Journal of Agricultural Science; Vol. 6, No.

10; 2014
ISSN 1916-9752 E-ISSN 1916-9760
Published by Canadian Center of Science and Education

Analysis of Chickpea Value Chain and Determinants of Market Options

Choice in Selected Districts of Southern Ethiopia
Tewodros Tefera1
1
School of Environment, Gender and Development Studies, College of Agriculture, Hawassa University, Ethiopia
Correspondence: Tewodros Tefera, School of Environment, Gender and Development Studies, College of
Agriculture, Hawassa University, Ethiopia. E-mail: [email protected]

Received: February 24, 2014 Accepted: July 15, 2014 Online Published: September 15, 2014
doi:10.5539/jas.v6n10p26 URL: http://dx.doi.org/10.5539/jas.v6n10p26

Abstract
Chickpea value chain study was conducted in selected districts of the south region where CIFSRF project was
operating. In total 227 respondents comprising 162 chickpea producers; 11 local assemblers; 8 rural wholesalers;
8 urban wholesalers and 38 retailers were interviewed. The study finding shows that the chickpea value chain
actors are broadly classified into three viz., inputs suppliers, direct market actors and enablers. Chickpea
producers sold their chickpea products to different market intermediaries and final consumers. About nine
chickpea marketing channels were identified. The total amount of chickpea that was transacted through these
marketing channels in 2012/13 was 9,181qts. Out of which the project site farmers supply 77% while 27% were
imported from other chickpea producing areas. The study result shows that the total gross marketing margin was
45.8% with producer participation margin of 54.2%. Approximately 10.2% out of a total gross marketing margin
of 45.8% constitutes the total marketing charges, giving a net marketing margin of 35.5%. Although this
suggests an appreciable level of profits reaped by traders, volumes traded are relatively low as compared to other
cereals. The market intermediaries incurred different marketing costs such as packing, processing, transportation
and loading unloading. Determinants of chickpea producer participation in alternative market options were
analyzed. A multinomial logit analysis results show that family size, landholding, access to market information
and Income from crops was positively influences wholesale market participation as compared to farm gate.
Similarly landholding, access to market information and extension services positively influence consumer
market participation than farm gate while access to information and income from crops positively influences
retails market participation than farm gate. On the other hand membership to cooperatives was negatively
influences wholesale, retail and consumer market participation than farm gate market option. Households
distance from nearest market negatively influences wholesale market participation than farm gate market option
while off farm activities negatively influences retail market participation than farm gate. The study suggested
that availing credit for agricultural marketing, promoting collective marketing, strengthening of extension
service, provision of market information and road development improves farmers marketing margin and
chickpea value chain performance.
Keywords: Chickpea, value chain, market margin, multinomial logit, chickpea marketing options
1. Introduction
Markets continue to be seen as the means for ensuring that smallholder producers of agricultural products are
effectively integrated into the mainstream of national economies, especially in developing countries. For one
thing, markets provide the opportunity for farm production to contribute to poverty reduction through the cash
income realized from sales of farm produce. In turn, markets drive production as farmers strive to meet the
demands of consumers and end-users in terms of quantity and quality. But their very existence, or how
effectively they function, cannot be guaranteed in many developing countries. In Ethiopia, there is a certain
urgency to address the real concern that, in spite of considerable investments into restructuring the sector since
1992 and directly tackle agrarian and land reform, poverty is still rife and there is the clear indication that much
of this arises from farmers not being able to sell produce at a profit. Unlocking markets for this group of farmers
is therefore considered a crucial developmental necessity. Research and case studies conducted in various parts
of the country point to the importance of the market access to smallholders (Chilot et al., 2010).
IDRC/CDRI has committed in improving food and nutrition security in Ethiopia and engaged in Research and

26
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Development work for more than a decade in partnership with Hawassa University. Thus far interventions such
as enhancing the soil health, promoting improved and bio-fortified varieties of pulse crops, introduction of better
processing and food preparation, market and value chain development, gender, nutrition education and human
capital development are the major ones. Cognizant to the importance of value chain approach to stimulate both
chickpea supply and demand side Equation, attention was given to study the marketing practice and value chain
of chickpea in call 3 project.
Pulse production is the major component of farming in the project sites. Haricot bean and chickpea are widely
produced by the majority of farmers and playing a crucial and diverse role in the farming systems and in the
diets of the community. They are ideal crops for simultaneously achieving four developmental goals in targeted
areas - reducing poverty, improving human health and nutrition, cash generation and enhancing ecosystem
resilience.
2. Objectives of the Study
This study addressed the following specific objectives: (i) identify different marketing channels and
intermediaries involved therein and their roles in chickpea marketing; (ii) determine the extent of value addition
in terms of marketing cost and margins in successive stages of chickpea movement; and (iii) determine factors
influencing households’ choice of market options.
3. Methodology
3.1 Description of the Study
The study is conducted in four districts: i) Halaba special district, ii) Damot Gale district of Wolita zone, iii)
Meskan district of Guraghe zone, and iv) Soddo district of Guraghe zone. A combination of qualitative and
quantitative methods was used for primary data collection. For household questionnaire survey a total sample
size of 227 respondents comprising of 162 Farmers, 8 rural wholesalers, 38 retailers, 8 urban wholesalers and 11
rural assemblers took part. A single Focus Group Discussion (FGD) per kebele with mixed community group
conducted. On Average 8 persons participated in the FGD.
3.2 Data Source and Type
This study used two kinds of data sources: (1). Interview of chickpea producers and intermediaries such as local
assemblers, rural wholesalers, urban wholesalers and retailers using structured questionnaire; (2). Focus group
discussion with farmers. The survey instruments were designed to explore the chickpea production, marketing,
product follow and distribution, marketing costs and margins. To complement the structured survey focus group
discussion was conducted with relevant chickpea value chain actors. Moreover personal observation and
secondary data collection were made.
3.3 Sampling Procedure and Methods of Data Collection
A multistage sampling procedure was adopted for this study. After the four districts had been selected, a stratified
sampling frame of chickpea producers and non producers were prepared. For the present study participants from
producers’ stratum were randomly selected until the required sample size was achieved. On the other had
chickpea traders such as local assemblers, rural wholesalers, urban wholesalers and retailers were sampled at
village and district towns by employing snow ball sampling techniques. Because of the limited number of traders
the sample exhaustively contained almost all chickpea traders in the study area. In total 162 chickpea producers;
11 local assemblers; 8 rural wholesalers; 8 urban wholesalers and 38 retailers were interviewed making a total
number of 227 respondents for the study.
Primary data were collected using four enumerators recruited from each district office of agriculture in the study
districts. Before the actual data collection went on one day training and orientation was given on the content and
approach of survey administration. These enumerators were closely and frequently supervised by the researcher
and the required data from the producers were gathered using a pretested interview schedule. For the traders, a
spate questionnaire was design and administered. The research design employed was a cross sectional one with
the aim of identifying market actors, market channels, marketing cost, margins and factors influencing the choice
of market channel by chickpea farmers.
3.4 Data Analysis
Primary data was entered in the SPSS spread sheet and cleaned for irregularities. The cleaned data was
summarized into descriptive format in terms of frequencies, percentages and central tendencies. STATA software
was used to analysis the Multinomial Logistic Model for the identification of factors influencing marketing
options. Based on the primary data, the average prices at different market levels were estimated, after which

27
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

average price margins for different market intermediaries were computed.

3.5 Estimation of Marketing Costs and Marketing Margins
Different types of marketing costs (including transport, packing, processing, loading unloading etc.,) relating to
transaction of chickpea for producers and each traders (local assemblers, rural wholesalers, urban wholesalers
and retailers) were collected per quintal basis.
The term marketing margin is commonly refers to the difference between producer and consumer prices of an
equivalent quantity and quality of a commodity (Tomek & Robinson, 1990). However, it may also describe price
differences between other points in the marketing chains. It is a price charged for providing a mix of marketing
services such as assembling, transportation, handling, packing, processing, storage etc. plus profit. Marketing
margins for the various chickpea traders were estimated using the following formulas.
Retailing Price – Farm gate Price
TGMM = × 100
Retailing/Consumer Price

Rural Assembler Price – Farm gate Price

GMMRA = × 100
Retailing/Consumer Price

Rural wholesalers – Rural Assembler Price

GMMRWS = × 100
Retailing/Consumer Price

Rural wholesalers – Rural Assembler Price

GMMUWS = × 100
Retailing/Consumer Price

Retailing Price – Urban Wholesalers

GMMR = × 100
Retailing/Consumer Price

NMM= TGMM – TMC

GMMP=100% – TGMM
Where
TGMM is Total Gross Marketing margin;
GMMRA is the percentage of the total gross marketing margin received by rural assemblers;
GMMRWS is the percentage of the total gross marketing margin received by rural wholesalers;
GMMUWS is the percentage of the total gross marketing margin received by urban wholesalers;
GMMR is the percentage of the total gross marketing margin received by retailers;
GMMP is the producers gross marketing margin;
TMC is the total marketing cost;
NMM is the net marketing margin.
3.6 Econometric Model
The primary goal of the econometric model used in this study is to explain the effects of the independent
variables Xi (i.e. farmers household, production and marketing characteristics) on the probability of choosing
among the various market channels exist in our study: rural assemblers, rural wholesalers, urban wholesalers and
retailers. The outcome is nominal because the categories are assumed to be unordered. Therefore the most
appropriate model to estimate farmers’ decision to sell in one of these four different market channels is a strategy
choice model, specifically a Multinomial Logistic Model (MNL). This model was used because it is the standard
method for estimating unordered, multi category dependent variables. It also assumes independence across the
choices, that is, it does not allow correlation or substitution between them (Wooldridge, 2008).
In this model each household makes only one choice from a group of available strategies, and this discussion is
based on a number of exogenous factors. Those factors include household level and area-specific variables. For
the ith farmer faced with j choices, suppose that the utility of choice j is:

28
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Uij= Zijβ + εij (1)

If farmer makes choice j in particular, then we assume that Uij is the maximum among the j utilities. Hence the
standard model will be driven by the probability that choice j is made which is,
Probability (Uij > Uik) for all other K # j (2)
Assuming that Yi represents the choice taken, then with j disturbances being distributed identically and
independently the multinomial logit model adopted. The multinomial logit is actually an extension of the binary
logit model, having more than two values for the dependent variable. Let (p0, p1 … pm) be the probabilities of
m+1 alternatives of choice. The probability of an individual i to choose the alternative j is given by:
exp( x ib i ) , where j = 0, 1 … m (3)
Prob(Yi  j ) 
1   mj 1 exp( x ib i )
Where xi is the vector of the independent variables associated to the individual i, and bj is the vector of
parameters associated to the alternative j.
Following Equation 3, the generalization of the logit model for the multinomial case is made by taking different
parameters bj depending on the alternatives of choice, such that the independent variables xi remain constants
depending on the products. Still, there is another possibility: the McFadden’s conditional logit model which
considers a constant vector of parameters b and allows the independent variables xij to depend on the alternatives
(McFadden 1974, 1980). The probability of an individual i to choose the product j is given by:

exp( x ij b ) exp( x *ij b ) , j=1,2, … m (4)

p ij  P ( y i  j )  
 exp( x ik b ) 1   mk 1 exp( x *ik b )
m
k 1

Where xij* = xij - xi0.

Based on Equation 4, according to Greene (2002) and Mugisha et al. (2004) and the fact that farmer’s
participation in different market options is categorized into alternatives, using those who participated in the
village market option (selling to rural assemblers) as the base alternative. The other alternatives include selling to
wholesalers (rural and urban wholesalers based in rural and district towns), retailers and consumers options. The
ratio of the probabilities, hence estimated as follows:
P( y i  j) exp( x*ijb ) exp( x ijb )
 m  j, l  1, 2, ... m (5)
P( y i  l )  k 1 exp( x*ik b )  mk 1 exp( xik b )
Which, as in the case of the multinomial logit is independent of the other alternatives of choice i.e. selling at
farm gate, wholesale, retailers or consumer market.
The marginal effects are obtained from the multinomial logit regression results by the following Equation:
 p (6)
 x ji
ji
 p    p 
ji i ki k

Where, β and P represents the parameter and likelihood, respectively, of the choices. Marginal likelihood gives
better indications and represents changes in dependent variable for a given change in a particular explanatory
variable whereas holding the other explanatory variables at their sample means. The models are estimated under
maximum likelihood procedure which yield consistent, asymptotically normal and efficient estimate.
4. Result and Discussion
4.1 Chickpea Value Chain Actors and Major Functions
Inputs supply: the major seed suppliers in the study areas were HwU CIFSRF project, traders, informal farmer to
farmer seed exchange, Debre Zeit Agricultural Research Centre. WoA was the major actor identified in supply of
fertilizer, insecticide and herbicide while MFI supply short term credit.
Chickpea grain supply: The chickpea value chain major function from intermediaries’ point of view was
analyzed. The result indicates that the major sources of chickpea supply are farmers. However, the degree of
dependence on farmers for chickpea grain supply varies from one to another. Local assemblers fully met their
chickpea supply requirements from farmers. Rural wholesalers depend on farmers for 50% of their chickpea
requirement while retailers and urban wholesalers obtained 34 and 25% of their chickpea grain from farmers
respectively. Intra and inter trade is common among chickpea intermediaries and chickpea flow depending on
price signals, marketing cost and concentration of customers. There was a statistical difference in chickpea grain
suppliers among chickpea intermediaries.
Credit supply: wholesalers are the major source of credit in the chickpea trade. Out of the total 22 traders who

29
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

took credit in 2012/13 production season, 77.2% of them were met their credit need from wholesalers. The
Microfinance institutions (MFI) supply credit for 18.2% of the chickpea traders while relatives supply for 4.6%.
Source of information: retailers are the major source of information. To limited extent retailers, brokers,
wholesalers and mass media were identified as source information in the chickpea market. The retailers’ greater
access to information might be due to their frequent contact with chickpea actors and consumers. The role of
mass media as source information was less important.
Chickpea buyers: the major buyers of chickpea are consumers. Wholesalers and retailers are also the secondary
buyers of chickpea to resale back to consumers and other needy buyers. There is a statistically significant
difference at 1% level in terms of chickpea buyers.

Table 1. Chickpea value chain actors and their major roles

Chickpea actors and Traders
Totals
major functions Rural wholesalers Retailers Urban wholesalers Local assemblers
Chickpea suppliers
Farmers 4 (50) 13 (34) 2 (25) 11 (100) 30 (55.3)
Retailers 0 (0) 6 (16) 0 (0) 0 (0) 6 (9.23)
Brokers 0 (0) 0 (0) 1 (12.5) 0 (0) 1 (1.54)
Other wholesalers 4 (50) 18 (47) 5 (62.5) 0 (0) 27 (41.5)
Pearson chi2= 31.66 Pr=0.007
Credit suppliers
IMF 30(100) 0 (0) 0 (0) 1 (50) 4 (18.20)
Wholesalers 0(0) 12 (100) 5 (100) 0 (0) 17 (77.26)
Relatives 0 (0) 0 (0) 0 (0) 1 (50) 1 (4.55)
Pearson chi2=22.00 Pr=0.001
Source of information
Retailers 5 (62.6) 19 (67.8) 4 (50) 11 (100) 39 (70.9)
Brokers 0 (0) 1 (3.57) 3 (37.5) 0 (0) 4 (7.27)
Wholesalers 3 (37.5) 6 (21.4) 1 (12.5) 0 (0) 10 (18.2)
Mass media 0 (0) 1 (3.6) 0 (0) 0 (0) 1 (1.8)
Pearson chi2=16.56 Pr=0.167
Chickpea buyers
Consumers 2 (25) 36 (97.3) 2 (25) 5 (45.5) 45 (69.2)
Wholesalers 0 (0) 0 (0) 0 (0) 6 (54.5) 6 (9.2)
Retailers 6 (75) 1 (0) 6 (75) 0 (0) 13 (20)
Pearson chi2=81.96 Pr=0.000
Figures in parenthesis are percentage.

30
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Inputs suppliers Direct market actors

HwU/CIFSRF project Local Rural Urban

Producers Retailers Consumers
collectors wholesalers wholesalers
Traders

Farmer to farmer
exchange

Woreda Office of Enablers (technical advice, service, finance, policy)

Agriculture (WoA)

Research center
Research
(Debre Zeit) CIFSRF MFI WoA NGos BoA
centers
MFI

Figure 1. Chickpea value chain actors

The chickpea value chain actors: The chickpea value chain actors are broadly classified into three viz., inputs
suppliers direct market actors and enablers. The inputs suppliers mainly deliver inputs such as seeds, fertilizers,
finance and pesticide. The major actors identified in the inputs supply functions are CIFSRF project, traders,
farmers, Debre Zeit Agricultural Research centers, MFI, and woreda office of agriculture. The direct market
actors are those involved the chickpea trade and they dictate the flow of chickpea in time and space. These
include producers, local collectors, wholesalers, retailers and consumers. The enablers are the inputs suppliers
are coincide most of the times as most of the input suppliers also involved in technical advice, service provision
and policy formulation and implementation of the same.
4.2 Marketing Channels of Chickpea
Chickpea is an important cash crop and it passes through the hands of many intermediaries. Based on the
direction of flow and volume of chickpea transacted, nine marketing channels were identified. The channel starts
from the producers (farmers) and ends in the terminal market passing through a number of marketing actors
along the chain. According to producers survey which involves 162 respondents in the four study districts in the
year 2012/13 about 318 quintals of chickpea was marketed. However, the trader survey shows that 6708 qt of
chickpea is purchased from farmers in the study areas while 2473 quintal of chickpea was imported outside the
project area. Hence, the total amount of chickpea that was transacted 2012/13 in the project districts was found
to be 9,181 qts. In order to quantify the volume of chickpea handled by each marketing actor along the marketing
chain, the total purchased amount was obtained from the producer and trader surveys. The trader survey shade
light that the study areas are not chickpea secured and 27% of the chickpea demand was met through import
from other regions of the country. Urban wholesalers played a critical role in filling this deficit.
The study identified nine chickpea marketing channels. The amount of chickpea transacted in these market
channels was different and out of the nine, four market channels were found to be dominant in terms of chickpea
volume of transaction. Marketing channel 1 starts from producers and ends with final consumer. In this market
channel about 350 qt chickpea (3.8%) was supplied. Channel 2 involves producers, retailers and consumers. It
was found to be the least dominant one in terms of volume of chickpea supply. In this market channel only 0.8%
of the total chickpea was supplied. Marketing channel 3 was the dominant one, about 36% of the chickpea was
supplied in this channel. The participants of this market channel include producers, rural assemblers, urban
wholesalers and consumers. Channel 4, 5, 6 and 7 supply 3.5%, 8%, 2.3% and 6.4% of chickpea respectively.
Marketing channel 8 and channel 9 were dominant as they accounted for the supply of 12% and 27% of the
chickpea respectively. Market channel 8 comprises of market actors such as producers, urban wholesalers,
retailers and consumers where as market channel 9 include centrally located wholesalers in the capital city Addis
Ababa and Debre Zeit, local urban wholesalers and consumers.

31
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Chickpea marketing channels and percentage of

chickpea transacted

Channel 1 3.8
Channel 2 0.8
Channel 3 36.2
Channel 4 3.5
Channel 5 8.0
Channel 6 2.3
Channel 7 6.4
Channel 8 12.1
Channel 9 26.9
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0

Figure 2. Chickpea marketing channels

The identified channels were:

Channel 1: Producers Consumers (350qt)
Channel 2: Producers Retailers Consumers (70qt)
Channel 3: Producers Rural Assemblers Consumers (3322 qt)
Channel 4: Producers Rural Assemblers Urban wholesalers Consumers (325qt)
Channel 5: Producers Rural wholesalers Consumers (735qt)
Channel 6: Producers Rural wholesalers Retailers Consumers (208qt)
Channel 7: Producers Urban wholesalers Consumers (586qt)
Channel 8: Producers Urban wholesalers Retailers Consumers (1112qt)
Channel 9: Central Wholesalers Urban wholesalers Consumers (2473qt)

32
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Producers (Farmers)
6708 qt.

3647

943

Rural assemblers 1698

3647 qt
70

Rural wholesalers
350 943 qt

208
Urban wholesalers
1112 325 4496 qt

735
Retailers 3322
1390 qt
586

2473
1390
2473

Local consumers (9181 qt)

Regional and federal wholesalers

2473 qt

Figure 3. Chickpea marketing channels; the figures represent volume of chickpea passing in the different market
channels

4.3 Marketing Margin of Chickpea

Results of the marketing margins analysis showed a total gross marketing margin of 45.8% with producer
participation margin of 54.2%. The gross marketing margin for chickpea is rather on the high side as compared
to that of hot pepper which is about 29% (Amare & Dawit, 2013). Approximately 10.2% out of a total gross
marketing margin of 45.8% constitutes the total marketing charges, giving a net marketing margin of 35.5%.
Although this suggests an appreciable level of abnormal profits reaped by traders, volumes traded are relatively
low as compared to other cereals.

33
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Table 2. Chickpea marketing margin

Price at various level of chain actors Birr/qt Gross marketing margins
Average farm gate price 663.30 -
Average rural assemblers price 752.20 -
Average rural wholesalers price 895.00 -
Average urban wholesalers price 1048.45 -
Average retailing price 1224.15 -
TGMM - 45.8%
GMMRA - 7.3%
GMMRWS - 11.6%
GMMUWS - 12.5%
GMMR - 14.4%
GMMP - 54.2%

Again disaggregating total net marketing margins by the key traders along the distribution chain indicate that
retailers absorb a higher proportion of the total which compensates for the low volumes traded (13.2%).
However, there is a fair representation of profit allocation among rural and urban wholesalers leaving very little
to rural assemblers (4.2%).

Table 3. Chickpea marketing costs

Districts
Marketing cost
Rural wholesalers Retailers Urban wholesalers Local assemblers
Packing 2.00 - 4.00 10.00
Processing - 6.44 - -
Transport 18.20 5.60 29.25 19.10
Loading unloading 5.60 0.57 3.18 2.18
Total marketing cost 25.8 12.04 36.43 32.28
Total as % of retail price 2.5 1.1 3.5 3.1
GMM (%) 11.6 14.3 12.5 7.3
NMM% 9.1 13.2 9.0 4.2

4.4 Market Participation

In subsistence production system producers have played dual roles of producing and consuming. And in such
system what is produced may not be sold or partly sold to market. According to the study findings the market
participation level for chickpea producers ranges between 69% in Halaba and 100% in Meskan. This indicates
chickpea is a cash crop and contribute much to the household cash generation. In all the study areas the
overwhelming majority of respondents confirm that their income from chickpea is increasing. Despite the cash
generation form chickpea has shown increase there is statistical difference among the districts at 5% level.

34
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Table 4. Producers’ participation in chickpea market

District
Chickpea marketing Chi2
Halaba Damot Gal Sodo Meskan
Percentage
Market participation 69 78 95 100 18.74***
Trends of income obtained from chickpea
Increasing 51 59 67 56
Decreasing 3 11 14 22
20.18**
Same 13 20 12 6
Difficult to tell 33 9 7 16
**, *** indicates that chickpea market participation and trends in income obtained from chickpea are
significantly different from zero among districts at the 5 and 1 percent level, respectively.

4.5 Determinants of Market Option Participation

For this analysis some of the market options (rural and urban wholesalers) are merged and rural assembler
market option is renamed village market. Accordingly the market option is categorized into four namely village
market, wholesale market, retailers and consumers. Out of the four options wholesale and retailers market are
dominant while village market and consumer market are limited in terms of market outlet options.

Table 5. Chickpea market option

Market option Freq. Percent Cum.
Village market 11 7.75 7.75
Wholesale market (R+U) 93 65.49 73.24
Retailers 26 18.31 91.55
Consumers 12 8.45 100.00
Total 142 100.00

Variables Used in Multinomial Logit Regression:

Membership in cooperatives: the majority of respondents who sold their chickpea products at the farm gate,
directly to consumers and retailers market options were member of cooperative whereas 73% of respondents
who sold their chickpea to wholesalers market option were nonmembers of cooperatives. The observed
difference in the choice of market options is statistically significant at 1% level.
Access to market information: the majority of the respondents were affirmed their access to market information.
From those who sold their chickpea to wholesales, retails and consumers the overwhelming majority (greater
than 50%) had access to market information while the majority of the respondents who sale their chickpea at
farm gate lack market information. The difference in market option choice is statistically influenced by market
access and the observed difference was statistically significant at 1% level.
Access to extension service: the majority of the respondents who sold to farm gate (81.82%), wholesale (89.25%),
retail (88%) and consumer (75%) market option had access to extension services. Access to extension service
influence on market option choice was not statistically significant.
Off-farm participation: only 9% of those who sold to farm gate market option participated in off-farm activities
while 29% of respondents each who sold to wholesales and retails participated in off-farm activities. The
differences in off-farm activities participation were not statistically influencing market options.
Access to credit: the majority of respondents who had access to credit participated in retails (56%) and consumer
(58%) markets. Among the respondents who sold their chickpea at farm gate and wholesales market options only
18% and 38% of them had access to credit. The difference in credit access to credit was statistically influencing
market options choice at 10% level of significance.

35
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Table 6. Descriptive statistics of variables used in the model (discreet variables)

Market option
Categorical variables Chi2
Farm gate Wholesale Retailers Consumer
Membership in coops
- Yes 9 (81.82) 25 (26.88) 14 (53.85) 7 (58.33) 18.55***
Access to market information
- Yes 5 (45.45) 80 (86.02) 19 (73.08) 6 (50) 15.82***
Access to extension service
- Yes 9 (81.82) 83 (89.25) 22 (88) 9 (75) 2.25ns
Participate in off-farm activities
- Yes 1 (9.09) 27 (29.03) 7 (29.17) 3 (25) 2.05ns
Access to credit
- Yes 2 (18.18) 36 (38.71) 14 (56) 7 (58.33) 6.30*
*, and *** Imply level of significance at 10 and 1% respectively.

Years of schooling in formal education: those who sold their chickpea at farm gate had 4.3 years of formal
education while those who sold in wholesales, retails and consumer market had 5.8, 6.45 and 4.8 years of formal
education respectively. The finding indicates that education statistically influence choice of market option at 10
% level.
Family size: the finding show that households having large average family size preferred to sale for consumers
and retail market options. However, family size was not statistically influence the choice of market options.
Total landholding: respondents own the largest average land holding sold their chickpea at farm gate while those
own the smallest average land holding sold to consumers market. However, landholding was not found to be
statistically influence the choice of market options.
Distance from home to nearest market: households sold to the farm gate market option was found on average 11
km away from the nearest market while those who sold for wholesales, retails and consumers market option
located on average 9, 8, and 10.9 km away from home. The observed difference in distance from the nearest
market influence the choice of market option and it was found to be statistically significant at 10% level.
Income from crops: households who got large income from crops prefer to sale for consumers and retails market
options. On the other hand those households who generate small average income from crops sold their chickpea
to wholesales and farm gate markets. Average crops income influence market options and the result was
statistically significant at 10% level.
Production costs: households incurred higher production costs sold their chickpea for wholesales and consumers
markets. Producers and wholesalers also interlined in the credit market as the wholesalers supply credit to
producers as the later expected to deliver their produces to them. The preference to consumer market is
motivated by higher price. However the observed difference was not statistically significant.
Marketing cost: households who sold at farm gate market on average incurred less marketing costs compared to
those sold to wholesales (11.8), retails (14.5) and consumers (15.4). The difference however, was not statistically
significant.

36
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Table 7. Descriptive statistics of continuous variables

Market option
Continuous variables F-value
Farm gates Wholesales Retails Consumers
Education (average years of schooling) 4.3 5.8 6.45 4.8 2.17*
Family size 7.66 7.58 8.65 9.27 1.53ns
Total land holding (ha) 2.20 1.82 1.86 1.74 0.5ns
Distance from home to nearest market (km) 11.1 9.03 8.25 10.9 2.19*
Income from crops 7,571.50 9,267.56 14,494.67 39,686.00 2.79*
Production cost (Birr) 208.1 437.6 196.21 382.75 1.05ns
Marketing cost (Birr) 5.25 11.77 14.46 15.42 0.66ns
*, and *** Imply level of significance at 10 and 1% respectively.

4.6 Factors Affecting the Choice of Chickpea Marketing Options

In this section, results from the estimation of the multinomial logit model are presented. Discussion of the results
focuses on the three comparisons within the model: the comparisons of mass wholesales, retails, and consumers
market options to the farm gate. In a next section, the three comparisons within the model are evaluated in terms
of individual variables that were found to be significant.
In terms of goodness of fit, the model had acceptable measures. The likelihood ratio chi2 was 79.41 and the
pseudo-R2 measures were 34.79. These measures were statistically significant at less than 1% level. In order to
develop broader understanding of how producers’ characteristics, their access to resources and services are
related to marketing options choice the 0.10 level was chosen to indentify significant coefficient estimates
Family size: according to the results, family size was positively influences wholesale market participation and
the result was significant at 10% level. As the family size of the households changes by one, the probability of
participating in the wholesale market than farm gate increases by 1.7% implying that the chickpea producers will
sell more chickpea in the wholesale market as compared to the farm gate market. The finding agree with the
prior expectation that large family size implies better labour endowment so that households are in a position to
travel to get wholesalers in the district or nearby town markets.
Cooperative membership: The finding reveals that, membership to cooperatives was statistically significant and
negatively influences wholesale, retail and consumer market participation respectively. As the membership status
of the households’ changes from not member to member, the probability of participating in the wholesale market
than farm gate reduces by 31.1% implying that the households will sell fewer amounts of chickpea in the
wholesale market as compared to the farm gate market. Similarly cooperative member’s participation in retail
and consumer markets reduces by 2.6 and 2.1% as compared to farm gate markets. This was inconsistent with
the prior expectation. It was assumed that cooperatives members are price conscious and they sale their chickpea
to markets which pay higher price.
Total landholding: the results indicate that the household’s ownership of total landholding positively influences
wholesale and consumer market participation. As the landholding size increases by one hectare, the probability
of participating in wholesale and consumer than farm gate increases by 0.7 and 0.2% respectively. This is
consistent with Machethe et al. (2008) who find that larger land sizes raise the probability of market participation
for farmers since land is a critical production asset having a direct bearing on production of a marketable surplus,
ceteris paribus. This implies that those with large parcels of land are likely to participate more in markets
especially in wholesale and consumers markets as they have scale advantage to reduce costs to take products to
distance market.
Distance from home to nearest market: the study finding shows that, households distance from nearest market
was significant at 10% level and negatively influences wholesale market participation. As the distance of the
household from nearby market increased by one kilometer, the probability of participating in the wholesale
market compared to farm gate market reduces by 4.4% implying that the households will sell fewer chickpea in
the wholesale market as compared to the farm gate market. This result concurs to the prior expectation, as
household’s location further from market places they prefer to sell their produce at farm gate.
Access to market information: Access to market information positively influences wholesale, retail and consumer

37
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

markets participation such that, a change in a farmer’s status from no access to market information to access
increases the probability of wholesale, retail and consumer markets participation than farm gate by 33.8, 1.7 and
3% respectively. This implies that access to information will make the farmer participate more in the wholesale,
retail and consumer than the farm gate market option. Market information enable the farmer to improve their
decision making on what to produce and to whom to sell which in turn increases his/her marketed surplus and
market participation (Lapar et al., 2002; Shepherd, A., 1997).
Crops income: income from crops influences wholesale and retail markets participation positively and
significantly. An increase in the yearly income by one Ethiopian birr increases the probability of participating in
the wholesale and retail market than farm gate by 1.7 and 0.2% respectively. The increase in cash resources will
make the households invest more in chickpea production and marketing activities resulting to more surplus
driving them to sell to wholesale which is a larger market compared to farm gate.
Participation in off farm activities: the results show that participation in off farm activities negatively influences
retail market participation. As the farmer involved in off farm activities the time he/she has to spare for
marketing of agricultural activities and to produce marketable surplus is less, hence this decreases the probability
of participating in retail market than farm gate increases by 13.3%.
Access to extension service: it positively influences consumer market participation such that, a change in a
farmer’s status from no access to extension to access increases the probability of consumer market participation
than farm gate increased by 29.3%. This implies that extension will make the farmer participate more in the
consumer than the farm gate market option. Extension services enable the farmer to improve his production
methods hence leading to more output which in turn increases his/her marketed surplus hence market
participation (Lapar et al., 2002), especially in terminal markets which provide lucrative price.

Table 8. Determinates of chickpea producer choice of market options

Market participation options Coef. (Std. Err.) z P>|z| Marginal effects
Wholesales option
Years of education -0.20 (0.265) -0.77 0.442 0.010
Family size 0.474 (0 .24) 1.92 0.055* 0.017
Co-ops membership -7.57 (3.64) -2.08 0.038** -0.311
Total landholding 0.15 (0.073) 1.97 0.049** 0.017
Distance from home to nearest market -0.49 (0 .25) -1.94 0.052* -0.044
Access to market information 7.39 (3.28) 2.25 0.024** 0.338
Income from crops 0.047 (0 .026) 1.76 0.078* 0.001
Marketing cost -0.134 (0 .094) -1.42 0.156 -0.008
Access to extension service -0.485 (1.68) -0.28 0.776 -0.050
Participation in off farm activities -1.63 (1.58) -1.03 0.302 0.061
Access credit 1.16 (1.65) 0.7 0.484 0.087
_cons -4.60 (7.67) -0.6 0.548
Retail market option
Years of education -0.267 (0 .27) -0.99 0.324 -0.005
Family size 0.389 (0 .25) 1.55 0.12 0.005
Co-ops membership -6.43 (3.65) -1.76 0.079* -0.026
Total landholding 0.11 (0 .075) 1.4 0.161 -0.003
Distance from home to nearest market -0.215 (0.26) -0.83 0.409 0.022
Access to market information 6.24 (3.31) 1.88 0.06* 0.017
Income from crops 0.057 (0.03) 2.1 0.036** 0.002
Marketing cost -0.12 (0.096) -1.24 0.216 -0.002
Access to extension service -0.686 (1.73) -0.39 0.693 -0.054
Participation in off farm activities -2.67 (1.63) -1.64 0.10* -0.133
Access credit 0.498 (1.72) 0.29 0.773 -0.060
_cons 1.22 (8.0) 0.15 0.879

38
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

Consumers market option

Years of education -0.486 (0.296) -1.64 0.10* -0.016
Family size 0.528 (0.261) 2.02 0.043** 0.008
Co-ops membership -6.66 (3.708) -1.8 0.073* -0.021
Total landholding 0.152 (0.080) 1.9 0.057* 0.002
Distance from home to nearest market -0.234 (0.275) -0.85 0.395 -0.007
Access to market information 5.593 (3.361) 1.66 0.096* 0.030
Income from crops 0.033 (0.028) 1.18 0.239 -0.001
Marketing cost -0.028 (0.103) -0.27 0.789 0.005
Access to extension service 1.19 (0.53) 2.24 0.023** 0.293
Participation in off farm activities -1.92 (1.725) -1.11 0.266 -0.013
Access credit 1.09 (1.682) 0.65 0.516 0.015
_cons -9.54 ( 8.47) -1.13 0.26
*, and *** Imply level of significance at 10 and 1% respectively.
Village market is the base outcome; Number of observation = 133; LR chi2 = 79.41; Log likelihood = -74.52;
prob. > χ2 = 0.000; pseudo R2 = 34.79.

5. Conclusion
Chickpea production is the major component of farming system in the project sites. It contributes significantly to
household consumption and cash income generation. In this study focus has been given on exploring the value
chain of chickpea in selected potential chickpea producing districts and identification of factors influencing
market outlet options.
The study identified three categories of value actors viz., inputs suppliers direct market actors and enablers. The
inputs suppliers mainly deliver inputs such as seeds, fertilizers, finance and pesticide. The major identified actors
in the inputs supply domain include CIFSRF project, traders, farmers, Debre Zeit Agricultural Research centres,
MFI, and woreda office of agriculture. The direct market actors are those involved the chickpea trade and they
dictate the flow of chickpea in time and space. The major value chain actors in this category include producers,
local assemblers, rural wholesalers, urban wholesalers, retailers and consumers. About nine chickpea marketing
channels were identified. The total amount of chickpea that was transacted through these marketing channels in
2012/13 was 9,181qts. Off which the project site farmers supply 77% while 27% were imported from other
chickpea producing areas.
Chickpea producers incurred in kind and cash expenses for hired labour, fertilizers, seed and pesticides. In
addition producers expended marketing cost such as sale tax, transportation, loading unloading and packing
charges. While the majority of respondents was reported their access to extension, limited number of them had
access to credit and participated in off farm activities. Involvement of some intermediaries seems to be redundant
whose presence just adds a cost to the consumer and a loss to the smallholder farmers by reducing their price
share. Moreover, the superfluous involvement of intermediaries keeps chickpea producers and markets separated
not allowing them to be market responsive.
The study result shows that the total gross marketing margin was 45.8% with producer participation margin of
54.2%. Approximately 10.2% out of a total gross marketing margin of 45.8% constitutes the total marketing
charges, giving a net marketing margin of 35.5%. Although this suggests an appreciable level of abnormal profits
reaped by traders, volumes traded are relatively low as compared to other cereals. The market intermediaries
incurred different marketing costs such as packing, processing, transportation and loading unloading.
Determinants of participation in alternative market option were analyzed. The multinomial logit analysis results
show that family size, landholding, access to market information and Income from crops was positively
influences wholesale market participation as compared to farm gate. Similarly landholding, access to market
information and extension services positively influence consumer market participation than farm gate while
access to information and income from crops positively influences retails market participation than farm gate.
On the other hand membership to cooperatives was negatively influences wholesale, retail and consumer market
participation than farm gate market option. Households distance from nearest market negatively influences
wholesale market participation than farm gate market option while off farm activities negatively influences retail

39
www.ccsenet.org/jas Journal of Agricultural Science Vol. 6, No. 10; 2014

market participation than farm gate.

Policy Implications:
1. The most common market option for majority of the farmers is the wholesale market because of its role in
interlocked credit market and relatively better price it offers.
2. Access to market information, extension service, and credit enables farmers to participate in market options
which secure them higher price for their chickpea produce
3. Households’ resource endowment such as labour, income from crops and landholding size also raises
households’ likelihoods to participate in wholesale, retailers and consumers market options.
4. Infrastructure development also ensure better benefit for markets as distance from market outlets is a major
constraint for households to participate in markets that ensure higher price.
Acknowledgements
The authors gratefully acknowledge the DFATD /IDRC for funding this study. The author has also benefited
from discussions with, and the support of, Dr. Sheleme Beyene, Dr. Carol Henery and Dr. Atul Nayyar.
References
Amare, H., & Dawit, A. (2013). Marketing channel and margin analysis: A case study of red pepper marketing at
Jabitehinan District in Northwestern Ethiopia. International Journal of Agricultural Economics and
Extension, 1(6), 31-40.
Chilot, Y., Shahidur, R., Befekadu, B., & Solomon, L. (2010). Pulses Value Chain Potential in Ethiopia:
Constraints and opportunities for enhancing exports. International Food Policy Research Institute (IFPRI).
Greene, W. H. (2002). Econometric Analysis (5th ed.). Pearson Education Inc., New Jersey.
Lapar, M. L., Holloway, G., & Ehui, S. (2002). Policy options promoting market participation of smallholder
livestock producers: A case study of the Philippines. Socio-economic and Policy Research Working Paper
47. ILRI, Nairobi.
Machethe, C., Jagwe, J., & Ouma, E. (2008). Transaction costs and smallholder farmers’ participation in
banana markets in the Great Lakes Region. Department of Agricultural Economics, Extension and Rural
Development, University of Pretoria.
McFadden, D. (1980). Econometric Models of Probabilistic Choice Among Products. Journal of Business, 53,
513-529. http://dx.doi.org/10.1086/296093
Mugisha J, Mbowa, S., & Kasibante, F. (2004). Determinants of use of Information and Communication
Technologies by Agribusiness Firms in Uganda. East Afr. J. Rural Dev., 20, 377-7103.
Shepherd, A. (1997). Market Information Services: Theory and Practice. FAO (Food and Agriculture
Organization), Rome.
Studenmund, A. H. (1992). Using Econometrics. Apract.cal Guide. New York. Harper Collins U.S Bureau of
Census. Country Business Patterns. Economics Letter, 38, 31-36.
Tomek, W. G., & Robinson, K. L. (1990). Agricultural Product Prices (3rd ed., pp. 107-8(360)). Cornell
University Press, New York.
Wooldrige, J. M. (2008). Introductory Econometrics: A Modern Approach (4th ed.). J. W. Calhoun: CENGAGE
Learning.

Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution
license (http://creativecommons.org/licenses/by/3.0/).

40