Livestock Research for Rural Development 19 (9) 2007 Guide for preparation of papers LRRD News

Citation of this paper

Determinants of smallholder dairy farmers' adoption of various milk marketing channels in Kenya highlands

L M Mburu, J W Wakhungu* and K W Gitu**

Ministry of Livestock and Fisheries Development, P.O. Box 47010-00100, Nairobi-Kenya
leonardmburu@yahoo.com
* University of Nairobi, Department of Animal Production, P.O. Box 29053-00625, Nairobi- Kenya
** University of Nairobi, Department of Agricultural Economics, P.O. Box 29053-00625, Nairobi- Kenya

Abstract

Understanding the factors affecting smallholder dairy farmers' adoption of various milk marketing channels is essential to implementation of dairy marketing liberalization policies in Kenya Highlands. Purposive multi stage using Probability Proportion to Size sampling design across different agro-ecological zones in Kenya highlands was used to evaluate the rationale underlying smallholders' milk marketing channel choice using Econometric Logit Models.

Logit models of milk marketing channels through itinerant traders (hawkers, neighbors and hotels) were non-significant (P > 0.05) but dairy cooperative was significant (P < 0.05). Eleven explanatory variables were significant (P < 0.05) in explaining farmers' adoption of milk marketing through the dairy cooperative channel. Leases land, average milk price [Kenya shillings (KES) /kg], total number of cow milked and farm acreage negatively influenced farmers' adoption of milk marketing through the dairy cooperative channel. Upper midlands, lower highlands, hired permanent labour, household head worked off-farm, average milk production per cow (kg / day), dairy cooperative as a source of animal production information, and availability of credit services had positive influence.

Farmers should be encouraged to undertake additional activities which stabilize household incomes to enable them adopt dairy technologies without exposing them to additional risk e.g. off-farm activities facilitate adoption of dairy technologies by the risk oversee farmers. Programs to improve and strengthen cooperatives can contribute to the development of dairy industry and substantially contribute to alleviating poverty.

Key words: Kenya highlands, marketing channels, smallholder dairy farmers


Introduction

Dairy production in Kenya is faced by a multitude of perceived and often experienced risks, which contribute to high costs of production and low average productivity (Muriuki et al 2003). These factors cause low profit to the producer and price fluctuations for the consumer.

The Kenyan dairy sector is made up of more than 600,000 smallholder dairy farms scattered around the country. These farmers account for 56% of the total milk production and 70% of the total marketed milk in the country (Omore et al 1999). Furthermore, livestock diversify production; provide year-round employment and spread of risk. Any factor that could lower or increase expenses is a source of risk to the economic performance of the dairy business (Bailey 2001). Some of these risks are: milk prices, purchased feed prices, hired labour, crop /forage production among others.

Most empirical studies using econometric models often relate the adoption decision to households and technological characteristics. Numerous studies have found that constraints imposed by these factors have discouraged technology adoption (Umali and Schwartz 1994; Nicholson et al 1999). These factors influence the awareness, availability, costs, benefits and risks associated with the different livestock technologies and management practices (Benin et al 2003).

Therefore, understanding the factors affecting the farmers' adoption of various milk marketing channels is critical to success of development and implementation of policies and programs in dairy industry marketing liberalization. But surprisingly little work has been done to examine the determinants of adoption of any milk marketing channel hence the objective of this study.
 

Materials and methods

Description of study site

The study was carried out in Kiambu district, Central province located in Kenya highlands from December 2004 to March 2005. The district occupies 1323.9 km2 with a population density of 562 persons per km2 with 189,706 households (CBS 2001). The Kenya highlands comprise areas with altitude 1200-2550 meters, annual mean temperatures of 13.40C to 21.90C. The rainfall is bimodal varying from 600-1200 mm per year depending on location and altitude. Fertile soils here have good potential for biomass production and intensively cultivated and food cropped 1.4-1.7 times per year (Jaetzold and Schmidt 1983). Dairy cattle's farming in the district includes the intensive (zero grazing), semi-intensive and extensive grazing production systems.

Data collection

The study used conceptual framework for dairy systems analysis of production-to-consumption approach developed by ILRI (Rey et al 1993). Primary data were collected through personal interviews by trained enumerators using a survey questionnaire covering measures from resources to parameters reflecting farm functioning from respondents with at least one dairy cow at the time of survey. All information collected referred to the situation of the day before the survey. The data from questionnaires was entered into Statistical Program for Social Scientists (SPSS) from SPSS Inc.

Sampling procedure

Purposive multi stage using Probability Proportion to Size sampling design was used. Three agro-ecological zones: Upper midlands, Lower highlands and Lower midlands were chosen purposively. Within the agro-ecological zones, eight research locations were selected based on household density: low, medium and high. Locations with a higher population size (CBS 2001) had a proportionately higher sample size in the survey. In order to capture as much local variations as possible, the sample in each zone was spread across the 27 sub-locations (smallest administrative unit of a district in Kenya) among farms selected as randomly as possible. In some, if the farmer could not be reached or did not wish to participate in the study, another one in the locality was substituted.

The sample size was obtained from estimating the number of observations potentially needed to distinguish between the three agro-ecological zones by a difference of 30% in some of the important farm/household variables. Assuming a desired confidence interval of 95%, and using a coefficient of variation of 68%, which was the observed co-efficient of variation of households in Kiambu dairy herd from previous studies (Kaguongo et al 1997); a minimum sample size of 40 in each agro-ecological zone was calculated. The calculation of sample size in each stratification class, to estimate a difference, was based on the equation (Poate and Daplyn 1993):

Where:

n = minimum sample size,

z = 1.96 for 95% confidence interval,
c = Coefficient of Variation,
d = Level of difference.

The chosen sample required then 14 observations in each location. After maintaining a minimum of 10 observations in each location, the total sample size obtained was 134 households (or 0.07 % of the households in Kiambu district).

Econometric models

The two models used in adoption studies are the logit and probit models both of which have a dependent variable bound between 0 and 1 and are convenient for dichotomous adoption variables. Probit model is particularly well suited to experimental data while logit model is for observational data (Rahm and Huffman 1984).

Logit models provide empirical estimates of how changes in exogenous variables influence the probability of adoption of any technology. The results of the logit model estimates are reported as the marginal effects of a change in the exogenous variables, that is, the change in the probability of choice due to a one-unit change in the exogenous variable. In the case of dummy variables (i.e. 0 or 1) such as households buying fodder, the marginal effect is the difference in probability due to belonging to one group rather than the other (e.g. households buying fodder versus households not buying fodder). For continuous variables such as the age of household head, the marginal effect is the change in probability due to an increase in one year in age. The impact of other categorical and continuous variables can be interpreted analogously. The magnitude, statistical significance and the signs (i.e. positive or negative influence on probability of choice of milk marketing channel) of the marginal effects are typically of most interest in evaluating the factors influencing the probability of adoption of any technology.

Farmers' socio-economic variables

The factors affecting milk marketing through the dairy cooperatives related to location, farmers resources (human and physical) and institutional factors (Table 1).


Table 1.  Mean descriptive statistics for variables in the econometric model for milk marketing channels

Explanatory variables 

Lower highlands

Upper
midlands

Lower  midlands

Overall

Location

 

 

 

 

Upper midlands, 1* = Yes, 0* = No

0

1

0

0.335

Lower highlands, 1* = Yes, 0*= No

1

0

0

0.358

Farmers resources (human and physical factors)

 

 

 

Household head age, Years

54.1

53.9

50.6

52.9

Household head education, 1= Post primary, 0= Primary

0.416

0.311

0.561

0.425

Household head works off-farm, 1 *Yes, 0*=No

0.333

0.644

0.39

0.455

Leases land, 1*= Yes, 0*= No

0.291

0.422

0.073

0.268

Number of farms cultivated

1.8

2

1.2

1.7

Total farm acreage, acres

3.05

1.6

3.45

2.65

Hires permanent labour, 1*= Yes, 0 *= No

0.333

0.067

0.341

0.246

Distance from farm to nearest market, Km

2

2.06

2.45

2.2

Number of cows milked

1.8

1.4

1.5

1.6

Average price of milk, KES /kg

17.5

19.3

17.9

18.2

Average milk production, kg/day/cow

10.25

7.65

8.25

8.75

Institutional factors

 

 

 

 

Credit services available, 1*= Yes, 0 *= No

0.604

0.266

0.609

0.492

Member of farmers’ group, 1*= Yes, 0*= No

0.687

0.844

0.829

0.783

Government extension agent, 1* = Yes, 0* = No

0.354

0.155

0.561

0.35

Sales agents, 1*= Yes, 0*= No

0.375

0.067

0.487

0.306

Dairy cooperatives information, 1*= Yes, *=No

0.833

0.089

0.487

0.477

* The dummy explanatory variables take the value of 1 if the farmer currently uses or had adopted the management strategy and 0, otherwise.


The dummy explanatory variables take the value of 1 if the farmer currently uses or had adopted the management strategy and 0, otherwise. The dependent variables index if the farmer has adopted of any milk marketing channel. The variable takes the value of 1 if the farmer currently uses or had adopted the channel and 0, otherwise.
 

Results

Determinants of milk marketing through the dairy cooperatives

The number of households using various milk marketing channels varied across the three survey sites respectively (Table 1). The econometric model was fitted with 18 variables. Logit models of milk marketing channels through itinerant traders (hawkers, neighbors and hotels) were non-significant (P > 0.05) but dairy cooperative was significant (P < 0.05). Eleven explanatory variables were significant (P < 0.05) in explaining farmers' adoption of milk marketing through the dairy cooperative channel. Leases land, average milk price (KES /kg), total number of cows milked and farm acreage negatively influenced farmers' adoption of milk marketing through the dairy cooperative channel. Upper midlands, lower highlands, hired permanent labour, household head worked off-farm, average milk production (kg / day/ cow), dairy cooperative as a source of animal production information, and availability of credit services had positive influence (Table 2).


Table 2.  Logit model of factors affecting farmers’ milk marketing through dairy cooperative channel in Kenya highlands

Explanatory variables

Parameter estimate ± Standard error

Intercept

3.99 ± 3.05

Upper midlands*

0.0741 ± 0.531**

Lower highlands*

1.36 ± 0.395**

Household head age, Years

-0.00531 ± 0.0104

Household head education level, 1= Post primary, 0= Primary    

-0.121 ± 0.265

Household head worked off-farm   *

0.0979 ± 0.301*

Leases land*

-0.568 ± 0.391*

Number of farms

0.319 ± 0.209

Total farm size, acres

-0.0276 ± 0.0629*

Hires permanent labour*

0.098 ± 0.348**

Distance to nearest market, Km    

0.112 ± 0.0742

Number of cows milked

-0.0212 ± 0.19**

Milk production, kg/cow/ day       

0.0325 ± 0.041**

Average milk price, KES /kg

-0.583 ± 0.173 **

Credit services available*

0.376 ± 0.307 **

Government extension agent*

-0.112 ± 0.356

Member of a farmers’ group*

0.0238 ± 0.298

Sales agent   *                                  

0.158 ± 0.291

Dairy cooperatives*

0.409 ± 0.413**

* The dummy explanatory variable takes the value of 1 if the farmer currently uses or had adopted the management strategy and 0, otherwise.

**Correlation significant at the 0.01, * Correlation significant at the 0.05 level.
Pearson Goodness-of-Fit Chi Square
= 559, DF = 115, P = 0.00

Discussion

Location factors

Upper midlands and lower highlands were significant and positively influenced the marketing of milk through the dairy cooperative channel. However, the magnitudes varied (Table 2). Households in upper midlands were less likely (7.4%) to market their milk through the dairy cooperatives than those in other areas due to its close proximity to Nairobi city (Staal et al 1998). In lower highlands, households were more likely to market their milk through the cooperatives than those in upper midlands probably due to lack of alternative competitive informal markets.

Farmers' resources (human and physical)

Total farm acreage influenced negatively the marketing of milk through the dairy cooperative channel (Table 2). This suggested that the probability of milk marketing through the cooperatives decreased as the total land size increased, although the effect of an additional acre of land was relatively small (2.75%). Large land sizes characterized old dairy farmers who had no alternative sources of income and thus the need for daily cash from milk sales through itinerant traders.

The probability of milk marketing through the dairy cooperatives increased if household head worked off-farm though marginally (Table 2). This was a reflection of employed farmers' increased interest in cooperative market. Off farm employment increased farmers exposure to opportunities for extra daily cash hence disposal of milk through cooperatives which paid monthly for milk delivered.

Hired permanent labour influenced positively the marketing of milk through the dairy cooperative channel (Table 2). This suggested that the probability of milk marketing through the cooperatives increased if household hired permanent labour. Perhaps permanent labour employment led to more efficient utilization of resources and hence more milk production. The cooperative market remained the only reliable channel for such excess milk. Employment generation through small-scale dairy marketing and processing depending on enterprise creates 2.0-0.3 direct and indirect jobs for every 100 liters of milk traded (Omore et al 2001).

Leasing of land influenced negatively the marketing of milk through the dairy cooperative channel (Table 2). This suggested that the probability of milk marketing through the cooperatives decreased with household leasing land. In Kiambu district 27% of households leased land (Table 1). Such households were involved in multiple tenure arrangements as they cultivated land belonging to someone else. Such parcels of land were not likely to near the homestead, hence requiring daily cash to travel to, fetch fodder from or carry out other agronomic practices like weeding. On these farms different crops either in pure stands or various combinations are grown. This was a risk management strategy, which leads to land fragmentation.

The number of cows milked and average milk production per cow (kg /day) influenced marketing of milk through the dairy cooperative channel negatively and positively respectively (Table 2). This suggested that the probability of milk marketing through the cooperatives decreased marginally (2.1%) with increase in the number of cows milked and increased marginally (3.25%) with milk yield per cow per day. Farmers selling milk to cooperatives are likely to have excess milk. The additional milk produced required a reliable market outlet that was only offered by the cooperatives.

There was a negative relationship between the average milk price (KES /kg) and the marketing of milk through dairy cooperative channel (Table 2). This suggested that the probability of milk marketing through the cooperatives increased with decrease in milk price. Perhaps unlike other channels that imposed milk delivery quotas during times of milk glut, cooperatives did not but offered lower prices. This degree of volatility made it difficult to plan cash flow needs for the dairy enterprise. Dairy farmers need to budget each month for feed purchases, hired labour and veterinary and artificial insemination expenses and any other expense (Bailey 2001). Cash flow problems occur when milk prices fall below expected levels.

Institutional factors

Adoption of milk marketing through the cooperative channel was influenced positively by credit availability (Table 2) suggesting that the probability of milk marketing through the cooperatives increased with ease of credit availability. It can be assumed that farmers who wanted credit from cooperatives were more likely to sell their milk there to improve their credit rating. Unlike banks that required security inform of title deeds, vehicle logbooks or quoted public company shares, cooperatives decided milk delivery track record as security.

Dairy cooperative influenced positively the adoption of milk marketing through the dairy cooperative channel (Table 2). Therefore, farmers marketing their milk through the cooperative were likely to be more knowledgeable than other farmers using other market channels. The dairy cooperatives thus were not only marketing channels, but also a significant source of other market information for farmers particularly with regard to concentrates, veterinary clinical drugs, and artificial insemination services and forage seeds. Consequently, they determined in many ways what breed of cattle should farmers keep and type of concentrates to feed in response to market demand. Cooperatives can thus unwittingly contribute to the failure or success of dairy industry.

Other policy related intervention like government extension agent as a source of extension information had relatively small negative (11.2%) and insignificant effects on marketing of milk through the dairy cooperative channel in Kiambu district. This implied that extension was playing a negative role in cooperatives marketing. Extension agents usually urged farmers to sell their milk to the highest bidder. Given the low milk prices offered by cooperatives, it was not one of their preferred options.

Others like membership to agricultural farmers' group had marginal positive influence (2.4%). Membership to farmers associations and saving societies has helped farmers to participate in trainings and agricultural events, which have formed a major source of knowledge and skills applied in the farm (Gichinga and Maluvu 2003).
 

Conclusions

Acknowledgements

The authors are grateful to Deutscher Akademischer Austausch Dienst (DAAD) for their research funds to run this study. Thanks also due to all enumerators and farmers whoparticipated in this study.
 

References

Bailey K 2001 The fundamentals of forward contracting, hedging and options for dairy producers in the North East. Staff Paper #338, June 5, 2001. College of Agricultural Sciences,The Pennsylvannia State University, U.S.A. http://agmarketing.extension.psu.edu/Commodity/PDFs/DryRiskMgt.pdf.

Benin S, Pender J and Ehui S (Editors) 2003 Policies for sustainable land management in the East African highlands. Summary of Papers and Proceedings of a Conference held at the United Nations Economic Commission for Africa (UNECA), Addis Ababa, Ethiopia, 24-26 April 2002. Socio-economics and Policy Research Working Paper 50. International Livestock Research Institute (ILRI), Nairobi, Kenya. Pp 90-95. http://www.ifpri.org/divs/eptd/ws/papers/eptws13.pdf

CBS 2001 The 1999 Population and Housing Census: Counting Our People for Development. Volume 1, 2001. Ministry of Planning and National Development, Central Bureau of Statistics, Nairobi, Kenya, Pp 1-5.

Gichinga L W and Maluvu J M 2003 Organic resource management in Kenya: Perspectives and Guidelines. Forum for Organic Resource Management and Technologies, Nairobi, Kenya. Pp 18.

Jaetzold R and Schmidt H 1983Natural conditions and farm management information: Farm management Hand book of Kenya, Volume II, Part B, Central Kenya. Ministry of Agriculture, Nairobi, Kenya Pp 510.

Kaguongo W N W, Staal S and Ackello- Ogutu C 1997Smallholder adoption of dairy technologies: Effect of risk with management strategies in a liberalizing economy. In: Fungoh P O and Mbadi G C O (Editors) Focus on Agricultural Research for Sustainable Development in a Changing Environment: Proceedings of the 5th Kari Scientific Conference, Nairobi, Kenya. Pp 243-251.

Muriuki H, Omore A, Hooton N, Waithaka M M, Ouma M, Staal S J and Odhiambo P 2003 The policy environment in the Kenya dairy sub-sector: A review. SDP Research and Development Report No. 2, Smallholder Dairy Research and Development (R and D) Project, Nairobi, Kenya, Pp 21. http://www.smallholderdairy.org/collaborative%20res%20reports.htm

Nicholson C F, Thornton P K, Mohammed L, Minge R W, Mwamwchi D M, Elbasha E H, Staal S J and Thorpe W 1999 Smallholder dairy technology: An Adoption and impact study. ILRI (International Livestock Research Institute), Nairobi, Kenya. Pp 51. http://www.ilri.org/InfoServ/Webpub/Fulldocs/CoastImp/Toc.htm

Omore A, Cheng'ole J M, Islam F, Nurah G, Khan M I, Staal S J and Dugdill B T 2001 Employment generation through small scale dairy marketing and processing: Experiences in Kenya, Bangladesh and Ghana. A joint study by ILRI Market-oriented Smallholder Project and FAO Animal Production Division, Rome, Pp 31. http://www.fao.org/docrep/007/y4860e/y4860e00.htm

Omore A, Muriuki H, Kenyanjui M, Owango M and Staal S J 1999 The Kenya Dairy sub-sector: A rapid appraisal. MoA/KARI/ILRI Smallholder (Research and Development) Project Report, Ministry of Agriculture (Kenya), Kenya Agricultural Research Institute (KARI) and International Livestock Research Institute (ILRI), Nairobi, Kenya. 51 pp. http://www.smallholderdairy.org/collaborative%20res%20reports.htm

Poate C D and Daplyn P F 1993 Data for agrarian development. Wye Studies in Agricultural and Rural Development. Cambridge: Cambridge University Press, U.K.

Rahm M R and Huffman W E 1984The adoption of reduced tillage: The role of Human capital and other variables. American Journal of Agricultural Economics: 405- 413.

Rey B, Thorpe W, Smith J, Osuji P, Mullins G and Agyemang K 1993 Improvement of dairy production to satisfy consumer demand in sub-Saharan Africa: A Conceptual Framework for Research. Market- oriented Smallholder Dairy Research Working Document No. 1. International Livestock Centre for Africa (ILCA), Addis Ababa, Ethiopia, pp 13. http://www.fao.org/wairdocs/ilri/x5465e/x5465e00.htm

Staal S J, Chege L, Kenyanjui M, Kimari A, Lukuyu B, Njubi D, Owango M, Tanner J, Thorpe W and Wambugu M 1998 Characterization of Dairy Systems Supplying the Nairobi market, KARI/ MoA /ILRI Collaborative Research Project Report, Nairobi, Kenya, 69 pp.

Umali D L and Schwartz L 1994 Public and private extension: Beyond traditional Frontiers. World Bank Discussion Papers, World Bank, Washington D.C., U.S.A. Pp 15-27. http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/1994/04/01/000009265_3970311123758/Rendered/PDF/multi_page.pdf



Received 2 March 2007; Accepted 21 March 2007; Published 5 September 2007

Go to top