Value chain analysis of smallholder milk producer in West Hararghe Zone, Ethiopia

The study was aimed at analyzing milk value chains in West Hararghe zone with 140 producers were randomly selected from list of dairy producer kebeles. Meanwhile, 30 traders from different markets and 30 consumers in Chiro, Bedessa Gemechis and Mieso towns were selected and interviewed. Heckman two stage models were applied to identify determinants of milk market participation and volume of milk supply to the market. The results shows among those variables hypothesized to affect milk market participation decision, distance to the nearest market, crop income and age of household head were negatively and signifi cantly affecting participation decision and also livestock unit is affect participation decision signifi cantly and positively. Land size, improved dairy cows, livestock unit were signifi cantly and positively infl uencing the value of milk marketed. The value chain analysis revealed that the major value chain actors are input suppliers, producers, collectors, rural and urban wholesalers, retailers, cafes and consumers. It is also found that milk passes through several intermediaries with value added products (boiled milk and yoghurt) before reaching the ultimate users. Therefore, policies aiming at increasing producer’s provision of modern inputs, number of milk cows and improving livestock production of milk cows and change of their attitudes towards milk production and processing, participating to dairy market were recommended to accelerate milk value chains development in the study area. Research Article Value chain analysis of smallholder milk producer in West Hararghe Zone, Ethiopia Azeb Lemma Kebede* Adune Dinku and Mohammed Sheko School of Agricultural Economics and Agribusiness, Oda Bultum University, Chiro, Ethiopia Received: 30 May, 2020 Accepted: 30 June, 2020 Published: 02 July, 2020 *Corresponding author: Azeb Lemma Kebede, School of Agricultural Economics and Agribusiness, Oda Bultum University, Chiro, Ethiopia, Tel: +251920913784; E-mail:


Introduction
The estimate of cattle for the rural sedentary areas at country level is about 53.4 million. About 64% or 34 million of these are aged 3 years or above and play a number of economic roles in the livelihood of smallholder farmers. Livestock is primarily kept on smallholdings where it provides draught power for crop production, manure for soil fertility, fuel and serves as a source of family diet and source of cash income (from sale of livestock and livestock products) particularly when markets for crops are not favorable [1].
Dairy production is crucial in Ethiopia as milk and milk products are important source of food and income. Despite the huge potential, dairy production has not been fully exploited and promoted in the country. A number of factors such as use of traditional technologies, limited supply of inputs (feed, breeding stock, artifi cial insemination and water), inadequate extension service, poor marketing infrastructure, lack of marketing support services and market information, limited credit services, absence of producers' organizations, and natural resources degradation [2] have contributed to unexploitation of dairy potential. In addition, policy decision on assurance of quality and standards, product marketing, among others is taken in the absence of vital information on how they affect the entire value chain [3].
Like most developing countries in Africa, Ethiopia did not have a clear livestock and livestock development policy for many years up until the establishment of Livestock Marketing Authority (LMA) in 1998. Furthermore, dairy product marketing channels and their characteristics have not yet been studied and analyzed for different parts of the country. Earlier studies on local and regional dairy products market in the country include [2-6] Hence, the study area of West Hararghe has high milk production marketed without further processing to other products. This implies that producers at farm level not get appropriate profi t share than other intermediary. Even if, there were high production of milk there were no studies in the study area related to milk value chain analysis since milk value chain studies become essential to provide vital and valid information on the operation and effi ciency of dairy product marketing system, for effective research, planning and policy formulation. Knowing the behaviour of each actor from input suppliers to the ultimate consumers would be give information to solve problems and give ultimate solution. Therefore, in line with the market-oriented production strategy of the country's policy, the study is intended at bridging the information gap with regard to milk value chains, factors affecting milk supply and to participate in the milk market and to access the profi t margin of different actors in West Hararegie Zone, Ethiopia.

Description of the study area
The study was conducted in West Hararghe zone, Oromia National Regional State of Ethiopia and the capital city of the Zone is located about 326 km east of the capital city of Addis Ababa along the main road. Based on the population projection value (2018) report, West Hararghe has a population of 1,951,706, an increase of 47.16% over the 1994 census, of whom 989,861 are men and 961,845 women; with an area of 15,065.86 square kilometers, the zone has a population density of 124.23. While 160,895 or 9.36% are urban inhabitants, a further 10,567 or 0.56% are pastoralists. The topography of West Hararghe is characterized by steep slopes in the highlands and midhighlands and large plains in the lowland areas. The highlands and mid-highlands are normally extensively cultivated but only partially protected by soil conservation structures and practices such as grass strips, alley cropping and bench terraces. The Zone is characterized by crop-livestock mixed farming system where livestock in general and dairy production in particular contribute signifi cantly to farmer livelihoods used as cash income generating purpose. Local cattle are the predominant breeds in the study area. Market oriented dairy production based on local dairy cows is also practiced. The study area has high production of milk potential in and selling to the district areas. The major crops grown in the area are food barley, wheat, sorghum and Khat production. Khat and Coffee is an important cash crop of this Zone. Over 50 square kilometers is planted with this crop of coffee and the climatic condition is conducive to livestock production.

Sample size and sampling technique
For this study, a representative sample was selected by using multi stage random sampling technique to select milk producer kebeles and sample milk producer households. In the fi rst stage from twelve milk producer districts, three of them were selected randomly. In the second stage nine kebeles are selected randomly based on probability proportional to size of kebeles in the districts and fi nally 140 producers were randomly selected from list of dairy producer based on probability proportional to size of milk producers households in the kebeles. Interviewed sample size is determined using a simplifi ed formula provided by [7] provided below.
The interviewers of sample traders were selected randomly counting from different markets based on different characteristics and 30 traders were selected from the three districts with probability proportional to size of the traders in the district.
Dairy product consumers were interviewed from Meiso, Chiro, Qunii, and Bedessa cities randomly. And a total of 30 households were selected from those cities. The total sample size of 200 was used for this study including producers, traders and consumers.
Where: n= sample size, N =Population size e=sampling error/ level of precision = 7% were used

Data type and methods of data collection
Both primary and secondary data types were used in the study under investigation. Primary data were collected using two types of structured questionnaire, one for milk producer farmers and the other for milk traders. A preliminary survey was also conducted through focus group discussions (guiding key informant's checklist) to obtain general information that could not be collected from dairy producers and traders of interviews. Primary data collected from households were focused on factors affecting milk market participation decisions, volume of marketed milk supply and general behavior of different value chain actors. Secondary data were also collected from journals, books, CSA, district reports and other organizations.

Methods of data analysis
Descriptive statistics such as such as means, percentage and total gross marketing are used to analyze demographic and socio-economic characteristics of smallholder milk producers. [8] defi nes market performance as to how well the market fulfi lls certain social and private objectives. This includes price levels and price stability in long and short term, profi t levels, cost, effi ciency and qualities and quantity of food commodities'. Because precise costs are frequently diffi cult to determine in many agricultural marketing chains for the reasons that costs are often cash and imputed, the Total Gross Marketing Margin (TGMM) is required to be calculated (Scott, 1995). It is expressed as a percentage of the difference between end buyer and fi rst seller prices (Mendoza, 1991).

Market performance analysis
Where, TGMM=Total Gross Marketing Margin The producer's margin is calculated as a difference: Where GMMp= the producer's share in consumer price or

Market participation equation
Models that include whether milk producing smallholder farmers participate in the market or not participate. In such circumstances, the probit model estimation is employed.
The Probit model is specifi ed as: Where Y 1i is the latent dependent variable which is not observed, Participated or not participated X 1i -is vectors that are assumed to affect the probability of sampled dairy household milk market participation. The model is specifi ed as: Y 2i is observed if and only if MMP = 1. The variance of u 1i is normalized to one because only MMP, not Y 1i is observed. The error terms, u 1i and u 2i , are assumed to be bivariat, normally distributed with correlation coeffi cient, ,  1 , and  2 are the parameter vectors. where: Y 2i is the observed dependent variable it is the volume of supply in the second-step X 2i is factors assumed to affect sale volume  2 is vector of unknown parameter in the supply equation u 2i is residuals in the supply equation that are independently and normally distributed with zero mean and constant variance.

Results and Discussion
Demographic and socio-economic characteristics of sample households Milk product movement: The primary objective of milk production for farmers in the district was for cash income generating and family consumption of dairy products. In  Actors, their roles, and relationships in milk value chain: The competitiveness of value chain is greatly infl uenced by the partnership and collaboration for innovation that can be realized by value chain actors [10]. Value chain actors include direct chain actors, which commercially involved in the chain and indirect actors, which provide fi nancial or non-fi nancial support services. The survey result shows that the dairy value chain actors in the study area are input suppliers, producers, collectors, rural wholesalers, urban wholesalers, retailers, cafes and ultimate consumers. Chain actors and their roles are described as follows. for whom to sell their produced milk products. Milk producers are the major actors who perform most of the value chain functions right from farm inputs preparation including buying of dairy cows or procurement of the inputs from other sources to handling and marketing. The major value chain functions that milk producers perform include feeding of their milk cows, cleaning of their milk cow barn, controlling and taking to veterinary services, harvesting or milking and transporting to purchasers of the product of different value chain actors. Majority of producers were selling their row milk to collectors and others of were also selling directly to consumers and retailers.
Milk producers of the study area mostly organized by local name /Ekub/, which is used to save time, to decrease transport cost and labour cost these is highly effective in Gemechis District. Those producers supply their milk to those members based on their order. The collected milk sells to their customers of collectors, retailers and for other consumers on average for the price of 21 Birr/ litre. In summer season, even if production of cow's milk and milk products increased, but the price were low the producer who gets because of high production of dairy products on the market. Because those traders set the purchase price and production cost of dairy cows were very high resulting into less benefi t from selling of this products. Feeding milk cow's price increasing, lack of modernized milk container to transport the market, and absence of dairy cooperatives were the main constraints in the study area. Shortages of grazing land and water in accessibility were problems, which affects milk participation decision and level of milk marketed. Even if, there were no extension and other training opportunities but the existence of well experienced farmers by their traditional and developed indigenous livestock management system, the rational power share that have women have in the milk production and marketing business in the study area was the main opportunity for different value chain actors. The current initiation done by the government to develop infrastructures relevant to livestock development and develop disseminate breed types was also the main opportunities for participants and non-participant of dairy producers.
Local collectors: These traders in assembly markets collect raw milk from milk producers in local kebeles milk markets for reselling it to rural wholesalers, café, retailers and consumers directly. They use their own fi nancial resources to buy milk from the surrounding area. They play important role for those producers to join them for other value chain actors in the milk value chain for the study areas. Collectors are the actors responsible for the trading of milk from production areas to other actors of different market. The value adding activities of collectors include buying, assembling and selling to different value chain actors.
Furthermore, local collectors stated that lack of testing quality like including water and other similar color products, adulteration were the major problems in their buying and selling activities. In addition, from the sample kebeles, the most severe problems were lack of transportation access, high transportation cost, lack of modernized milk transporting container, spoilage cost were the main constraints.
Rural wholesalers: Rural wholesalers are mainly involved in buying raw milk from merely collectors and selling to different value chain actors. These are suppliers of bulk of milk products to urban wholesalers, retailers and cafes after boiling and based on their customer demands fulfi lling their container of jerkin of three litres then transport to Chiro and Qunni, from Gemechis, and to Chiro and Bedessa from Oda Bultum and to Chiro from Mieso sometimes to Awash from Mieso. Survey result indicates that three of rural wholesalers were found in the district market. Chiro city is the main assembly center for milk from collectors in their surrounding rural kebeles of the three districts. They have high supply and quality milk products from those sample kebeles. Hence, for those kebeles of rural wholesalers have an opportunity to get high supply of milk from these two kebeles such as Kasheja and Kuyobo, which is a remote area, but high quality milk were supplied from these areas. Lack of support from governmental or nongovernmental organizations, high transport cost, inability to extend shelf life of milk products, poor transporting and packing material are among the major problems for those rural wholesalers.
Urban wholesalers: Urban wholesalers are those who reside in terminal market of Chiro, Bedessa and sometimes from Awash buy processed or value added and unprocessed milk from rural wholesalers of Gemmechis, OdaBultum and Mieso districts directly from other collectors. At present, urban wholesalers are major milk suppliers to retailers and cafes of Chiro and Bedesa because of transport cost, loading and unloading cost and other costs the price were very different from rural wholesalers and transporting vehicle transport were let so, they face a problem of deteriorate milk and they get loss of profi t like other traders. They have an opportunity of access of transportation as supplying milk from other markets of their customers and taking milk container again for them. This transportation system decreases their transportation cost, energy, time, and may enable them to involve in other economic income activities.
Retailers: Retailers involvement in the chain includes buying of milk directly from, producers, collectors, urban and rural wholesalers and selling directly to consumers. The retailers mostly whom they have shops and refrigerator have selling by value adding to other product form of yoghurt. Others sell without further processing. Directly consumers by value adding of yoghurt and further processing of boiling milk again to control quality. Retailers are key actors in dairy value chain in the district and town of Bedess and Chiro. They are the last link between producers and consumers other than cafes. Consumers usually buy the product from retailers as they offer according to requirement and purchasing power of the buyers. Retailers can be divided in to urban and rural. In the case of rural, retailers are based in village market and mainly purchase milk from farmers, and sell to consumers. Urban retailers purchase from urban wholesalers and sell to urban consumers. They have an opportunity because of high demand of dairy products in their area. They also have an opportunity as customers of urban wholesalers are there. Marketing margin can be used to measure the share of the fi nal selling price that is captured by a particular agent in the value chain. In order to calculate the marketing margin of an agent, the average price of milk for that particular agent was taken.
Marketing margins, associated costs and benefi t share of value chain actors and marketing margins through different main channels were presented bellow Table 2.

Econometric result
Determinants of milk market participation: The Heckman's procedure results for both outcome and selection variables are presented and discussed below. Moreover, it is important to check multicollinearity problem before running the model for continuous variables. VIF for the variables was in the ranges of 1.16-2.63. The result shows that multicollinearity was not a problem among the hypothesized variables (Appendix Table 1, 2) Table 3.
Age of household head: As hypothesized, it infl uenced milk market participation decision negatively at fi ve percent signifi cance level. The marginal effect revealed that when the household age increases by one year, the probability of participating in the milk market decreases by 2.56 percent. As realty aged households may have shortage of labour force to milk production and marketing activities those households are instead of selling and generating cash, they prefer consuming row milk. This study result resembles the fi ndings of [26].
Distance to the nearest market: It was hypothesized that it infl uences milk market participation decision negatively. This variable was found signifi cant to infl uence participation decision in milk market negatively at less than one percent signifi cant level negatively. This implies that as dairy producer residents distance increases by a kilometre from the milk market, the probability of participation decision to the market decreases by 7.5 percent. This may imply that milk product is perishable and hence it may not be marketed to long distance markets. This fi nding is consistent with that of [11].
Crop income: Previously it was hypothesized that crop income affect milk market participation positively. It infl uences Citation: Kebede  participation of milk market signifi cantly and negatively at less than ten percent signifi cance level. This is because if crop income of households is adequate for their consumption and economically enough, household do not sell their milk and milk products to the market this fi nding is similar with the fi nding of [3]. Since, the result revealed that as income from crop increases by one percent participation decision of milk to the market decreases by the predicted value of 30 percent.
Tropical livestock holding: This variable was found to infl uence the probability of participating in the milk market positively and signifi cantly at one percent signifi cance level. Livestock are important sources of cash in dairy households to purchase different input; this fi nding is consistent with the fi nding of [2,12]. The marginal effect revealed that when the TLU increases by one unit, the probability of participating in the milk market increase by 27.7 percent.
Determinants of level of participation in-farm level milk marketing: According to [9] a sample selection bias refers to the problems where the dependent variable is only observed for a restricted and non random sample.

Inverse mills ratio (LAMBDA):
The p-value of this variable is 0.012as presented in table below this implies that the correction for selectivity bias is highly signifi cant at less appropriate for this data and Mill's ratio corrects biased. Hence, this result suggests that there appears to be unobserved factors that might affect negatively both probability of dairy household market entry decision and marketable milk volume.
Land size: As expected, this variable infl uences milk market volume positively and signifi cantly at fi ve percent. The result shows that, the availability of land enables the owner to earn more agricultural output, which in turn increases the marketable supply. The implication is that as the farmer use more land for crop production more output is obtained and the residue also used for the feed of animals and also enables farmers to buy more cows and result in increases volume of milk production and this indirectly increases the volume of milk supplied to the market so the coeffi cient shows that as land size increases by one hectare leads to 25.87 litres increase in volume of milk marketed. This fi nding is consistent with that of [3,4].  Table 4.  Regarding the cost of the value chain actors, milk producers in the study areas incur costs mostly during production periods rather than marketing their produce. Producer's profi t margin was very low because of high production cost. As compared to each channel from all value chain, actor's cafes take the highest share of profi t of 8.49 Birr per liter. Retailers also shared the second profi t margin than other actors.

Tropical livestock unit:
The model result shows that, age, crop income, distance to the nearest milk market and total livestock unit infl uenced producers' market participation signifi cantly negatively and positively respectively. Moreover, the result indicated that marketed milk volume was also affected by improved dairy cows, land size and tropical livestock unit signifi cantly and positively. So governments and other stakeholders must improve the breeds of dairy cows to increase milk production.
Those farmers who have another adequate crop income generates but not know the value or the advantage of milk income benefi ts those farmers who must change their production system to milk products as because of demand of milk products increases. Other farm income of crop may be destroyed by natural damage and may be seasonal so, this production of dairying of damage is controlled, and owning of land is scarce from the district.
Distance to the nearest milk market infl uences marketed milk volume negatively. Because milk producers living were far from the milk market, rather than supplying in to the market consume directly. Separating butter and selling butter and consuming the other dairy product will be the best alternative advantageous. And also governments and those stakeholders