What Informs Farmers’ Choice of Output Markets? The Case of Maize,Cowpea and Livestock Production in Northern Ghana

Abstract

This study analyses the main determinants of output market choices by rural farmers in northern Ghana amidst growing concerns of lack of lucrative markets for smallholder farmers. Using recent survey data collected from 448 households, the study applied the multinomial logistic regression (MLR) model with village markets as the base outcome. The findings revealed that association membership, access to storage facilities, openness to new production and marketing methods, access to financial services, knowledge of sustainable intensification (SI) practices, access to guaranteed market, availability of quality market services and distance to output market would likely influence the choice of a farmer to sell at a farm gate over village markets. Access to extension services was found significant in influencing farmers’ decisions to sell by the roadside. Also, gender, association membership, access to processing facilities, availability of quality market services and distance to output market would likely influence the choice of a farmer to do private sales. Furthermore, association membership, access to processing facilities, access to extension services and market information significantly influences the decision of farmers to use other market outlets (e.g., regional/district markets). The study concludes that the choice of market outlet used by farmers depends much on institutional and channel-specific characteristics. These findings have policy implications for the development of market policies, providing rural market infrastructure services, promoting SI practices and strengthening extension service delivery.

Keywords

Access determinants Northern Ghana output markets smallholder farmers

Introduction

Markets play an important role in the rural economy of farming households. In sub-Saharan Africa, market access is a key determinant of economic development and directly linked to increased farmer productivity and food security in most countries (Corsi, Marchisio, and Orsi 2017; Garretsen and Bosker 2012; Obisesan 2018). For instance, a percentage increase in market access has been found to be associated with a 0.03% increase in Gross Domestic Product growth per worker (Garretsen and Bosker 2012). The importance of market access has increased in the past two decades and efforts to improve farmers’ access to markets remain a key feature in rural development strategies especially in agrarian economies. This is evident in numerous collective actions such as institutional arrangements and national level policy interventions aimed at improving smallholder access to markets (Markelova et al. 2009).

In recent years, the policy focus in most developing countries is centred on how to improve the efficient functioning of agricultural value chains for better market access for nutritional and health outcomes (FAO 2013; GAIN 2013). Recent evidence shows that access to rural markets is positively linked to farmers’ welfare through improved food security and household consumption (Stifel and Minten 2017). However, improved diets resulting from market access appears to be insignificant in delivering the needed nutritional outcomes.

Smallholder farmers are faced with multiple constraints in accessing both input and output markets. These constraints are institutional, social and technical in nature (Thindisa and Urban 2018). ILO (2017) in assessing market access for smallholder farmers found that (a) the absence of lucrative markets with the requisite support functions for smallholder farmers to access hampers efforts aimed at moving them from subsistence to commercial production, (b) contract farming is a useful approach to linking smallholder farmers to value chains as it enhances access to improved inputs, technical assistance, secured markets and stable prices and (c) large income gains and improvements in productivity are associated with contract farming. However, sustainability of the impacts remains a challenge due to high dropout rates in most contract farming schemes. Furthermore, the marginalisation of farmer groups at the conceptual/formulation stage of policy interventions aimed at improving market access often undermines the impact they were originally intended to create.

In northern Ghana, most smallholder farmers prefer to use village markets over the district and regional markets for the sale of their produce due to proximity and lower transfer costs of market participation (Amikuzuno 2015). Empirical evidence of what drives farmers choice of output markets and the key constrain that limit their participation in output markets, especially in northern Ghana is limited. This study contributes to the output market access literature in two important ways: (a) using the MLR modelling approach, we show that access to essential marketing services (storage, cleaning, grading, standardisation, finance, information), association membership, knowledge of sustainable intensification (SI) practices and distance to input sources would likely dominate the decision of a farmer to sell at the farm gate over the use of village markets and (b) the choice of market outlet used by farmers is diverse and cannot be explained by a single factor. The study, therefore, contributes to enhancing our understanding of the factors that influence smallholder farmers’ access to output markets in northern Ghana and the underlying patterns that inform their choice of market participation.

The rest of the article is structured as follows: The section ‘Output Market Choices and Determinants: A Review’ reviews the related relevant literature on farmers’ access to output markets and drivers. The section ‘Methodology’ presents the data and methodology employed in the study. The results obtained from the analysis and the discussions that ensued are captured in the section ‘Results and Discussions’. The main conclusions and policy relevance are detailed in the section ‘Conclusions and Recommendations’.

Output Market Choices and Determinants: A Review

Several studies have analysed the constraints and determinants of farmers’ access to output markets in both rural and urban markets (Fikadu, Duguma, and Mitiku 2019; Musemwa et al. 2008; Panda and Sreekumar 2012). Jari and Fraser (2009) analysed the technical and institutional factors that affect smallholder agricultural marketing in South Africa and revealed that tradition, group participation, market infrastructure, social capital, contractual arrangements, expertise in grades and standards and market information influence households marketing behaviour. Amankwah et al. (2012) and Binge, Mshenga, and Kgosikome (2019) reported that institutional constraints (high mortality, dry season water shortages, poor interaction among stakeholder institutions and communities, lack of proper markets, pasture scarcity, theft) affects the market participation of livestock farmers. Low prices for livestock, poor roads and lack of market information constitutes marketing constraints (Binge, Mshenga, and Kgosikome 2019). Panda and Sreekumar (2012) reported that institutional, economic and technical factors influence vegetable farmers’ market choices. Technical factors that contribute to good quality products for consumers include storage facilities, physical infrastructure, transport ownership, good road network and adding value to produce. Institutional and economic factors likely to influence farmer participation in formal markets are access to market information, guaranteed markets, grading of produce, training and education. However, the study did not consider other economic and political factors such as risk management strategy which influences market participation. Previously, Musemwa et al. (2008) analysed the marketing constraints and opportunities associated with Nguni cattle based on its social and economic importance to resource-poor farmers. The main marketing constraints identified were distance to formal markets, poor infrastructure, high transaction costs, poor road network and limited information. Existing market opportunities revealed which actors in the livestock value chain could take advantage of includes private sales/informal markets, auctions, butcheries and abattoirs. The study emphasised the need to promote locally adopted breeds and work towards addressing the constraints identified for the sustainability of the livestock industry.

Kihoro et al. (2016) analysed the market channels (retailers, wholesalers, assemblers) used by farmers in Ethiopia. The results show that the age of farmer, price of green grams, access to credit and selling as individuals positively influences the choice of rural assemblers marketing channel. Gender of household head, cost of production and use of mobile phones for market information, positively influences the probability of using rural retailers over wholesale marketing channels. The study concluded that marketing channels with low production and transport costs and higher prices are preferred by farmers. Identification and prioritisation of farmer–trader relationships that are unique and promote more market options for farmers are desirable.

The role of Information and Communication Technologies (ICTs) in enabling farmers’ access to output markets through market information dissemination is highlighted by various studies (Akar 2008; Amaya and Alwang 2011; Dagne and Oguamanam 2018; Goyal 2010; Overa 2006). The use of cell phones and networks is affecting traditional means of getting information especially with the heavy involvement of women in markets (Amaya and Alwang 2011). Market information has entered the digital age and is crucial in improving the incomes and wellbeing of farmers. Cellular technology is reported to broadened access to information, quickened the flow of information among actors and reduced search costs (Akar 2008; Amaya and Alwang 2011; Overa 2006). The low information costs have the potential to change market choices, the role of gender and influence output prices received by farmers. For instance, in India, the provision of price information by Kiosks to farmers in rural villages led to an increase in output price received by farmers from 1% to 5% (Goyal 2010). This does not only impact on the income level of farmers but also helps to link farmers to other market actors. Expanding the information content of existing networks and providing efficient technical support for the functioning of markets is, therefore, advocated for improved access to output markets (Amaya and Alwang 2011). However, it is worth noting that, despite the apparent high penetration of mobile technologies, the limited ICT knowledge among farmers, especially those in developing countries may undermine the capacity to improve access to market information (Dagne and Oguamanam 2018; Ogbeide and Ele 2015). Limited access to reliable internet services and the apparent lack of money to hook on to data services provided by telecom companies and other internet service providers constitute yet another market access constraint (Mwombe et al. 2014).

Recent studies have focused on collective action as an enabler to increasing farmers’ access to markets (Fikadu, Duguma, and Mitiku 2019; Fischer and Qaim 2012; Gyau et al. 2014; Markelova et al. 2009). Collective marketing is underpinned by two key motives: reductions in transaction costs and countering the market power of opportunistic buyers (Fischer and Qaim 2012). Gyau et al. (2014) synthesised the key lessons learned in using collective action to improve market access for agroforestry products in Cameroon. The authors noted that improvements in market access led to increased incomes and improved food security for smallholder farmers. However, market failures impact negatively the potential of linking farmers to output markets. Distance has also been shown to discourage dairy farmers from joining cooperatives for collective marketing (Fikadu, Duguma, and Mitiku 2019). The success of collective action, therefore, depends on the motivation of farmers, the design of group activities to include social activities and the environment (Gyau et al. 2014). Previously, Markelova et al. (2009) provided evidence on how the poor could benefit from improved market access through institutions engaged in collective action. The findings revealed that (a) collective action overcomes market failures, facilitates information access and enables large scale production and marketing (resource pooling), (b) group characteristics affect performance, (c) government and the private sector play complementary roles in overcoming the marketing constraints faced by producer groups and (d) better output prices and adoption to the changing global supply chains can result from collective action. However, to make collective marketing effective, it must be profitable and sustainable. Furthermore, Fischer and Qaim (2012) found positive income effects of collective marketing on banana farmers in Kenya. Collective marketing serves as a catalyst for adopting innovations due to better information flows and inclusiveness. Land ownership, access to a mobile phone, distance to paved roads and access to credit were found to significantly increase the likelihood of farmers’ seeking to join a collective marketing group. However, the price gains arising from collective marketing was very marginal.

Markets also have spillover effects on general wellbeing. The positive effects of market access on nutritional outcomes have been well documented (Koppmair, Kassie, and Qaim 2016; Stifel and Minten 2017). Koppmair, Kassie, and Qaim (2016) analysed the link between farm production and dietary diversity in rural Malawi and reported a positive significant relationship. Access to output markets is an important factor for dietary diversity, hence, the need for improvements in market access. Furthermore, Stifel and Minten (2017) analysed the link between individual/household wellbeing, nutrition and market access in Ethiopia and reported that distance affects food security and educational outcomes. Households residing in distant and remote areas were found to be more food insecure with lower rates of school enrolments than those closer to markets. These differences have been attributed to low production activities of households and unfavourable terms of trade which adversely reduce the size of the marketable surplus. Thus, market access is positively associated with household production and marketing activities, and the need for increased investments in agriculture exists. However, greater market access was insufficient in triggering improved nutritional outcomes.

Ma and Abdulai (2016) analysed the link between farmers’ access to markets and its determinants in China using the multinomial logit approach with two-stage selection criteria. The study revealed that access to credit, timely repayment, output price, extension contact and the volume of transactions positively and significantly influence farmer choice for written contracts, which increased their net returns. Cooperative sales and distance to markets were found to positively drive farmers’ choice for oral contracts. Thus, farmers who utilise written marketing contracts tend to benefit more compared to those using oral contracts.

In analysing the dimensions of market access following the post-liberalisation period (1997–2010) in Kenya, Chamberlin and Jayne (2013) found low correlations between indicators of market access with rural people experiencing a high degree of variations over space and time. Thus, changes in market access conditions are directly linked to the behaviour of market agents. Market access has multiple dimensions and hence difficult to be measured with a single index (Chamberlin and Jayne 2013). Previously, Zeller, Diagne, and Mataya (1997) reported that the transaction cost incurred by households’ in accessing the nearest input and output market outlet is negatively related to the share of the area cropped to hybrid maize. The authors concluded that improving rural infrastructure and facilitating access to markets is essential in transforming subsistence agriculture and enhancing the adoption of new technologies. Amrouk et al. (2013) showed that extension access, training and demonstration activities and building productive assets of private agricultural sector agencies impact significantly on market access and participation. Social and human capital factors also impact positively on the market participation of agro-processors (Thindisa and Urban 2018). Nonetheless, the main barriers to expanding farmers’ access to markets that require urgent address are related to quality, standards and export markets.

In summary, the review showed that farmer’s market choices are influenced by various factors: technological, institutional and socio-economic in nature. Key factors that limit farmer’s participation in markets are limited access to market information, poor roads, absence of guaranteed markets, limited access to extension services and credit. The use of mobile phones and existing networks have positive impacts on market information access likely to change the market choices and output prices received by farmers and other value chain actors. Collective marketing has positive income effects with spillover effects on food security, nutritional outcomes and market access for smallholder farmers. However, the profitability and sustainability of collective actions hinge on farmers’ motivation, the environment, design of group activities and distance. Hence, dealing with issues of market failures remains unresolved in the literature. The use of written contracts through beneficial to farmers is less used and the role of political factors in marketing is less explored in the literature. Finally, the use of various market channels (private sales, farm gate, village markets and regional markets) which serve as marketing opportunities for farmers has not been analysed in the literature and it is unclear what factors will influence a farmer to choose one market outlet over another. This remains a major gap which the current study seeks to address.

Methodology

Sampling Design and Data

The data used for this study was collected from a household survey conducted between February and May 2019 which covered six districts across northern Ghana. Two districts each were purposively selected from the Northern, Upper East, Upper West Regions of Ghana to analyse the determinants of farmers’ access to output markets. A total of 13 rural communities and 448 households were covered based on a programme intervention that seeks to promote SI practices among farming households for increased incomes and food security. The regions and communities were purposively selected whilst the households which were interviewed using a semi-structured questionnaire were randomly selected.

Variable Definitions and Measurement

Table 1 presents the variables used in the study, their measurements and descriptive statistics. The choice of these variables was informed by literature drawn from previous works that analysed farmers’ access to markets. This is followed by a discussion on the variables and their justification.

Table 1.

Variable Definitions and Measurement.

Variable	Description/Measurement	Obs.	Mean	SD
Sex of respondent (SEX)	1 if a respondent is male, 0 otherwise	445	1.202	0.40
Association membership (ASM)	1 if a farmer belongs to any producer/marketing/processing group, 0 otherwise	448	1.410	0.49
Processing facilities (PRF)	1 if a farmer has access to processing facilities in the community, 0 otherwise	447	1.348	0.47
Extension Services (EXT)	1 if a farmer has access to extension services, 0 otherwise	448	1.379	0.48
Storage facilities (STF)	1 if a farmer has access to storage facilities in the household/ community, 0 otherwise	445	1.337	0.47
New practices (NEP)	1 if a farmer is opened to new farming/marketing methods, 0 otherwise	447	1.277	0.44
Market quality (MKQ)	1 if cleaning, grading and drying services exists in the market, 0 otherwise	443	1.724	0.44
Access to financial services (AFS)	1 if a farmer has access to savings and loans locally, 0 otherwise	445	1.485	0.50
Sustainable intensification (SIT)	1 if a farmer has knowledge of sustainable intensification practices, 0 otherwise	448	1.292	0.45
Type of farming (TPF)	1 if a farmer is engaged in crop production, 0 otherwise	441	0.954	0.20
Type of farming (TPF)	1 if a farmer is engaged in livestock production, 0 otherwise	441	0.258	0.43
Guaranteed markets (GMK)	1 if there is guaranteed market for a farmer produce, 0 otherwise	446	1.820	0.38
Distance to market (DIM)	Average distance to the nearest input/output market in kilometres	444	18.31	16.84
Access to market information (AMI)	1 if a farmer has access to market information, 0 otherwise	448	1.238	0.42
Bargaining skills (BAS)	1 if a household has sufficient bargaining skills to secure good prices, 0 otherwise	443	1.613	0.48

Source: Various literature and field survey, 2019.

The sex of the farmer influences marketing activities, and females are mostly engaged in the sale and marketing of farm produce. The heavy involvement of women in markets constitutes a potential force that could alter market choices with a greater impact on gender (Amaya and Alwang 2011). The expectation is that women’s involvement will positively influence the choice of market channel to use in output markets.

Socially, association membership by smallholder farmers results in better advocacy and collective marketing (Fikadu, Duguma, and Mitiku 2019; Fisher and Qaim 2012; Gyau et al. 2014). Farmers belonging to production, processing, or marketing group often have better access to resources (inputs, finance, information), price advantages and markets with positive income effects. However, managing members’ expectations, enforcing rules and norms and addressing governance issues remain critical concerns in the proper functioning of such associations. Positive effects on output market choices are anticipated.

Access to processing facilities can impact significantly on output market decisions due to the potential for value addition. Processing improves the form of the commodity and makes it more attractive to consumers as they meet their needs. Investing in processing facilities, however, can be capital intensive which mostly falls outside what individual farmers can afford (Baloyi 2010). The weak nature of most farmer groups is also a disincentive for joint ownership and management of infrastructure facilities or services. Nonetheless, positive effects on marketing channel choices can be expected.

Access to extension services influence farmers’ production and marketing decisions through information sharing. Constraints in accessing efficient extension services affect knowledge of appropriate technologies and reduce productivity (Stifel and Minten 2017). Extension access impacts significantly on market participation (Amrouk et al. 2013) and positive effects are anticipated. Access to storage facilities prolongs the shelve lives of produce and therefore impact significantly on farmers’ access to output markets. Where appropriate storage facilities exist, farmers are able to store their produce especially during periods of glut and take advantage of higher prices during the lean season. Storage facilities also serve as a buffer in improving the local food security situation (Tefera 2012). The type of storage or facility, and safe storage practices adopted had significant effects on the marketing behaviour and participation of farmers (Chuma, Mudhara, and Govereh 2020).

Knowledge of new farming and marketing methods increases the participation of farmers in markets. The adoption of improved farming technologies/practices often leads to yield increases and increased marketing activity among farmers (Kassie et al. 2013). Positive effects on farmers’ market access channel choices are anticipated. The quality of markets (grading, drying and cleaning services) is expected to impact positively on output market decisions of farmers in different locations. For instance, in markets where cleaning services exist, farmers can secure good prices due to improved quality of produce.

Access to financial services can have a tremendous effect on farmers’ access to output markets. Stock acquisition and meeting marketing costs make financial services integral to reaching output markets. Access to credit significantly influences farmers’ access to markets (Amrouk et al. 2013), increased the chance of farmers’ membership to a group (Fischer and Quim 2012) and influences the choice of marketing channel used by rural assemblers positively (Kihoro et al. 2016). The positive effects of savings and loan services on farmers’ choice of market channels are anticipated. Farmers’ knowledge of SI practices would likely influence their market participation decisions especially with climate change impacts on production and rising soil infertility. SI practices (crop rotation, composting and maize stripping) improves soil fertility, increase yields and incomes of farmers. Positive effects are expected from farmers who adopt SI practices on their marketing decisions.

The type of farming activity engaged by a farmer influences their access to output markets. The type of crop/livestock, variety/breed and kind of value addition influences demand and supply patterns as well as market decisions. Farmers engaged in the production of commodities demanded by consumers are better placed to have access to lucrative output markets. The presence of guaranteed markets can serve as a catalyst for farmers to access output markets due to economies of scale. Through guaranteed markets, farmers are able to shift from non-market to formal market participation (Panda and Sreekumar 2012). Written contracts which are an element of guaranteed markets have also been shown to increase the farmers’ net profits (Ma and Abdulai 2016).

Distance impacts significantly on farmers’ access to output markets especially those residing in remote and distant locations (Stifel and Minten 2017). Distance to paved roads influences the decision of farmers to join a marketing group (Fischer and Qaim 2012) and travelling over a long distance to reach buyers impact significantly on farmers’ choice of market contracts (Ma and Abdulai 2016). However, Chamberlin and Jayne (2013) argue that distance is not a good measure of farmers’ access to markets but rather the behaviour of marketing agents reflects improvements in market access. Mixed effects are anticipated.

Access to market information impact on farmers’ choice of market outlets to use. Market information influences the marketing behaviour of households (Jari and Fraser 2009), the prices received (Goyal 2010), a variety of crops grown (Deressa et al. 2009) and the extent of market participation (Maponya et al. 2018; Obisesan[CE]:Author: The citations ‘Obisesan 2018’ and ‘Ma and Abdulai 2016’ are not present in the reference list. Please include them in the list with complete bibliographic details (article/chapter title, journal/book title, issue/volume number, page numbers, publisher’s information, whichever is applicable) or else allow us to delete them from the text. 2018). A direct relationship between farmers’ access to information and choice of output market is expected.

The level of bargaining skills of farmers influence their market participation and prices received. Farmers’ market participation and terms of trade impact negatively on the size of agricultural surplus marketed and the number of food items purchased (Stifel and Minten 2017). This suggests that bargaining skills matter for farmers as it impacts on their incomes. This study hypothesised that farmers have weak bargaining skills and therefore are unable to secure good output prices.

Multinomial Logistic Regression Model

The multinomial logit framework was developed mainly for modelling discrete choice or market shares (McFadden 1974). However, its application now extends to analysing budget share (Theil 1969), cost-shares (Considine and Mount 1984) and land shares. The approach is useful in examining strategic choices with multiple outcomes.

This study utilised the MLR in line with recent studies (Kihoro et al. 2016; Ma and Abdulai 2016; Panda and Sreekumar 2012). The model is an extension of the binary logistic regression and allows for more than two categories of the outcome variable. Independent variables can be binary, interval/ratio in scale, or continuous. The probability of categorical membership is evaluated following the maximum likelihood estimation procedure and care is needed in determining the sample size and estimation for outlying cases (Bayaga 2010).

The model assumes independence among the dependent variable choices rather than linearity, normality or homoscedasticity. Thus, the choice of one category is unrelated to the choice of any other category (El-Habil 2012).

The general form of the MLR is given as:

P r o b [Y i = j] = \frac{\exp (B^{'} j X t}{\sum_{m = 1}^{J} \exp (B^{'} m X t, j = 0, …, J}

(1)

where,

Prob [Yi = j] = the probability of choosing the market outlet (farm gate, village market, private sales, regional/district market)

Xi = a vector of the predictor variables (see Table 1)

Bj = vector of the parameters to be estimated; and

J = the number of market outlet in the choice set.

Thorough initial data analysis was conducted including bivariate, multivariate and univariate assessment. Specifically, multicollinearity was checked with simple correlations among the independent variables. Multicollinearity can occur in logistic regressions leading to inflated standard errors of the logit parameters. Where multicollinearity exists, the reliability of the estimated parameters is affected (Garson 2011). The absence of multicollinearity paved the way for the model estimation to be done. Interpretation of the logistic coefficients is based on the idea that a unit change in the predictor will result in some amount of expected change in the logit. The closer the value of the coefficient to zero, the less the predictive power of the logit.

Results and Discussions

Descriptive Statistics

The main market outlets used by farmers were village markets (59.75%), private sales (20.4%), regional/district markets (11.5%), farm gate (6.45%) and roadside (1.9%). Improving market infrastructure especially of village markets will greatly enhance farmers’ access to input and output markets. In-depth interviews conducted with various District Assemblies showed that efforts were underway in developing the rural market infrastructure. However, the unwillingness of traders to relocate to some new market sites constructed is a major challenge due to cultural beliefs and the fear of losing their customers. This suggests the need for proper participatory consultations in the sitting of new markets in communities.

Access to extension services by farmers was low (61.5%). Public extension agents typically provide information to farmers on good agricultural practices, output prices, certified seeds and fertilizers. Most farmers (76.55%) have access to inputs due to the government subsidy programme on seeds and fertilizers coupled with interventions by private agricultural companies and NGOs in the area. There are no gender barriers in access to inputs. Most farmers (75.35%) were using planting seeds obtained from their own farms with only a few (14.55%) relying on Agro-input dealers for planting materials. Some communities indicated that they do not have input dealers. The quality and availability of inputs were found not to be an issue, but the price of fertilizers was reported to be high despite the government subsidy programme. About half (51.05%) of the farmers had access to savings and loan services from group-based associations (such as Village Savings and Loan Association) but access to formal finance was non-existent. From the viewpoint of farmers, most financial institutions are not willing to provide credit to farm-based enterprises due to the high rates of default (bad credit history) and the inability of most farmers to provide collateral demanded by these financial institutions.

Farmers’ access to guarantee market was low (18.3%) with a likely impact on their market participation decisions. These were mainly farmers with verbal contract agreements with either input dealers or traders who pre-finance their production activities and in return receive grains after harvest. About 59.45% of the households belong to one form of association or the other. The social groups/associations are useful in providing communal farm labour and savings mobilisation that support members. Food processing dominates in value-added activities in the area (61.7%) but the absence of processing equipment remains a major hindrance to most farmers.

Most livestock farmers covered in the study owned poultry (34.03%), goats (29.87%) and sheep (20.79%) with less than 10% keeping pigs and cattle. Poultry and livestock play important social, economic and cultural roles in the lives of households in northern Ghana. Access to credit for livestock production and marketing is limited and the majority (80.2%) of livestock farmers were dependent on their own finances. Only about 8.6% of farmers had access to formal credit and only 10% of farmers could self-finance their operations (mostly dependent on friends and relatives). Theft, high mortality rates and limited availability of veterinary drugs and services are the main constraints hampering the development of the livestock sector.

Specific livestock markets exist within most (70%) districts and prices of livestock were mainly determined through bargaining (69.3%) and farmers with good bargaining skills are better placed in securing good prices. The main factors that limit the efficient participation of farmers in livestock markets include pests and diseases outbreak and limited access to finance (see Figure 1). This is in line with Amankwah et al. (2012) empirical finding that weak support systems for animal health service delivery and production are a key structural constraint hindering small ruminant farmers’ participation in markets. Major players in the livestock market were individual traders/buyers (52%), farmers (18%), assemblers (12%) and food sellers (17%).

Figure 1.

Source: Field survey, 2019.

Factors Influencing Farmers’ Choice of Output Market Outlets

Table 2 presents the results obtained from the MLR using Village Market as the base outcome. The probability chi-square value as well as the pseudo R² statistic shows that the model is well fitted to the data with a high predictive power (57%).

Table 2.

Multinomial Logistic Regression of Farmers’ Choice of Output Market Outlet.

Variables	Farm Gate			Private Sales			District/Regional Markets
Variables	Coeff./Std. Err	z	P>\|z\|	Coeff./Std. Err	z	P>\|z\|	Coeff./Std. Err	z	P>\|z\|
Sex of respondent (SEX)	−0.587(2.152)	−0.27	0.785	1.335**(0.643)	2.07	0.038	0.487(0.764)	0.64	0.523
Association membership (ASM)	−5.755*(3.197)	−1.80	0.072	1.952***(0.590)	3.31	0.001	1.852***(0.748)	2.48	0.013
Processing facilities (PRF)	13.005**(5.135)	2.53	0.011	0.352(0.628)	0.56	0.575	1.863***(0.606)	3.07	0.002
Extension services (EXT)	13.483*(6.896)	1.96	0.051	−2.848***(1.056)	−2.70	0.007	−0.844(0.903)	−0.94	0.350
Storage facilities (STF)	10.068**(4.765)	2.11	0.035	−0.681(0.646)	−1.05	0.291	−1.314**(0.677)	−1.97	0.052
New practices (NEP)	−9.804**(4.576)	−2.14	0.032	0.011(0.666)	0.02	0.986	0.440(0.742)	0.59	0.553
Market quality (MKQ)	−12.786**(5.648)	−2.26	0.024	1.065(0.724)	1.47	0.141	−0.525(0.918)	−0.57	0.567
Access to financial services (AFS)	−3.018(2.253)	−1.19	0.234	−3.069***(0.545)	−5.63	0.000	−0.267(0.752)	−0.36	0.722
Sustainable intensification (SIT)	−12.940**(5.687)	−2.28	0.023	−0.611(0.524)	−1.17	0.244	0.8987(0.5984)	1.50	0.133
Crop production (TYP)	1.424(2.970)	0.48	0.632	−0.931(1.018)	−0.91	0.360	0.492(1.694)	0.29	0.771
Livestock production (TYP)	−7.828*(4.099)	−1.91	0.056	−0.242(0.700)	−0.35	0.729	−17.69(4210.4)	−0.00	0.997
Guaranteed markets (GMK)	−16.974**(7.524)	−2.26	0.024	−0.9489(0.750)	−1.26	0.206	18.385(7145.6)	0.00	0.998
Distance to market (DIM)	−0.076(0.077)	−0.99	0.323	0.009***(0.002)	3.51	0.000	−0.001(0.003)	−0.57	0.572
Access to market information (AMI)	−1.457(2.367)	−0.62	0.538	−0.905(0.960)	−0.94	0.346	−4.174***(1.359)	−3.07	0.002
Bargaining skills (BAS)	−5.380*(2.708)	−1.99	0.047	1.432*(0.768)	1.86	0.062	19.418(3517.3)	0.01	0.996
Constant	11.138*(5.593)	1.99	0.046	0.270(1.316)	0.21	0.837	−39.911(7964.4)	−0.01	0.996

Source: Estimation from field data, 2019.

Note: *, **, *** represent 10%, 5% and 1% level of significance, respectively.

Also, standard errors are in parenthesis. Number of observations = 448; LR Chi2 ( 64 ) = 398.03; Prob > Chi2 = 0.000; Pseudo R2 = 0.5779; Log Likelihood = −145.36556

The results in Table 2 are discussed under three main sub-headings by comparing each market outlet (farm gate, private sales and regional/district markets) with the base outcome. The interpretations are based on the fact that a negative coefficient (less than one) signifies a choice of the base outcome over the comparison group and vice versa. 1.

Farm gate versus village market

The results show that access to processing and storage facilities, belonging to an association, extension service access, exposure to improved farming and marketing methods, access to finance, guaranteed markets, knowledge of SI practices, livestock production and bargaining skills are significant factors that influence farmers choice of using farm gate over village markets in selling their produce. A farmer not belonging to a production/processing/marketing group in the community is associated with a 5.755 decrease in the relative log odds of using farm gate as their main market outlet. Thus, farmers who do not belong to a producer or marketing group are more likely to sell at the village market over the farm gate.

Farmers’ access to processing services is posited to have positive effects on their access to output markets. The results obtained here are in line with this theoretical expectation and a farmer not having access to processing facilities is associated with a 13.005 increase in the relative log odds of using farm gate as the main market outlet. This means that farmers without access to processing facilities are more likely to sell at the farm gate. This is intuitive since in the absence of value addition the drive to sell in village markets diminishes due to additional handling costs likely to be borne by the farmer. As noted by Musah, Bonsu, and Seini (2014), smallholder farmers often prefer to sell at the farm gate for reasons of convenience despite the fact that better prices are offered in distant and well-developed markets.

Access to extension services has a positive significant effect on market access. A farmer without access to extension services is associated with a 10.069 increase in the relative log odds of using farm gate as the main market outlet. Thus, the likelihood of farmers to market their products through the farm gate channel are higher where they lack access to extension services. Weak animal health extension delivery has been reported as a structural limitation to livestock farmers’ participation in markets (Amankwah et al. 2012). As such, only a few farmers engaged in market-oriented livestock production. Access to storage facilities is associated with a 9.804 decrease in the relative log odds of using farm gate as the main market outlet. Thus, the absence of storage facilities in a household or community levels is associated with a greater likelihood of farmers to sell at the village market over the farm gate.

A farmer not having access to new farming and marketing methods is shown to result in a 12.786 decrease in the choice of using the farm gate as the main market outlet. Thus, farmers’ access to new marketing and farming methods is associated with an increased likelihood of selling at the village market. This is supported by Achandi and Mujawamariya (2016) observation that the existence of a market locally impacts on the quantity of products marketed. Access to quality marketing services (cleaning, grading and standardisation) can impact farmers’ market access. Access to quality marketing services is shown to be associated with a 12.786 decrease in using village markets. Farmers with limited access to quality marketing services are more likely to sell at the farm gate. This outcome is in line with the findings of a recent study by Anthony and Lenah (2020) that tarpaulin ownership and sorting of produce positively influence smallholder maize farmers participation in formal markets.

A farmer not having knowledge in SI practices is associated with a 9.045 decrease in the relative log odds of using farm gate as the main market outlet. Thus, a farmer with knowledge in SI is more likely to use the village market over the farm gate in marketing their produce. SI practices such as organic farming increase the incomes of farmers and also create new market opportunities especially in urban areas. Mpombo (2018) found that the use of inorganic fertilizers and animal traction negatively influences the market orientation of farmers.

The rearing of livestock is shown to decrease the likelihood of using farm gate as the main market outlet by 7.828. Thus, livestock farmers are more likely to sell at the village market over the farm gate. This is explained by the relatively large number of buyers in the village markets which makes it more competitive as compared with the farm gate. Dlamini and Huang (2019) found that herd size, type of cattle breed and experience in production significantly affects the sales and market participation of smallholder beef farmers.

Access to guaranteed markets is associated with 16.975 decreases in the relative log odds of using farm gate as the main market outlet. Thus, farmers without access to guaranteed markets will prefer to sell at the village market than at the farm gate. Where there are guaranteed markets, farmers will like to keep their profit levels by curtailing transport and other costs involved in moving the product from the farm gate to the village market. They will, therefore, prefer to sell at the farm gate.

Farmers without sufficient bargaining skills to secure good prices are associated with a 1.458 decrease in the relative log odds of using farm gate as the main market outlet. Thus, farmers without sufficient bargaining skills are more likely to use village markets over the farm gate to sell their outputs. Groups can, however, influence market prices through collective bargaining (Musah, Bonsu, and Seini 2014) irrespective of the market channel. 2.

Private sales versus village markets

The results of Table 2 further indicated that gender, association membership, distance to output market and bargaining skills of the farmer are significant factors that influence the choice of farmers to do private sales over village markets. However, access to extension services and finance will likely influence the choice of farmers to use village markets over private sales.

Females are associated with a 1.335 increase in the relative log odds of using private sales as the main market outlet. Thus, female farmers are more likely to utilise private sale outlets than village markets to sell their produce. A farmer not belonging to a producer, processing or marketing group in a community is associated with a 1.952 increase in the relative log odds of using private sales as the main market outlet. Thus, farmers who are not members of these associations or groups would prefer private sales to the use of village markets. This could be attributed to the level of trust that is built between farmers and individual off-takers over time.

Access to extension and financial services decreases the odds of using private sales as the main market outlet by 2.848 and 3.009, respectively. Farmers without access to the extension or financial services would likely choose to sell their output in village markets over private sales. Access to credit influenced market participation positively (Anthony and Lenah 2020; Musah, Bonsu, and Seini 2014). Also, access to extension services has been reported to impact positively on farmers’ market participation (Aliyi, Tadesse and Demise 2018; Mpombo 2018).

The average distance to the nearest output market covered by a farmer is associated with a 0.009 increase in the relative log odds of using private sales as the main market outlet. Thus, a unit increase in the farmer’s average distance to the nearest output market would more likely influence their choice for private sales over village markets. Fisher and Qaim (2012) showed that distance to paved roads increases the likelihood of farmers seeking to join collective marketing groups. Also, the bargaining skills of a farmer is related to a 1.432 increase in the log odds of using private sales as the main market outlet. This means that without sufficient bargaining skills farmers are more likely to use private sales over village markets. 3.

District/regional markets versus village markets

Table 2 revealed that access to processing facilities and belonging to an association are significant factors that influence farmers’ choice of using district and regional markets for the sale of their produce. Similarly, farmers access to storage facilities and access to marketing information decreases the likelihood of using village markets over district/regional markets.

Association membership by farmers is characterised by with 1.852 decrease in the log odds of using district and regional markets as the main market outlet. Thus, farmers who do not belong to any group are more likely to use village markets over the district and regional markets. This could be explained by the lower transaction costs associated with selling through the village market though price disadvantages exist. Cooperative membership positively influences market participation (Aliyi, Tadesse, and Demise 2018) and groups enable collective bargaining which results in higher prices for members (Musah, Bonsu, and Seini 2014).

A farmer not having access to processing facilities (either personally or in the community) increases the likelihood of using district/regional markets as the main market outlet by 1.863. Thus, farmers without access to processing facilities would prefer to sell in the district and regional markets to village markets. Aliyi, Tadesse, and Demise (2018) found that value-added activity has a positive influence on the intensity of market participation. Farmers having access to storage facilities is associated with 1.314 decreases in the log odds of using district and regional markets as the main market outlet. This suggests that households without access to storage facilities are more likely to sell their outputs in the village market over the district and regional markets. This is partly attributed as distance and the associated high transaction costs that come with selling in regional markets. Though output prices have been found to influence selling via formal markets (Anthony and Lenah 2020), the absence of storage facilities remains a hindrance in the development of rural markets.

A farmer’s access to market information in the locality is linked to a 4.174 decrease in the relative log odds of using district and regional markets as the main market outlet. Thus, a farmer without access to market information would more likely sell at the village market rather than use district and regional markets. Market information is, therefore, crucial to farmers as it is able to direct their activities to lucrative markets for participation. This finding lends support to previous studies that analysed the role of market information on market participation and reported positive influence (Aliyi, Tadesse, and Demise 2018; Musah, Bonus, and Seini 2014). However, it contradicts that of Anthony and Lenah (2020) who reported that the source of market information negatively influences formal market participation. District and Metropolitan Assemblies need to pay more attention to providing the needed storage facilities in markets for increased patronage by farmers.

Conclusions and Recommendations

This study contributes to the agricultural marketing literature by analysing the factors that influence farmers’ choices for various market outlets (private sale, farm gate, village market and regional markets) using the MLR. End markets are critical for smallholder farmers and other value chain actors as they impact on their profit margins, incomes, crop choice and production decisions. Yet, studies that focus on analysing and identifying the main factors that influence farmers’ choice of markets has remained very limited. By using village markets as the base category, various factors were found to influence farmers’ choices for the market outlets considered. The main significant factors that influence farmers’ choice to sell output at the farm gate was access to inputs, access to processing facilities, inputs and extension services. Effective extension service delivery is critical in influencing farmers’ decisions to invest in sustainable agricultural practices for increased yields (Kassie et al. 2013). Also, access to storage facilities, association membership (production, processing, or marketing), openness to new production and marketing methods, knowledge of SI practices, access to financial services, availability of quality market services (cleaning, grading and drying), access to guaranteed market and distance to output market would likely influence the choice of farmers to sell in village markets. As noted by Dalton et al. (2014) and Tappan and McGahuey (2005), the adoption of SI practices decreases farmers’ cost of production, improves productivity and reshapes natural resources. Additionally, the sex of the farmer, association membership, access to processing facilities, availability of quality market services and distance to output market would likely dominate the choice of a farmer to do private sales. These findings are relevant for farmers, marketers, input dealers and policymakers, especially those at the local level interested in promoting farmer participation in output markets. One area in which this work could be extended is to analyse the gender aspect with a focus on livestock markets.

Footnotes

Acknowledgements

The authors are grateful for the support received from the Africa RISING project team members both in STEPRI and IITA Tamale offices during data collection. Special thanks to Dr Fred Kizito and Dr Hoeschle-Zeldon Irmgard of IITA.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

This research work is an output of the Africa RISING Project supported by the United States Agency for International Development under the Feed-the-Future initiative [AID-BFS-G-11-00002].

ORCID iD

Abdulai Adams

References

Achandi

E. L.

, Mujawamariya

2016. ‘Market Participation by Smallholder Rice Farmers in Tanzania: A Double Hurdle Analysis.’ Studies in Agricultural Economics 118: 112–15. https://dx-doi-org.web.bisu.edu.cn/10.7896/j.1528.

Aker

J. C.

2008. ‘Does Digital Divide or Provide? Information Technology, Search Costs and Cereal Market Performance in Niger.’ Bureau for Research and Economic Analysis of Development (BREAD), Working Paper No. 177 Duke University, Durham, NC.

Aliyi

, Tadesse

, Demise

2018. ‘Determinants of Intensity of Market Participation of Smallholder Rice Producers: The Case of Shebe Sombo District, Southwest Ethiopia’. Journal of Agricultural Economics and Development 7(3) : 13–22.

Amankwah

, Klerkx

, Oosting

J. S.

, Sakyi-Dawson

, Van der Zijpp

A. J.

, Millar

2012. ‘Diagnosing Constraints to Market Participation of Small Ruminant Producers in Northern Ghana: An Innovation Systems Analysis.’ NJAS—Wageningen Journal of Life Sciences 60–63: 37–47.

Amaya

, Alwang

2011. ‘Access to Information and Farmers Market Choices: The Case of Potato in Highland Bolivia.’ Journal of Agricultural Food System and Commodity Development 1 (4): 35–53. https://dx-doi-org.web.bisu.edu.cn/10.5304/jafscd.2011.014.003.

Amikuzuno

2015. ‘An Analysis of Pro-poor, Gender-sensitive Market Access Opportunities and Initiatives for Smallholder Farmers in the Upper East and West Regions of Ghana.’ Market Survey Report of the RESULTS Project , pp. 1–81.

Amrouk

, Poole

, Mudungwe

, Muzvondiwa

2013. ‘The Impact of Commodity Development Projects on Smallholders’ Market Access in Developing Countries: Case Studies of FAO/CFC Projects.’ FAO Commodity and Trade Research Working Paper No 35, FAO.

Anthony

, Lenah

2020. ‘Determinants of Smallholder Maize Farmers’ Choice of Selling Point: A Case of Kwanza District, Trans-Nzoia County, Kenya.’ Journal of Economics and Sustainable Development 11 (10): 147–51.

Baloyi

J. K.

2010. ‘An Analysis of Constraints Facing Smallholder Farmers in the Agribusiness Value Chain: A Case Study of Farmers in the Limpopo Province.’ Doctoral dissertation, University of Pretoria.

10.

Bayaga

2010. ‘Multinomial Logistic Regression: Usage and Application in Risk Analysis.’ Journal of Applied Quantitative Methods , 5(2): 288–297.

11.

Binge

, Mshenga

, Kgosikome

2019. ‘Production and Marketing Constraints of Small Stock Farming: Evidence from Limited and Non-Limited Farmers in Boteti Sub-District, Botswana.’ Journal of Agribusiness and Rural Development 3(53): 195–201.

12.

Chamberlin

, Jayne

T. S.

2013. ‘Unpacking the Meaning of “Market Access”: Evidence from Rural Kenya.’ World Development 41: 245–64. https://dx-doi-org.web.bisu.edu.cn/10.1016/worlddev.2012.06.004.

13.

Considine

T. J.

, Mount

T. D.

1984. ‘The Use of Linear Logit Models for Dynamic Input Demand Systems.’ Review of Economics and Statistics 66 (3): 434–43.

14.

Corsi

, Marchisio

L. V.

, Orsi

2017. ‘Connecting Smallholder Farmers to Local Markets: Drivers of Collective Action, Land Tenure and Food Security in East Chad.’ Land Use Policy 68: 39–47.

15.

Chuma

, Mudhara

, Govereh

2020. ‘The Effects of Grain Storage Technologies on Maize Marketing Behavior of Smallholder Farmers in Zimbabwe.’ Journal of Agriculture and Rural Development in the Tropic and Subtropics 121 (1): 1–12.

16.

Dagne

T. W.

, Oguamanam

2018. ‘ICTs in Agricultural Production and Potential Deployment in Operationalizing Geographical Indications in Uganda.’ Ottawa Faculty of Law Working Paper No. 2018-24. https://dx-doi-org.web.bisu.edu.cn/10.2139/ssrn.3241169.

17.

Dalton

T. J.

, Yahaya

, Naab

2014. ‘Perceptions and Performance of Conservation Agriculture Practices in Northwestern Ghana.’ Agricultural Ecosystem and Environment 187: 65–71.

18.

Deressa

T. T.

, Hassan

R. M.

, Ringler

, Alemu

, Yesuf

2009. ‘Determinants of Farmers’ Choice of Adaptation Methods to Climate Change in the Nile Basin of Ethiopia.’ Global Environmental Change 19 (2): 248–55.

19.

Dlamini

S. I.

, Huang

2019. ‘A Double Hurdle Estimation of Sales Decisions by Smallholder Beef Cattle Farmers in Eswatini.’ Sustainability 11: 15158. http://doi.org/10.3390/su11195185.

20.

El-Habil

A. M.

2012. ‘An Application on Multinomial Logistic Regression Model.’ Pakistani Journal of Statistics and Operations Research 8 (2): 271–91.

21.

FAO. 2013. Synthesis of Guiding Principles on Agricultural Programming for Nutrition . Rome: Food and Agriculture Organization.

22.

Fikadu

, Duguma

, Mitiku

2019. ‘Pull and Push Factors for Producers’ Membership in Dairy Marketing Cooperatives in Jimma Zone, Oromia, Ethiopia.’ Journal of Agribusiness and Rural Development 1 (51): 21–34. https://dx-doi-org.web.bisu.edu.cn/10.17306/J.JARD.2019.01162.

23.

Fischer

, Qaim

2012. ‘Linking Smallholders to Markets: Determinants and Impacts of Farmer Collective Action in Kenya.’ World Development 40(6): 1255–68. https://dx-doi-org.web.bisu.edu.cn/10.1016/worlddev.2011.11.018.

24.

GAIN. 2013. Routes for Nutrition Secure Cash Crop Value Chains . Amsterdam: Global Alliance for Improved Nutrition.

25.

Garretsen

, Bosker

2012. ‘Market Access: A Key Determinant of Economic Development in Sub-Saharan Africa.’ The World Bank. Available at: https://blogs.worldbank.org/developmenttalk/market-access-a-key-determinant-of-economic-development-in-sub-saharan-africa(Accessed on 07 May 2019).

26.

Garson

G. D.

2011. Logistic regression. Available at: http://faculty.chass.ncsu.edu/garson/PA765/logistics.htm#sigtets (Accessed on 12 February 2019).

27.

Goyal

2010. ‘Information, Direct Access to Farmers, and Rural Market Performance in Central India.’ American Economic Journal of Applied Economics 2 (3): 22–45.

28.

Gyau

, Franzel

, Chiatoh

, Nimino

, Owusu

2014. ‘Collective Action to Improve Market Access for Smallholder Producers of Agroforestry Products: Key Lessons Learned with Insights from Cameroon’s Experience.’ Current Opinion in Environmental Sustainability 6: 68–72.

29.

ILO. 2017. ‘Improving Market Access for Smallholder Farmers: What Works in Out-Grower Schemes – Evidence from Timor-Lester.’ Issue Brief No. 1.

30.

Jari

, Fraser

G. C. G.

2009. ‘An Analysis of Institutional and Technical Factors Influencing Agricultural Marketing amongst Smallholder Farmers in the Kat River Valley, Eastern Cape Province, South Africa.’ African Journal of Agricultural Research 4 (11): 1129–37.

31.

Kassie

, Jaleta

, Shiferaw

, Mmbando

, Mekuria

2013. ‘Adoption of Interrelated Sustainable Agricultural Practices in Smallholder Systems: Evidence from Rural Tanzania.’ Technological Forecasting and Social Change 80(3) : 525–40.

32.

Kihoro

E. M.

, Irungu

, Nyikal

, Maina

I. N.

2016. ‘An Analysis of Factors Influencing Farmers’ Choice of Green Gram Marketing Channel in Mbeere South Sub-Country, Kenya.’ Poster Presented at the 5^th International Conference of AAAE, September 23–26. Addis Ababa, Ethiopia: AAAE.

33.

Koppmair

, Kassie

, Qaim

2016. ‘Farm Production, Market Access and Dietary Diversity in Malawi.’ Public Health Nutrition 20(2): 325–35.

34.

, Abdulai

2016. ‘Linking Apple Farmers to Markets: Determinants and Impacts of Marketing Contracts in China.’ China Agricultural Economic Review 8(1): 2–21. https://dx-doi-org.web.bisu.edu.cn/10.1108/CAER-04-2015-0035.

35.

Maponya

, Kekana

, Senolo

G. M.

, Venter

S. L.

2018. ‘Socio-Economic Factors Influencing Market Participation of Horticultural Smallholder Farmers in the Alfred Nzo District, Eastern Cape, South Africa.’ Journal of Agribusiness and Rural Development 4 (50): 421–7. https://dx-doi-org.web.bisu.edu.cn/10.17306/J.JARD.2018.00421.

36.

Markelova

, Meinzen-Dick

, Hellin

, Dohrn

2009. ‘Collective Action for Smallholder Market Access.’ Food Policy 34: 1–7. http://doi.org/10.1016/j.foodpol.2008.10.001.

37.

McFadden

1974. ‘Conditional Logit Analysis of Quantitative Choice Behavior.’ In Frontiers of Econometrics , edited by P. Zarembka. New York: Academic Press.

38.

Musah

A. B.

, Bonsu

O. A. Y.

, Seini

2014. ‘Market Participation of Smallholder Farmers in the Upper West Region of Ghana.’ African Journal of Agricultural Research 9(31): 2427–35.

39.

Musemwa

, Mushunje

, Chimonyo

, Fraser

, Mapiye

, Muchenje

(2008). ‘Nguni Cattle Marketing Constraints and Opportunities in the Communal Areas of South Africa: Review.’ African Journal of Agricultural Research 3(4): 239–45.

40.

Mpombo

L. B.

2018. ‘Determinants of Market Participation among Smallholder Farmers in Tanzania.’ M.A dissertation, University of Dar es Salaam.

41.

Mwombe

S. O.

, Mugivane

F. I.

, Adolwa

I. S.

, Nderitu

J. H.

2014. ‘Evaluation of Information and Communication Technology Utilization by Smallholder Banana Farmers in Gatanga District, Kenya.’ The Journal of Agricultural Education and Extension 20(2): 247–61.

42.

Obisesam

A. A.

2018. ‘Market Participation and Food Security of Cassava Farmers in Rural South West Nigeria’. Journal of Agribusiness and Rural Development 1(47): 57–64. https://dx-doi-org.web.bisu.edu.cn/10.17306/J.JARD.2018.010399.

43.

Ogbeide

O. A.

, Ele

2015. ‘Smallholder Farmers and Mobile Phone Technology in sub-Sahara Agriculture.’ Mayfair Journal of Information and Technology Management in Agriculture 1(1): 1–19.

44.

Overa

2006. ‘Networks, Distance and Trust: Telecommunications Development and Changing Trading Practices in Ghana.’ World Development 34 (7): 1301–15.

45.

Panda

R. K.

, Sreekumar

2012. ‘Marketing Channel Choice and Marketing Efficiency Assessment in Agribusinesses.’ Journal of International Food and Agribusiness Marketing 24(3): 213–30. https://dx-doi-org.web.bisu.edu.cn/10.1080/08974438.2012.691812.

46.

Stifel

, Minten

2017. ‘Market Access, Well-Being, and Nutrition: Evidence from Ethiopia.’ World Development 90: 229–41. https://dx-doi-org.web.bisu.edu.cn/10.1016/worlddev.2016.09.009.

47.

Tappan

, McGahuey

2005. ‘Tracking Environmental Dynamics and Agricultural Intensification in Southern Mali.’ Agricultural Systems 94: 38–51.

48.

Tefera

2012. ‘Post-Harvest Losses in African Maize in the Face of Increasing Food Shortage.’ Food Security 4(2): 267–77.

49.

Theil

1969. ‘A Multinomial Extension of the Binary Logit Model.’ International Economic Review 10 (3): 251–59.

50.

Thindisa

L. M. V.

, Urban

2018. ‘Human-Social Capital and Market Access Factors Influencing Agro-Processing Participation by Small-Scale Agripreneurs: The Moderating Effects of Transaction Costs.’ Acta Commercii 18(1): a500. http://doi.org/10.4102/ac.v18i1.500.

51.

Zeller

, Diagne

, Mataya

1997. ‘Market Access by Small Smallholder Farmers in Malawi: Implications for Technology Adoption, Agricultural Productivity and Crop Income.’ IFPRI FCND Discussion Paper No. 35. Washington DC.