Marketing system and channels for scavenging local chickens in Lilongwe, Malawi

T N Gondwe*, C B A Wollny* and W Kaumbata

Department of Animal Science, Bunda College of Agriculture, University of Malawi, P.O. Box 219, Lilongwe, Malawi;
^*Institute of Animal Breeding and Genetics, Animal Breeding and Production in the Tropics, Georg - August Universität Göttingen, Kellnerweg 6, 37077 Göttingen, Germany.
tgondwe@gwdg.de  or   tgondwe@bunda.sdnp.org.mw

Abstract

A study was conducted to identify marketing structure, players, prices and profit margins of local chickens in rural and urban Lilongwe, Malawi. One rural market (Mitundu) and two urban markets (Lilongwe and Kawale) were visited three times during the dry season and two times during the wet season from 2002 to 2003.

Male middlemen sold both male and female local chickens of all phenotypes at the three markets. Middlemen at Mitundu bought their chickens from farmers and trading centres in surrounding villages and from Mozambique. Middlemen at urban markets bought chickens from rural and district markets and used public transport for the chickens. Purchasing, selling prices and profit margins were higher for urban markets than for the rural market. Selling prices for all markets were higher during the dry season. Profit margins at urban markets were higher during the wet than the dry season. Chickens sold during the dry season were heavier (1.50 kg ± 0.03) than during the wet (1.38kg ± 0.03) season. Live weights positively influenced pricing and profit margins.

It is concluded that a marketing chain exists for local chickens. Farmers transact in form of cash and barter at village level, whereas afterwards male middlemen control the market.

Keywords: local chicken, marketing chain, profit margins, urban and rural market

Introduction

Selling of local chickens is one of the functions of keeping free-range chickens observed during surveys and monitoring studies (Gondwe et al 1999; Gondwe et al 2003). The cash from sales is used to buy household needs including food, to improve food security at household level (Kyvsgaard et al 1999; Kondombo et al 2003). Some farmers barter their free-range local chickens for food and household items. Missohou et al (2002) reported that in Senegal, farmers exchanged six local chickens for one goat. Regardless of the mode of sales, this function ranks among the top three most important roles (food, income and socio-cultural) that are played by local chickens for the wellbeing of the household and community (Dessie and Ogle 2001; Mwalusanya et al 2001; Ekue et al 2002).

Marketing channels for local chickens include selling of chickens and eggs at households within the villages, on road sides, during entertainment ceremonies and even in local and city markets (Safalaoh 1997; Ekue et al 2002; Missohou et al 2002). The market channels are described as informal and poorly developed (Branckaert and Guèye 1999; Mlozi et al 2003). On the other hand, free-ranging local chickens are claimed to be on demand and fetch high market prices in urban markets of Malawi, Nicaragua and many developing countries in Africa and Asia due to preferred attributes such as being tastier than improved broiler strains (Aini 1990, Safalaoh 1997; Kyvsgaard et al 1999; Branckaert and Guèye 1999).

Analysis of the marketing system for free-range chickens will help to determine the economic value and importance of local chickens. This information is required to characterise, conserve and develop the poultry genetic resource and to justify resource allocation to rural poultry improvement and conservation projects. Branckaert and Guèye (1999) reported that an established market structure for free-range chickens is a prerequisite for developing family poultry. Even in breeding program development, indices require appropriate economic values that could be derived from such market studies. The current study focused on the existing marketing structure of free-range chickens surrounding the catchments where community participatory village poultry studies were initiated (Gondwe et al 2003). The objectives were to: (i) identify marketing of free-range chickens in local and urban markets; (ii) analyse the marketing chains and players; (iii) determine market prices and estimate profit margins.

Materials and Methods

Choice of markets of study

The study was conducted at three markets, Mitundu, Lilongwe and Kawale between 14 July and 23 August 2002 (during dry season) and between 21 January and 5 February 2003 (wet season). The choice of these markets was to cover catchments of Mitundu and Mkwinda Extension Planning Areas (EPAs) of Lilongwe Agricultural Development Division (LADD) where monitoring studies on growth performance and constraints of free-range chickens were conducted (August 2002 to August 2003) assuming that at least a sample of the monitored chickens are sold through one of these markets. Mitundu is a local (rural) market open for sales of various food and non-food products. This market is within 15 km of  the furthest villages of Mitundu and Mkwinda EPAs. Most smallholder farmers in the area sell and buy their products from this market. The market is approximately 55 km distant from the Mozambique border. Informal cross border trade, especially of foodstuffs, is taking place. Lilongwe market is located in the centre of the city and the distance to the Mkwinda EPA is about 35 km. Kawale market is located on the outskirt of Lilongwe city, five km from Lilongwe market. Both Lilongwe and Kawale markets serve urban communities.

Study protocol

The markets were visited in the morning at between 8:00 and 10:00 a.m. once per week. Each market was visited three times during  the dry season and two times during thewet season. At the markets, locations where live free-range chickens were sold were identified with the help of market officials. After introductions, those selling chickens were briefed on the purpose of the visit. These sellers were asked questions written on a structured questionnaire. Information collected by asking or through observation included: (i) demography of the sellers in terms of their names, gender and education; (ii) number of chickens each seller had for sale; (iii) phenotypes and sex; (iv) sources of chickens; (v) selling price; (vi) purchasing price; (vii) mode of transport.

After requesting permission from the sellers, chickens were individually weighed using a digital scale (CA Kern, 5K5; Kern & Sohn GmbH, Germany) and prices, sex and phenotype were recorded. Based on interviews with traders, transport costs were recorded during the rainy season only. Market levies were fixed at MK5.00 (1 US$ = MK85.00) per seller per day. In total, 42 middlemen were interviewed during the study. Of the observed sales, 53 % were during the dry season; 47% during the wet season.

Price portfolios and profit margins

Since prices were pegged per live bird, these were converted into price per unit live weight for each chicken by dividing price by live weight. Transport costs and levies were considered as transaction costs. Mortality of chickens during transport was not reported by sellers and was, therefore, not included in the analysis as a form of loss. Sellers' purchase price of chickens at farmer level was taken as farm-gate price. Profit margins were calculated by subtracting purchasing prices and transaction costs from the selling price.

Data analyses

Quantitative data were subjected to analysis of variance using the general linear model (GLM) procedure of the Statistical Analysis System (SAS 1999). Qualitative data were analysed for descriptive statistics using frequency procedures and cross-tabulation (SAS 1999). Appropriate statistical tests were applied to see the effect of factors such as season, phenotype, market and sex of chickens, that may be associated with marketing of local chickens. The following linear model was used during analysis of quantitative data

y_ijklm = µ + m_i + s_j + p_k + ß_ijkl + ε_ijklm

Where:

y_ijklm is the market parameter (price, weights) estimate for bird m on market i

µ is the overall mean

m_i is the fixed effect of market (i = 1,2,3)

s_j is the fixed effect of season (j = 1,2)

p_k is the fixed effect of phenotype (k = 1,2,3,4,5,6)

ß_ijkl is the fixed effect of sex of bird (l = 1,2)

ε_ijklm is the residual error

Results

Market channels and players in local chicken marketing

Local chickens were sold at the three markets. In addition, live goats were also sold at Mitundu market, and sometimes at Kawale. Chickens were the only species sold alive at Lilongwe market. At Kawale and Lilongwe markets, live broiler and culled layer chickens were also sold. All other livestock were sold as meat at the three markets. Presence of local chickens showed that existing commodity markets are used to sell the chickens.

Only male middlemen were involved in selling local chickens (Table 1).  At Kawale market, the majority of middlemen had attained secondary education. The educational level of the middlemen was highly variable ranging from illiterate to secondary school and differed widely by market.

Number of chickens available for sale, their live weights and prices

Mean numbers of chickens sold per middleman per day on each market are shown in Table 2 while their live weights; purchasing and selling prices by season are presented in Table 3. Local chicken market prices per chicken at Lilongwe and Kawale were significantly (p<0.05) higher than prices at Mitundu. Kawale offered highest prices per live chicken. Seasonal selling price differences were significant (p<0.05) for Mitundu market only. For all markets, dry season prices for chickens were higher than wet season prices. Selling prices per kg followed the trend of live chicken prices during the dry season.

Sellers at Mitundu market bought their chickens at significantly lower purchasing prices than those who were selling at Lilongwe and Kawale. Wet season purchasing prices were significantly (p<0.05) lower than dry season prices.

Live weights did not differ significantly (p>0.05) between markets during the dry season. During the wet season, chickens found at Kawale market were significantly heavier (p<0.05) than those at Lilongwe and Mitundu markets. Chickens sold at Mitundu were significantly heavier (p<0.05) during dry season than wet season.

Sources and demography of chickens sold

Middlemen bought their chickens from 22 different sources (Table 4). It was found that these sellers bought chickens from different places. Buying appears to follow a pattern specific for each market. For example, common sources for chickens sold at Mitundu came from trading centres that surround the market. These are places within rural areas servicing village markets for rural communities that comprise several neighbouring villages.

For Mitundu, other chickens came from villages across the border in Mozambique. Chickens sold at Lilongwe and Kawale markets came mostly from local rural markets. Other sources were reported just as districts. The districts and other sources mentioned are far from the markets (Figure 1).

Demographic characteristics of chickens sold at the markets included description of their sex and phenotypes (Tab le 5).  Both sexes of chickens were sold at the markets. More females than males were available for sale at Mitundu and Lilongwe markets. All common phenotypes for local chickens were sold at the markets and included Yakuda, Kawangi, Mawanga, Yofiira and Yoyera. Chiphulutsa was least available in all markets. Kansilanga, Kameta and Kachibudu were only found at Lilongwe and Kawale markets.

Observations from Mitundu and Mkwinda EPAs showed that farmers sold chickens to fellow farmers and middlemen, either directly for cash or through exchange with household items. Female members of households dominated in selling chickens at farm level, especially where sales were through barter. Bartering took place at households especially between farmers and middlemen. Farmers themselves also sold their chickens at trading centres. Participation of farmers in marketing of free-range chickens is at their households and at trading centres. Middlemen control the rest of the channel.

Profit margins of the middlemen

Profit margins before and after transaction costs (TC, Table 6) at different markets are presented per live chicken and per unit of live weight (Table 7).

Significant differences and trends observed in the profit margins follow the pattern observed for prices (Table 3). Seasonal differences in profit margins before TC were significant (p<0.05) for all markets. Both margins per live chicken and per unit weight were higher in wet than dry seasons for Lilongwe and Kawale. The reverse was observed for Mitundu market.

Effects of sex, phenotype and live weights of chickens on price and margins

Effects of sex of chickens (Table 8) showed that male chickens had significantly (p<0.05) higher purchasing and selling prices, and profit margins per live chicken than female chickens. But when these were compared on per unit live weight basis, female chickens had higher values than male chickens (p<0.05).

Pearson correlation coefficients were negative between live weight of chickens and selling price per unit weight (r = -0.45) and between live weight and purchasing price per unit weight (r = -0.31). Phenotype did not have significant effects on pricing and margins for local chickens. The effect was observed on numbers of each phenotype available for sale.

Live weights of local chickens positively influenced purchasing and selling prices but not profit margins. Pearson correlation coefficients were 0.76 between live weight and purchasing price, 0.74 between live weight and selling price and 0.29 between live weight and profit margins (p<0.05, n = 468). Correlation coefficient between selling price and profit margin were 0.59 and between purchasing price and profit margins 0.25.

Discussion

Marketing of local chickens

The results show a connected marketing structure that transfers free-range chickens from producers in rural households to consumers in both rural and urban areas. Middlemen operating in the marketing of free-range chickens were also reported in Tanzania (Mlozi et al 2003), Ethiopia (Dessie and Ogle 2001) and other countries in Africa (Kitalyi 1998). Dessie and Ogle (2001) observed that farmers could sell their chickens directly at local markets; thereafter, middlemen took over. Though Mlozi et al (2003) reported presence of women selling processed chickens at the markets, these were not farmers but vendors. The players observed in the study, therefore, agree with those reported in other countries. The market is informal at household level and village trading centres where farmers participate and transactions through barter are  also common. After that, middlemen operate local chicken marketing as their main business occupation.

The seasonal trends on numbers available in the markets, pricing and profit margins differed between the rural and the urban markets. Ekue et al (2002) reported that farmers sell chickens to meet their household needs especially food during the wet season. During the dry season, farmers sell their chickens instead of losing them through death from Newcastle disease (Safalaoh 1997; Dessie and Ogle 2001; Mlozi et al 2003). These factors operate at household (farm) level, influence bargaining power of farmers when selling their chickens (Mlozi et al 2003), and contribute to the seasonal variations. Supply and pricing of chickens at a local market are bound to be affected, reflecting what was observed at the rural market. Urban markets counteract these forces by diversifying sources of their chickens depending on market information. The different sources of chickens supplied to urban markets also provided opportunity for more phenotypes observed at these markets than those sold at the rural market.

Transaction costs and profit margins

Transaction costs per bird per middleman were a fixed cost related to size of market (Figure 3). These TCs decline at a disproportional rate as market size increases.

The farmers sold less than two chickens per month and this would mean high TCs if individually transported and offered at the market. High TCs associated with fewer chickens sold are the possible constraint of farmers from selling their chickens at the rural markets. Middlemen at the rural markets, such as Mitundu,  minimised TCs by using low (or no) cost form of transport such as bicycles or by just walking to the market. Middlemen in urban markets used public transport to reach distant local market places, so their market sizes had to be large enough and their prices had to absorb TCs. Risks of middlemen purchasing sick birds and therefore, dying before being sold need to be investigated in future studies.

The higher positive profit margins observed in urban markets than in rural market are comparable to other countries. In Tanzanian markets, middlemen generated 65 % profit margins above what farmers got (Mlozi et al 2003). Kitalyi (1998) reported that household level prices for chickens were one third that offered at urban markets, also reflecting higher profit margins for middlemen. Most likely, seasonal food shortage results in weak bargaining power of farmers.

Effect of sex, phenotype and live weights

The current system of the live chicken markets attracts higher profit margins for male than female birds but not per unit of live weight. The influence of phenotype on number of chickens sold, just like the sex effect, merely reflected the relative prevalence of different phenotypes among flocks in rural households (Gondwe et al 1999). All phenotypes were available for sale, hence improving all free-range chicken genetic resources have marketing justification. On per live chicken basis, pricing of free-range chickens was strongly and positively influenced by size of chickens. Sellers do not use weighing scales but estimate weight of chicken by assessing its size and by handling the chicken. The higher positive correlation between selling price and profit margin (r = 0.59) than between purchasing price and profit margin (r = 0.25) shows that selling price determines the profit margins realised.

Conclusions

• This study has found that markets for local chickens exist in both rural and urban areas.

• Both sexes and varied phenotypes are sold.

• Marketing at village level takes place through cash and direct barter transaction. Thereafter, marketing of local chickens is in cash and is controlled by middlemen.

• Farmers sell chickens to obtain household needs, but middlemen operate to make profits and reduce transaction costs.

• The study has established prices, transaction costs and profit margins of the market chain from producer to consumer of local chickens at different levels. This information could be utilised in valuing the local chicken genetic resources, guiding production and marketing management interventions as well as in developing breeding programs.

Acknowledgement

The Norwegian Agency for International Development (NORAD) is appreciated for financial support to the present study, and through their support to Bunda College of Agriculture, University of Malawi. City officials and middlemen are acknowledged for their assistance and cooperation during the study.

References

Aini I 1990 Indigenous chicken production in South - East Asia. World's Poultry Science Journal, 46: 51 - 57.

Branckaert R D S and Guèye E F 1999 FAO's programme for support to family poultry production. In Poultry as a Tool in poverty Eradication and Promotion of Gender Equality - Proceedings of a Workshop. http://www.husdyr.kvl.dk/htm/php/tune99/24-Branckaert.htm

Dessie T and Ogle B 2001 Village poultry production systems in the Central Highlands of Ethiopia. Tropical Animal Health and Production, 33: 521 - 537.

 

Ekue F N, Pone K D, Mafeni M J, Nfi A N and Njoya J 2002 Survey of the traditional poultry production system in the Bamenda area, Cameroon. In Characteristics and Parameters of Family Poultry Production in Africa. FAO/IAEA Co-ordinated Research Programme on Assessment of the effectiveness of vaccination strategies against Newcastle disease and Gumboro disease using immunoassay-based technologies for increasing farmyard poultry production in Africa. IAEA Vienna. http://www.iaea.or.at/programmes/nafa/d3/public/2-survey-ekue.pdf

 

Gondwe T N P, Ambali A J D, Chilera F C, Lwesya H and Wollny C 1999 Rural poultry biodiversity in Lilongwe and Mzuzu Agricultural Development Divisions (ADD), Malawi. Malawi Journal of Science and Technology, 5:17 - 25.

 

Gondwe T N P, Wollny C B A, Safalaoh A C L, Chilera F C and Chagunda M G G 2003 Community - based promotion of rural poultry diversity, management, utilisation and research in Malawi. In: FAO Community - Based Management of Animal Genetic Resources. Proceedings of the workshop held in Mbabane, Swaziland, 7 - 11 May 2001. pp 69 - 76.

 

Kitalyi A J 1998 Village chicken production in rural Africa, household food security and gender issues. Animal Production and Health Paper No. 142. Food and Agricultural Organisation of the United Nations, Rome.

 

Kondombo S R, Nianogo A J, Kwakkel R P, Udo H M Y and Slingerland M 2003 Comparative analysis of village chicken production in two farming systems in Burkina Faso. Tropical Animal Health and Production, 35: 563 - 574.

 

Kyvsgaard N C, Luna L A and Nansen P 1999 Analysis of a traditional gain and scavenge based poultry system in Nicaragua. In: Proceedings of a Workshop on Poultry as a Tool in Poverty Eradication and Promotion of Gender Equality (Dolberg F and Petersen P H, Editors.), Tune, Denmark. http://www.ihh.kvl.dk/htm/php/tune99/9-Kyvsgaard.htm

 

Missohou A, Dieye P N and Talaki E 2002 Rural poultry production and productivity in Southern Senegal. Livestock Research for Rural Development, 14, 2:2002. http://www.cipav.org.co/lrrd/lrrd14/2/miss142.htm

 

Mlozi M R S, Kakengi A V M, Minga U M, Mtambo A M and Olsen J E 2003 Marketing of free - range local chickens in Morogoro and Kilosa urban markets, Tanzania. Livestock Research for Rural Development, 15, 2:2003. http://www.cipav.org.co/lrrd/lrrd15/2/mloz152.htm

 

Mwalusanya N A, Katule A M, Mutayoba S K, Mtambo M M A, Olsen J E and Minga U M 2001 Productivity of local chickens under village management conditions. Tropical Animal Health and Production, 34 (5): 405 - 416.

 

Safalaoh A C L 1997 Characteristics of indigenous chickens of Malawi. Animal Genetic Resource Information, 22: 61 - 69.

 

SAS 1999 SAS /STAT Users Guide. Version 8.1,  SAS Institute Inc. ,Cary, NC, USA.