying of cattle by  farmers to control ectoparasites, treatment of sick cattle, and deworming. In addition, for the farms keeping exotic breeds, regular monitoring by veterinary officers was reported. Veterinary services were provided by owners of agrovets  shops  (shops  selling  agricultural  inputs),  and animal health assistants, but for farms keeping indigenous breeds, untrained veterinary personnel were often used. The main sources of information for livestock keeping and management included owners of agrovet shops, animal health assistants, veterinary officers, attendance of agricultural seminars; however, for the farms keeping indigenous cattle breeds, they would use untrained veterinary personnel as sources of information, which was a constraint to dairy development in the area since these untrained personnel cannot provide the right information on disease control and the correct dosages of veterinary medicines. The farms keeping exotic breeds mostly used artificial insemination as a breeding technique; however, for the indigenous farms, the preferred breeding method was natural mating. The general challenges that faced these farms included presence of diseases, frequent drought conditions, lack of market for milk, poor road network for transportation of milk to markets, high costs for veterinary and breeding services, high cost of commercial feeds, shortage of capital to expand the farms, inadequate veterinary officers, unreliable artificial insemination services, and lack of knowledge on proper feeding of livestock.

Disease challenges that affected cattle production in the Homa Bay County included tick-borne diseases (East Coast fever, Anaplasmosis, Babesiosis and heart water), trypanosomiasis, lumpy skin disease and foot and mouth disease. According to the County Director of Veterinary Services, Anaplasmosis, Babesiosis, heart water and East Coast fever were the most commonly reported cattle diseases in farms in that order. But lumpy skin disease, foot and mouth disease were not frequently reported in farms.

Partnerships with County government of Homa Bay to support crop and livestock sub-sectors

Opportunities were presented through investment by County government and its partners in agricultural activities (Table 5). For example, the country government of Homa Bay had invested in promotion of technology transfer by modifying the environment through investment in building green houses and provision of seed and fertilizer subsidies given to crop farmers. Within the livestock sub sector, the County government implemented an accelerated value chain development with the aim of increasing milk production which bridges the gap between the estimated demand and the current level of milk production. Furthermore, the introduction of Brachiaria grass to smallholder livestock farmers, which is a drought resistant source of fodder with high protein content, would alleviate challenges of supply for quality in water harvesting through construction of water pans to alleviate challenges associated with frequent and prolonged dry spells, which hampers availability of livestock feeds.

Indigenous cattle population were the main source of milk produced within Homa Bay County, but there is a significant contribution in milk production from the emerging farms that are keeping exotic breed of cattle with their crosses and dairy goats. However, the level of milk production within the County cannot match the estimated demand for milk, the shortage which is mostly supplied with processed milk distributed by the major milk processing companies in Kenya (Stirling Consultants Limited, 2014). Reliance on indigenous cattle breed for milk production is also a bottleneck for dairy development in the region, since the level of milk yield from these breeds of cattle is generally low, a finding which was previously reported for other cattle systems in Kenya (Muhuyi et al., 1999; Onono et al., 2013). In these areas, improved breeding techniques such as the use of artificial insemination and embryo transfer are less frequently used, except for the few farms that have adopted the use of these techniques for improved livestock breeding (Karanja, 2003; ILRI, 2016). Indeed, the farms which were using these techniques for breeding generally had a higher level of milk production, as compared to the farms keeping indigenous cattle breeds and who were practicing natural mating. Furthermore, the accelerated dairy value chain project which is implemented through a partnership between the County government of Homa Bay and the International Livestock Research Institute which aims at upgrading genetics of the indigenous cattle through fixed time artificial insemination will go a long way in increasing the dairy productivity (ILRI, 2016). In these farms, the average number of cattle raised for the farms keeping exotic breeds were 3 herds of cattle in a land holding of about 4 acres, and this estimate compared favorably with the description of smallholder dairy farms reported in central Kenya, which have a median farm size of three acres and a median of three dairy cattle kept per farm, with an average milk production of about 6 kg per day (Gitau et al., 1994).

The review has further identified several challenges affecting dairy cattle production in Homa Bay County. The main challenges for dairy cattle production included frequent droughts, lack of quality feeds and high costs of commercial feeds, high costs of veterinary services and unreliable services provision, and presence of livestock diseases. The diseases of cattle reported in these areas include Anaplasmosis, Babesiosis, and Heart water, in addition to East Coast fever and trypanosomiasis, of which majority of the diseases were tick-borne infections. Indeed, from a study on molecular detection of tick-borne pathogen diversities in ticks from livestock and reptiles along the shores and adjacent islands of Lake Victoria and Baringo in Kenya, disease causing pathogens responsible for causation of the listed diseases were identified (Omondi et al., 2017). These diseases have been   reported  to  occur  with  a  high  frequency  across different production systems in Kenya. For example, Okuthe and Buyu (2006), reported the occurrence of Anaplasmosis, East Coast fever and Babesiosis in smallholder farms within the highlands of Rift valley, while the occurrence of East Coast fever have been reported in the central parts of Kenya, where they have been reported to be the cause  of high mortality in calves (Gitau et al., 1997). With regard to the high costs and unreliable veterinary services delivery to the indigenous cattle farmers, the findings are in agreement with challenges which have been reported for animal health delivery systems in sub Saharan Africa. Indeed, bottlenecks associated with the provision of veterinary services and provision of extension services to livestock farmers in sub Saharan Africa, and Kenya in particular have previously been described, whereby the supply of veterinary services for the most part are through non qualified personnel, with several non-governmental organizations supporting delivery of veterinary services through community based animal health workers or disease reporters who have no basic training in animal health. Apart from the disease control interventions by the County government, there were no programmatic interventions to control the widespread livestock diseases by the development partners. This is in contrast to interventions on livestock within the arid and semi arid lands in Kenya, where livestock development projects have always focused on enhancing veterinary services delivery  besides   the  supply  of  feeds  (Leonard,  2000; Okwiri et al., 2002; Abdilatif et al., 2018).

Cattle were mostly raised on extensive grazing, but for the exotic dairy cattle breeds, they were entirely under zero grazing systems of management. These results therefore reveal different level of intensification with projects targeting both ecological and genetic approaches (Chagunda et al., 2016). These levels of intensifications are characterized by projects targeting upgrading of genetics of local indigenous cattle breeds through the fixed time artificial insemination and introduction of Brachiaria grass which is a high quality fodder grass tolerant to drought conditions and a high protein content (ILRI, 2016; USAID, 2017). In addition, the County government of Homa Bay had implemented the construction of water pans within the County to mitigate the frequent occurrence of dry spells. The success of this project would help alleviate challenges of perennial dry conditions and poor feed quality as well as their unavailability during droughts that affect cattle productivity. Indeed, the presence of these challenges greatly impacts milk production which is a consequence of feeding poor quality feeds, coupled with the poor genetic potential of animals. Also, the milk produced from these farms were mainly sold to neighbours and in local market places with the excess consumed in households.

Indeed, the post liberalization period of the dairy industry sector in Kenya had reportedly resulted to smallholder farmers selling their milk through non formal outlets, characterized by poor enforcement of regulations, and which market outlets were still used by farmers for milk marketing (Kaitibie et al., 2010).

Indigenous cattle breed currently produces most of the milk in Homa Bay County, however there are farms keeping exotic breed of cattle with potential for high milk production. Some of the challenges facing these cattle farmers include poor quality feeds and their unavailability during the drought periods, occurrence of livestock diseases that affects productivity, poor veterinary services provision for the farms keeping indigenous cattle which are the main source of milk consumed within the County; and lack of a robust extension service provision within the areas which also results to inadequate provision of advise to farmers on feeding especially for exotic pure breeds of cattle. Opportunities that exist include public - private partnerships between the County government and nongovernmental actors which have the potential to alleviate challenges of unavailability of feeds and poor livestock feed quality alongside upgrading of livestock genetics through the dairy value chain enhancement project to improve quality of dairy animals and enhanced milk yield in cattle herds. Although this report has presented a set of challenges facing these farmers and various interventions currently provided to the farmers, future studies should evaluate level of effectiveness of these public - private partnerships on development and on the alleviation of challenges that currently face livestock farmers under similar settings. In addition, the roles of other dairy value chain actors should be investigated to identify the factors that influence the pricing of milk.

The authors have not declared any conflict of interests.

The authors appreciate the cattle farmers in Rangwe who were interviewed during the course of this study, the local animal health services providers who assisted in the identification of the farmers; and the county director of veterinary services in Homa Bay County.