Smallholders’ vulnerability in the maize market: An analysis of marketing channels to improve the role of cooperatives in Benin

Abstract

Smallholder households in developing countries often face challenges related to market access. Despite being proposed as a potential solution, cooperatives in the north of Benin have reportedly failed to effectively address this issue. The alternatives channels used by smallholder famers and the factors that influence their choices remain unclear. This paper adopts a mixed approach, combining qualitative analysis with quantitative methods, particularly the multivariate probit model, to investigate the distribution channels of maize used by farmers in the Kandi district of Benin. The study aims to identify the current distribution channels of maize in Kandi and analyze factors that affect, smallholder producers choice of marketing channel. Initially, the study identified four primary channels for marketing maize—namely, collectors, wholesalers, brokers, and cooperatives—and producers typically commonly engage in multiple channels simultaneously. Subsequently, It becomes evident that factors such as age, cooperative membership, distance to market, reliance on informal credit, average maize bag prices, and the timing of sales significantly influence producers' selection of marketing channels. Interestingly, despite cooperatives offering comparatively higher prices, the majority of farmers opt for collectors due to the informal credit they provide and their practice of purchasing large volumes of maize bags at pre-agreed prices. In conclusion, enhancing joint-selling capabilities of cooperatives and establishing a credit provision business tailored to maize producers are identified as crucial steps. These measures aim to alleviate producers’ financial strain, diminish their dependence on collectors for credit, and bolster their bargaining power in the market.

1. Introduction

Maize is central to Benin's agricultural landscape, serving as the nation's primary staple food [1,2], and the most cultivated cereal, constituting up to 72.99% of the national cereal production. Its significance extends beyond mere subsistence, as it plays a pivotal role in ensuring food security; with an annual per capita consumption surpassing 136 kg, the sector is largely dependent on smallholder production [3]. Traditionally cultivated in the southern and central regions, maize production has progressively expanded into the northern cotton-growing areas [4]. This geographical shift underscores the crop's evolving importance, transitioning from being a staple for self-consumption to emerging as a substantial cash crop, particularly in the northern regions.

As a result, between 2010 and 2014, maize production in Benin increased by 30% owing to a confluence of factors, including declining sorghum yields, crop rotation benefits [5], and decreasing cotton profitability due to volatile international market prices [6]. The shift towards maize presents both an opportunity and a challenge, especially in marketing and sales, which are pivotal to the livelihoods of smallholder farmers. The expansion of maize production areas by 40.12% in the Kandi region over the last decade (see Appendix 2) reflects a broader trend toward market-oriented agriculture [7,8]. However, this has been met with an absence of structured marketing channels, especially concerning agricultural cooperatives that are intended to enhance farmers' bargaining power [9].

Previous studies on maize marketing in Benin have primarily focused on the southern regions [[10], [11], [12], [13]], revealing that maize is marketed primarily in the form of dry grains [14], with a high degree of intra-seasonal price variability. Prices tend to surge dramatically, increasing by up to 200% from the harvest period to the lean season in certain areas [15]. Moreover, the maize marketing system is characterized by private sector trading among farmers, with traders assuming a prominent role. Within this framework, wholesalers and retailers play crucial roles in the distribution of maize [11,12]. Traders typically procure maize during the harvest season when it is abundant, and prices are comparatively low [16] and subsequently sell it during the lean season when prices are significantly higher.

Furthermore, factors such as access to credit and storage facilities emerged as pivotal determinants of producers' revenue when marketing maize endeavors [17]. These resources enable producers to store the product, alleviate financial pressure, and effectively navigate through the lean season, thereby contributing to the sustainability of their livelihood. However, smallholders’ vulnerability is exacerbated by microfinance institutes' limited allocation of credit to agriculture, with only 16.4% of funds earmarked for this sector. Similarly, data from the census indicated that a mere 6.9% of farm households in Benin have access to credit [18]. Consequently, persistent challenges such as low productivity and post-harvest losses, which can amount to as much as 40%, constrain smallholder farmers' achievable outputs and incomes [19]. These challenges underscore the precarious situation faced by smallholders, further amplifying their vulnerabilities within the agricultural sector.

Theoretically, agricultural cooperatives in developing countries are recognized for their role in increasing producers' bargaining power in facilitating access to market and offering better prices [[20], [21], [22], [23]]. Well-organized farmers will be able to bypass assemblers and brokers in rural markets and connect directly with urban wholesalers, high-value retailers, and processors or exporters [24]. Despite efforts towards fostering joint selling within farmers’ organizations evidenced by previous studies, the reality in Benin presents a stark contrast [17]. In most cases, farmers in Benin perform maize marketing individually [25], signaling either the absence or the failure of maize cooperative joint-selling initiatives. This scenario is particularly pronounced in the northern regions, where prior research has revealed challenges related to the development of maize-producer cooperatives, including their limited capacity to engage in joint-selling activities [26]. To the best of our knowledge, no prior studies have investigated maize marketing in this geographical area. Previous research conducted in Benin has primarily focused on the southern regions and did not extensively analyze the decision-making factors influencing farmers' choice of marketing channels.

The purpose of this study is to comprehensively analyze the dynamic of maize market, with a particular emphasis on the role of maize cooperatives and avenues for improvement. By adopting a mixed-methods approach, the study aims to achieve two objectives. Firstly, based on qualitative data gather through insights from multiple stakeholders, the study seeks to identify maize marketing channels, price fluctuations and mechanisms in each channel. Secondly, using econometric analysis based on farm household data, the study aims to analyze factors influencing smallholder decision-making in maize marketing. The study diverges from the predominant reliance on quantitative data from farm households seen in the existing literature. It contributes in enhancing knowledge on maize marketing practices in the region by capturing a broader spectrum of perspectives and insights. Also, it identifies specific strategies and services that cooperatives can use for a sustainable and inclusive joint-selling business.

The paper is structured into five sections. The following Section (Section 2) synthesizes relevant literature pertaining to smallholders' vulnerability, maize marketing channels, and factors influencing farmers’ choices in the context of agricultural marketing. Section 3 outlines the research design and methodology employed in this study. Section 4 presents the empirical findings derived from both qualitative and quantitative analyses. Lastly, Section 5 encapsulates the conclusions drawn from the study and delineates their implications for practice, policy, and future research.

2. Literature review

2.1. Smallholders’ vulnerability in market access

In developing countries, smallholder farmers face significant challenges in accessing markets to sell their produce and purchase quality inputs, which lower productivity, making them highly vulnerable to income and food insecurity [27,28]. Market access is hampered by poor road infrastructure, lack of storage and processing facilities, high transportation costs, inadequate market information, and limited bargaining power [[29], [30], [31]]. This reduces farmers' ability to obtain fair prices for their crops, resulting in low and volatile incomes.

Moreover, the reliance on informal local markets characterized by few buyers exposes smallholders to exploitation by middlemen, further exacerbating their already weak bargaining position [32,33]. This vulnerability leaves farmers ill-equipped to cope with various shocks, such as extreme weather events, pest infestations, and disease outbreaks, often compelling them to resort to distress sales and undergo livelihood transitions. Additionally, poor market linkages exacerbate food insecurity as farmers struggle to access diverse foods or afford nutritious diets [34].

Furthermore, the COVID-19 pandemic has exacerbated these challenges by disrupting market chains, leading to a reduction in farmgate prices and revenues. Smallholders have been particularly impacted by the inability to sell perishable produce, resulting in substantial losses [35]. Furthermore, smallholders' ability to effectively adapt to climate change hinges not only on access to climate-related knowledge but also on financial resources derived from their access to markets [[36], [37], [38], [39]]. This underscores the critical role that robust market linkages play in buffering farmers against systemic shocks.

Efforts to enhance smallholder market access must focus on infrastructure development, aggregation through cooperatives, contract farming arrangements, digital platforms, and public procurement. Such initiatives can improve market efficiency, reduce income volatility, raise incomes, enhance resilience to shocks, and diminish livelihood vulnerability among smallholder farmers [[40], [41], [42]]. Targeted policies and interventions aimed at addressing market constraints are essential for safeguarding smallholder livelihoods and fostering sustainable agricultural development.

2.2. Marketing channels and decision-making

Marketing channels encompass the pathways through which agricultural producers, such as grain smallholder producers in Africa, bring their products to market and make them accessible to consumers [43]. In rural agricultural markets, smallholder farmers navigate various channels, including wholesalers, brokers, assemblers, and cooperatives, each playing distinct roles within the market ecosystem.

Wholesalers wield considerable influence in rural agricultural markets, primarily due to their financial prowess and control over market information. This control allows wholesalers to manipulate information to their advantage, potentially disadvantaging smallholder farmers [44].

Brokers and assemblers also play pivotal roles in providing market access to remote smallholders. However, they are susceptible to exploiting information asymmetries and the weak bargaining positions of farmers, often extracting unfairly high rents [45,46]. While these intermediaries offer essential marketing services, their actions can contribute to exploiting vulnerable smallholders.

In contrast, cooperatives offer a potential avenue for enhancing smallholder competitiveness. By pooling resources and leveraging collective action, cooperatives can reduce transaction costs and increase bargaining power [47]. Additionally, the aggregation of smallholder produce by cooperatives makes them attractive suppliers to buyers seeking traceability and reduced procurement costs [48].

Smallholder grain producers' selection of marketing channels is shaped by a multitude of factors that impact their ability to actively participate in the market [49] and ultimately determine their success and profitability. These determinants can be broadly categorized into three main groups—producer characteristics, market characteristics, and institutional factors [50].

Producer characteristics encompass a range of personal attributes such as age, gender, dependency ratio, and farm size. These factors influence producers' decision-making processes and their capacity to engage effectively in marketing activities.

Market characteristics pertain to the size and nature of the target market, including demand patterns, price volatility, and the level of competition. Understanding these market dynamics is crucial for smallholder farmers in determining the most suitable marketing channels for their produce.

Institutional factors encompass the presence and effectiveness of support institutions such as farmer associations, cooperatives, government policies, and access to finance and market information. These institutional arrangements play a critical role in shaping smallholder farmers' access to markets and their ability to engage in joint selling initiatives.

Given maize producers' challenges in accessing markets, cooperatives are viewed as a potential solution. However, encouraging joint selling within cooperatives requires a thorough assessment of the current conditions of the maize market and an understanding of the factors influencing smallholder producers' choice of marketing channels in Benin, particularly in the Kandi region. This understanding is essential for developing effective strategies to support smallholder farmers and promote inclusive and sustainable agricultural development within cooperatives. Policymakers and organizations can leverage this knowledge to tailor their interventions and address the specific challenges faced by smallholder farmers, thereby facilitating their integration into the market and fostering their economic resilience.

3. Research design and methodology

3.1. Targeted area of study

The study area, the district of Kandi (see Fig. 1), is situated in the central region of the department of Alibori, covering a vast expanse of 3421 square kilometers. Comprising 39 villages and 9 boroughs, Kandi exhibits diverse geographical characteristics. According to the most recent national census conducted in 2013, the total population of Kandi stands at 179 290 inhabitants [51]. The district's economy is heavily reliant on agriculture, with approximately 16 046 households engaged in agricultural activities, of which 15 980 are actively involved in crop cultivation [18].

Fig. 1.

Map of the study area.

Source: Computed with ArcGIS Pro, based on data from the National Forestry Inventory, 2008

Kandi holds significant prominence in maize production within Benin, being acknowledged as the country's second-largest maize-producing district [52]. Over the years, maize production in Kandi has witnessed substantial growth, with output increasing from 90 289 tons in 2016 to 119 401 tons in 2021, according to data from the Agency for Territorial Development and Statistics (ATDA) (refer to Appendix 2). This remarkable expansion underscores Kandi's pivotal role in maize production and its contribution to Benin's agricultural sector's development.

Furthermore, Kandi's strategic geographical location presents promising opportunities for agricultural products, particularly maize, owing to its proximity to neighboring countries such as Burkina Faso, Niger, and Nigeria. This proximity holds the potential to foster enhanced market access and trade prospects for agricultural produce originating from Kandi.

3.2. Overview of market conditions around kandi

According to the Ministry of Agriculture statistics, four types of agricultural markets were identified in Benin: primary markets, secondary or aggregation markets, and terminal or border markets (MAEP, 2022). Kandi falls under the category of secondary markets because of its location and the category of actors who operate there. Agricultural products, including maize grains, are collected and aggregated in Kandi before being transported to Malanville, a border market. From Malanville, these products are exported to Niger, Nigeria, and Burkina Faso. For instance, Maize is the main principal product Benin exported to Niger with an estimated export volume of 6082 T in 2010 [53].

Fig. 2 shows maize price variation—between 2021 and 2022—over the different markets. A consistent trend in maize prices is observed across all markets with the evolution of maize prices is delineated into 3 distinct periods. Firstly, the harvest season spans from November to December, during which maize price reach their lowest level due to the abundance of maize supply on the market. Secondly, the post-harvest season ranges from January to March, characterized by a slight increase in maize grain prices as the initial supply start diminishing. Thirdly, the lean season, from April to September, during which maize grains become less abundant on the market, resulting in a notable uptick in prices.

Fig. 2.

Maize price variations on agricultural markets.

Source: (MAEP, 2022)

Notably, maize prices tend to be more attractive in secondary and border markets due to their pivotal role as distribution hubs, serving as key points for the aggregation and transportation of agricultural products. Consequently price fluctuations in these market can significantly impact inter-border trade. Moreover, secondary and border markets offer opportunities for price arbitrage, wherein traders exploit price differential to maximize profits.

3.3. Type of data, sampling techniques, and sample size

Qualitative and quantitative data were collected from primary and secondary sources between May and July 2023 for this research. Qualitative data was obtained through interviews with representatives of the Communal Union of Maize Traders (UCCM) and Maize Producers Cooperatives (MPCs). Valuable insights were gained through the UCCM interview about the various stakeholders involved in maize marketing in the district, their respective roles, and market shares. Interviews with MPC representatives allow for capturing their role in maize marketing, mainly information related to joint selling ventures—including the identity of customers, delivery requirements, the quantity of maize bags sold, and payment schemes.

A structured questionnaire was used to collect primary data from maize producers. To identify the factors influencing marketing channels’ choice, the questionnaire covered socioeconomic, market, institutional, and motivational aspects. Additionally, we complemented our survey with secondary data obtained from various sources. These sources included published materials like academic papers, books, and official statistics from the Ministry of Agriculture, Breeding, and Fisheries. Moreover, unpublished data from entities such as the ATDA, the UCCM, and MPCs.

The present research adopts a multi-stage sampling approach to investigate the determinants of marketing channel choice by farmers. Firstly, Kandi was deliberately chosen as the study area due to its prominence as the second-largest maize-producing district in the country [52] and its crucial significance in the northern region. Based on recommendations from the regional office for agricultural development, the research narrowed its focus to villages with the most significant maize production and marketing records.

Secondly, eight villages within five boroughs were carefully selected based on their output in maize production and the existence of producers engaged in maize marketing activities. Together with officials of MPCs and UCCM, maize producers actively involved in maize marketing during the 2022/2023 cropping season were identified and listed, irrespective of their membership status in MPCs.

Based on these comprehensive lists from each village, the desired sample size was determined using the probability proportional to size (PPS) sampling method. Utilizing the Yamane [54] formula summarized in Equation ($1$), the research finalized a sample of 246 maize farmers, aiming for a 95% confidence level, 5% degree of variability, and a precision level of 6%. The study's multi-stage sampling strategy ensures a representative and robust dataset that will be analyzed to explore the critical factors influencing farmers' choices of marketing channels for maize.

$$\begin{array}{c}n=\frac{N}{1+N{\left(e\right)}^{2^{\prime }}}=\frac{15980}{1+15980{\left(0.006\right)}^{2^{\prime }}}\simeq 246\end{array}$$ (1)

n is the sample size, N=15 980 is the total size of maize producers, and e = 0.006 is the level of precision (6%).

3.4. Econometric model specification

Various factors influence farmers’ decisions on which marketing channels to choose. Two significant factors are the nature of the agricultural commodity and the local context. The perishability of the commodity and the available preservation technologies play an essential role in determining the marketing strategy. The local context, including the stakeholders, their numbers, and relationships with the producers, also influences whether farmers choose a single or multiple marketing channel.

In previous research, two methods—multinomial models and multivariate probit—were utilized to determine the factors that affect the selection of marketing channels. The multinomial model is applicable when farmers are limited to choosing only one market outlet from a set of mutually exclusive options. This model considers the independence of irrelevant alternative assumptions and the risks associated with selecting a particular channel [55,56]. On the other hand, the multivariate probit model is used when producers want to maximize their profit by selecting one or more marketing channels from the available options based on explanatory variables [[57], [58], [59], [60]]. Our survey found that maize grain producers use multiple channels for marketing and, therefore, adopted the multivariate probit model to analyze marketing channel choice determinants. Stata 14 package was used to run the multivariate probit analysis using variables summarized in Table 1.

Table 1.

Variable used in the Multivariate Probit Model (MVP).

Variables	Descriptions	Mean	Std Dev
Dependents variables
Brokers	1 if household choose brokers for maize marketing, 0 otherwise	0.63	0.48
Wholesalers	1 if household choose wholesalers for maize marketing, 0 otherwise	0.26	0.46
Cooperatives	1 if household choose cooperatives for maize marketing, 0 otherwise	0.23	0.42
Collectors	1 if household choose collectors for maize marketing, 0 otherwise	0.40	0.49
Independent variables
Gender	1 if the gender of the household head is male, 0 otherwise	0.95	0.21
Age	Age of household head in years	42.12	8.06
Dependents	Number of dependents of the head of household	10.45	6.02
Coop membership	1 if head of household is member of a maize producer cooperative	0.86	0.33
Distance market	Distance to the district market in kilometers	4.54	5.12
Use of tractor	1 if use tractor, 0 otherwise	0.48	0.50
Use of storage	1 if use storage for maize grains, 0 otherwise	0.97	0.16
Informal credit	1 if use informal credit, 0 otherwise	0.67	0.47
Maize land size	Household total size of land devoted to maize in kilometers	6.50	4.77
At harvest	1 if sell maize grains immediately at harvest, 0 otherwise	0.54	0.50
Post-harvest	1 if sell maize grains after harvest, 0 otherwise	0.72	0.44
Lean period	1 sell maize grains during the lean season, 0 otherwise	0.55	0.50
Avg price H	Average price of maize at harvest (XOF1)	10 782	143
Avg price PH	Post-harvest average price of maize (XOF)	13 842	127
Avg price LS	Lean season average price of maize (XOF)	20 868	568

Source: Authors' calculations.

The selection of maize marketing channel i by producer j is $Y_{ij}^A$ defined as the choice of producer_j to market maize grains through market channel i $Y_{ij}^A=1$ or not $Y_{ij}^A=0$ is expressed following Equation ($2$):

$$\begin{array}{c}{Y}_{ij}^A=\{\begin{array}{c}1\text{if}{Y}_{ij}^A=x_{ij}^A{\alpha }_{ij}+{\epsilon }^A\ge 0⟺x_{ij}^A\ge -{\epsilon }^A\\ 0\text{if}{Y}_{ij}^A=x_{ij}^A{\alpha }_{ij}+{\epsilon }^A<0⟺x_{ij}^A<-{\epsilon }^A\end{array}\end{array}$$ (2)

Where ${\alpha }_{ij}$ is a vector of estimators and ${\epsilon }^A$ is a vector of errors terms under the assumption of normal distribution; $Y_{ij}^A$ symbolizes the dependent variable for maize marketing channel choice of brokers, wholesalers, cooperatives, and collectors; $x_{ij}^A$ is the combined effect of the explanatory variables.

Univariate probit estimation of choice of each type of market outlet would be misleading for the expected problem of simultaneity. The selection of a kind of market outlet would be dependent on the selection of the other since smallholder farmers' choice decisions are interdependent, suggesting the need to estimate them simultaneously. To account for this problem, a multivariate probit simulation model was employed [10,57]. Since smallholder farmers’ market outlet choice decisions were expected to be affected by the same set of explanatory variables. Equation ($3$) below symbolize the multivariate probit equations.

$$\begin{array}{c}\{\begin{array}{c}\begin{array}{c}{Broker}_j={X_1^{\prime }\beta }_1+{\epsilon }^A\\ {Wholesaler}_j={X_2^{\prime }\beta }_2+{\epsilon }^B\\ {Cooperative}_j={X_3^{\prime }\beta }_3+{\epsilon }^C\end{array}\\ {Collector}_j={X_4^{\prime }\beta }_4+{\epsilon }^D\end{array}\end{array}$$ (3)

Where broker_j, wholesaler_j, cooperative_j, and collector_j are binary variables taking values 1 when farners_j selects brokers wholesalers, cooperatives, and collectors respectively, and 0 otherwise; X₁ to X₄ are vector of variables; ${\beta }_1$ to ${\beta }_4$ a vector of parameters to be estimated, and $\epsilon$ disturbance term. The symmetric covariance matrix Ω is given by Equation ($4$) as follows:

$$\begin{array}{c}\Omega =\left[\begin{array}{cccc}1& {\rho }_{12}& {\rho }_{13}& {\rho }_{14}\\ {\rho }_{21}& 1& {\rho }_{23}& {\rho }_{24}\\ {\rho }_{31}& {\rho }_{32}& 1& {\rho }_{34}\\ {\rho }_{41}& {\rho }_{42}& {\rho }_{43}& 1\end{array}\right]\end{array}$$ (4)

Where Ω represents the correlation between different types of market outlets. The non-zero off-diagonal allows for correlation across error terms of numerous latent equations, representing unobserved characteristics that affect the choice of other channels.

4. Description of results

4.1. Maize marketing channels and distribution volumes

Maize marketing in Kandi occurs through multiple channels that can be classified into marginal or main channels. Marginal channels are weekly market and direct sales to consumers, typically involving small quantities of maize, usually under 100 kg for a single transaction. Main channels handle large volumes, measured in tons of maize bags. In-depth interviews revealed 4 main channels—(1) brokers, (2) Collectors, (3) wholesalers, and (4) Maize producer cooperatives. Fig. 3 below shows the different marketing channels used by maize farmers.

Fig. 3.

Marketing channels and volume of maize distribution in Kandi

Note: Percentages along the chains refer to maize farmers' approximative share of maize sales through each marketing channel.

Source: Based on data from UCCM and MPCs, 2023.

Brokers play a crucial role in the maize marketing chain in Kandi, primarily dealing with small quantities of maize on behalf of producers who entrust them with a portion of their harvest. These brokers are predominantly local youths equipped with transportation means, such as motorbikes or tricycles, which allow them to transport between 2 and 6 bags of maize per trip. The process typically begins with brokers agreeing on a price with maize farmers before collecting maize bags for marketing or sale to retailers. Upon reaching the weekly market or retailers, brokers negotiate a higher price to earn a commission on the transaction. Dealing with brokers offers several benefits for maize farmers such as transport of maize bags, quick sales and availability over the year.

The main advantage cited by farmers who choose to engage with brokers for maize marketing is the ability to quickly sell small volumes of maize to address immediate financial needs. However, a notable disadvantage is that brokers' sale prices often exceed farmers' required prices, leading farmers to believe they could have earned more if they had marketed the maize themselves.

Collectors in the maize market of Kandi encompass two main types: large-scale farmers equipped with extensive storage facilities and local residents who assemble significant quantities of maize on behalf of wholesalers, often operating through informal contract farming arrangements with farmers. These collectors play a crucial role in the maize marketing ecosystem by supplying credit to maize farmers, typically in exchange for maize bags. The arrangement is often facilitated through informal contract farming agreements, wherein collectors provide farmers with credit, and in return, they receive maize bags as payment. The terms of these agreements, including the price of maize bags and repayment terms, are typically negotiated and settled upfront.

To address concerns about opportunistic behavior and ensure fair treatment of farmers, a minimum price of 10,000XOF was established among market actors starting from 2021. This minimum price serves as a safeguard against collectors taking advantage of farmers by offering unreasonably low prices for their maize. However, despite this regulation, there remains a risk of opportunistic behavior from some farmers who may fail to honor their contractual obligations or deliver maize bags in installments. This non-compliance often stems from the actual market price of maize exceeding the agreed minimum price of 10,000XOF, leading to disputes between collectors and farmers.

Wholesalers play a pivotal role in the maize marketing landscape of Kandi, acting as the primary players in the market. The analysis identifies two distinct categories of wholesalers: domestic-oriented wholesalers and cross-border-oriented wholesalers. Domestic-oriented wholesalers focus their operations on catering to the domestic market within Benin, while cross-border-oriented wholesalers specialize in exporting maize to neighboring countries. Both categories of wholesalers handle large volumes of maize bags, leveraging their substantial budgets to afford robust logistics infrastructure, including storage facilities and truck rental for transportation. These wholesalers primarily source their maize from collectors, who serve as intermediaries between smallholder farmers and larger market players. According to the president of the maize trader's union, UCCM, approximately 70% of the maize marketed in the district is sold to cross-border-oriented wholesalers, who then market it in neighboring countries such as Burkina Faso, Niger, and Nigeria. The strategic geographical location of the Department of Alibori, where Kandi is situated, facilitates easy access to these neighboring countries, which are major destinations for maize bags.

On average, there are 13 weekly shipments of maize bags sent to these neighboring countries, with each shipment containing approximately 650 bags of maize, each weighing 100 kg. The peak season for these shipments occurs from December until the end of June, corresponding to the period of heightened demand and market activity. The dominance of cross-border-oriented wholesalers underscores the significance of international trade in the maize market of Kandi, highlighting the region's role as a major exporter of maize to neighboring countries.

The Maize Producer Cooperatives (MPCs) in Kandi act in facilitating collective marketing efforts among smallholder farmers. However, despite their potential benefits, many MPCs face challenges in effectively engaging in joint selling activities. Out of the 27 MPCs in Kandi, only five were able to perform joint selling in 2022, highlighting the prevalent difficulties encountered by these cooperatives. In principle, MPCs are responsible for collecting maize bags from their members, conducting quality checks, removing impurities such as dust, straws, and stones, and recording the final weight before storage according to established protocols. These cooperatives have contractual agreements with organizations such as the World Food Program (WFP) and Catholic Relief Services (CRS), which purchase maize from them for various programs, including school feeding initiatives. However, these agreements often require farmers to deliver high-quality maize, which imposes additional requirements compared to other distribution channels.

One significant challenge faced by MPCs is the lengthy and bureaucratic process involved in receiving payments for maize deliveries. Despite fulfilling their contractual obligations, delays in paperwork processing have resulted in payment delays, with some MPCs waiting for more than three months after delivery to receive payment. This extended wait period, coupled with the need for additional storage time, poses significant challenges for farmers and reduces the attractiveness of selling maize through MPCs. Despite offering an attractive price of 30,000XOF for a bag of 100 kg of quality maize, the prolonged wait for payment and storage requirements deter many farmers from supplying maize to cooperatives.

4.2. Maize price intra-seasonal variations

Table 2 below shows the average quantity of maize sold along with the proportion of producers over the different periods of the crop season. Maize is sold over the year through 3 distinct periods—at harvest, post-harvest, and in the lean season. Around half of producers sell an average of 37.26 bags of 100 kg of maize at harvest at an average price of 10 782 XOF. The primary motivation mentioned by producers is the financial pressure for 62.85% of them and the contract farming for 35.71% of them.

Table 2.

Average quantities and prices of maize sold over seasons.

Time of sale	Sale parameters	Mean	Std	[95% Conf. Interval]		Proportion of farmers
Harvest	Quantity	37.26	3.7	29.93	44.59	56.91
	Price	10 782	143	10 498	11 065
Post- harvest	Quantity	95.69	16.23	63.64	127.73	76.01
	Price	13 842	127	13 591	14 093
Lean season	Quantity	100.98	10.3	80.60	121.37	57.72
	Price	20 868	568	19 744	21 992

Source: Own survey, 2023

76.01% of respondents declare that they sell maize grains during the post-harvest season. They sell an average of 95.69 bags of maize of 100 kg at an average price of 13 842 XOF. 61.5% and 36.36% of them, respectively, reported financial pressure and the increased maize price as the primary motivations.

Maize marketing in the lean season involved 57.72% of producers since many actors engaged in maize transactions more often store maize grains, expecting prices to increase. Our analysis revealed that producers sell more maize bags in the lean season and at a better price. Averagely 100.98 maize bags are sold during that period at an average price of 20 868 XOF.

4.3. Marketing channel selection patterns

Analysis revealed that maize producers use different market channels for maize marketing. Additionally, they use more than one marketing channel. The marketing channels are brokers, collectors, cooperatives, and wholesalers. Table 3 presents the predicted probabilities for each channel, displaying joint probabilities for failure and success as the results of the multivariate probit correlation.

Table 3.

Multivariate Probit correlation estimates across market outlets.

Variables	Marketing outlets
	Brokers	Wholesalers	Cooperatives	Collectors
Predicted probability	0.245	0.490	0.211	0.573
Joint probability (success)
0.930
Joint probability (failure)
0.006
Estimated correlation matrix
	ρ1	ρ2	ρ3	ρ4
ρ1	1.000
ρ2	−0.797***(0.064)	1.000
ρ3	−0.225*(0.105)	−0.059(0.142)	1.000
ρ4	−0.123(0.140)	−0.368**(0.129)	−0.510***(0.133)	1.000
Likelihood ratio test of: ρ21 = ρ31 = ρ41 = ρ32 = ρ42 = ρ43 = 0

χ2 (6) = 61.94
Prob > χ2 = 0.0000***	0.0000***
Number of draws (#)	6
Number of observations	246
Log likelihood	−347 077
Wald (χ2(40))	343.95
Prob > χ2	0.0000***

t statistics in parentheses.

*p < 0.05, **p < 0.01, ***p < 0.001.

Note: indexes refer to equation (1) = brokers, 2 = wholesalers, 3 = cooperatives, 4 = collectors.

Source: Authors' calculations.

The simulated maximum likelihood (SML) estimation result in Table 3 shows that the probabilities of maize producers choosing brokers, wholesalers, cooperatives, and collectors were, respectively, 24.52%, 49.8%, 21.18%, and 57.33%. Joint probabilities of success or failure of the four market channel choices also suggest that households’ likelihood of jointly choosing the 4 outlets simultaneously is 9.3%. In contrast, the joint probability of failure in doing so is nearly null.

4.4. Factors influencing the choice of marketing channels

Various factors influence the choice of marketing channels for maize grains among farmers. The MVP model results in Table 4 show that some variables were significant at multiple market channels. Age emerged as a determinant, with older farmers more likely to opt for wholesalers in maize marketing at 5% significance.

Table 4.

Results of multivariate Probit regression model of marketing channel.

Variables	Marketing outlets
	(1) (2) (3) (4)
	Brokers Coeff (Se)	Wholesalers Coeff (Se)	Cooperatives Coeff (Se)	Collectors Coeff (Se)
Age	−0.0021(0.11)	0.0310**(0.117)	−0.002(0.11)	−0.001(0.012)
Dependents	−0.0195(0.221)	−0.0122(0.011)	0.023(0.242)	−0.005(0.023)
Coop membership	−0.942**(0.324)	−0.190(0.372)	1.175**(0.341)	0.325(0.347)
Distance market	0.199***(0.040)	0.064(0.023)	−0.021(0.031)	−0.203**(0.064)
Use credit	−0.117(0.206)	0.188(0.234)	−0.367(0.215)	0.727**(0.249)
Maize land size	−0.0091(0.298)	0.024(0.030)	−0.038(0.034)	0.010 (0.036)
At harvest	1.597***(0.320)	1.109**(2.96)	−1.484***(0.341)	1.453***(0.419)
Post-harvest	2.072***(0.410)	1.123*(0.318)	−1.723***(0.499)	1.596**(0.549)
Lean period	3.580***(0.600)	0.407(0.502)	−1.778**(0.616)	0.282(0.736)
Average price	4e−4***(6e−5)	e−4 (6e−4)	28e−4***(72e−5)	−12e−4(97e−5)
_Cons	1.081(0.610)	−4.231***(0.633)	−0.959(0.665)	−1.985*(0.751)

*p < 0.05, **p < 0.01, and ***p < 0.001.

Coeff = Coefficient, Se=Robust standard errors in parentheses.

Source: Authors' calculations.

Additionally, cooperative membership significantly reduces the likelihood of selling through brokers, which is notable at 5% significance level. Instead, being a member boosts the probability of choosing cooperatives as a marketing channel (p < 0.10), indicating the impact of aligned collective interests.

Distance to market also shapes channel selection. Remoteness positively influences brokers usage (p < 0.05) since they facilitate transport from isolated areas. In contrast, it negatively affects selecting collectors (p < 0.05) who rely on farmgate purchases.

Credit dependence drives maize producers' choice of collectors (p < 0.05), reflecting their vital role as rural lenders. Timing of sales further influences channel preference. Sale during and just after harvest raises the likelihood of selecting wholesalers (p < 0.05; p < 0.10) when they purchase bags. Meanwhile, brokers are preferred throughout the year (p < 0.01), offering frequent income conversion.

Pricing also affects channel decisions. Higher prices boost the probability of choosing brokers and cooperatives (p < 0.01), known respectively to offer prices aligned on and above the market rate. Surprisingly, amongst the hypothesized variables, the number of dependents and the maize land size did not significantly influence the choice of any marketing channels.

4.5. Discussions

The research findings shed light on the main marketing channels for maize and the factors affecting farmers' choices, particularly in the context of smallholder farming in Kandi. Findings displayed the prevalence of multiple marketing channels used by maize producers. It is evident that farmers are influenced by specific factors and employ a diversified approach, engaging with brokers, collectors, cooperatives, and wholesalers simultaneously.

Access to credit and liquidity emerges as a key driver shaping channel preference. The high reliance on collectors for maize sales is linked to their provision of monetary advances that ensure a reliable maize supply. This shows the existence of contractual agreements between maize producers in Kandi and collectors. The same findings were obtained in diverse settings in the South [61,62]. As described by farmers, these informal contract farming arrangements guarantee access to capital but also risk locking them into low prices vulnerable to exploitation.

Wholesalers similarly dominate trade through capital-enabled aggregation. older maize producers' preference for wholesalers may be because they act as collectors or have similarities, such as extensive farmland and substantial capital, with collectors. This suggests that these similarities might foster trust and enduring relationships, as revealed by the qualitative analysis. Our finding is consistent with Dessie et al. (2018), who found that older grain producers in Ethiopia prefer to deal with wholesalers.

Transportation factors also impact channel decisions. Brokers are preferred throughout the year due to their flexibility in purchasing even small maize volumes and connecting producers to markets via competitive pricing. Degefa et al. (2022), and Nwafor [63] found the same results regarding grain marketing respectively in Ethiopia and Ghana. Smallholder farmers are more likely to choose a marketing channel closer to their farms, as it reduces transportation costs and time [64]. As one farmer said, "Brokers help me sell a few bags quickly when I need money." Their adaptability to frequent sales suits the needs of remotely located farmers. Conversely, cooperatives are chosen by their members partly due to reduced transport costs, given their proximity. However, they handle relatively minuscule volumes compared to the cross-border-oriented wholesalers exporting the majority of regional output.

Timing of sales and resulting price differences also affect choices. Wholesalers and collectors purchase early in the harvest when prices are lower, influenced by initial contractual agreements. As a farmer explained, "I already took a loan from my collector before planting, so I have to sell to him at the agreed price even if it's lower than the market rate." In contrast, cooperatives could provide higher prices by holding produce until lean periods when market prices are higher. However, this benefit is counteracted by delayed payments from cooperatives compared to prompt payments by other buyers upon delivery. The same constrains were observed in eastern Kenya regarding grain marketing [65].

Institutional buyers like the WFP also offer good prices but with protracted payment timelines that deter farmers. As one cooperative leader noted, "Three months after delivering our maize, we still haven't been paid, so farmers don't want to wait that long." Such constraints showcase cooperatives' challenges in providing competitive and timely payment terms compared to other channels.

While no single channel provides an optimal solution, enhancing cooperatives remains vital given the imbalanced power relations entrapping smallholders. However, addressing pricing alone without tackling underlying social dynamics is insufficient for creating truly empowering and equitable maize value chains.

4.6. Limitations

It is essential to mention that this research has two limitations regarding its scope and methodology. Firstly, the results cannot be applied nationwide as the sample size only represents maize farm households and market stakeholders in the district of Kandi. However, these findings can serve as a basis for further research to compare practices in Kandi with those in other districts of the country. Therefore, it is suggested that further research be conducted to elaborate on the strategies and techniques that drive maize farmers' access to the market. This could improve the market opportunities for maize farmers in Benin.

Secondly, this study may be limited by the restricted selection of variables when examining the factors that influence the choice of marketing channel. Given that various factors can affect the decision-making process, it is essential to broaden the scope of future research to include a broader range of variables. For instance, considering additional variables, such as maize farm households’ total revenue and access to market information, could provide valuable insight. Further, the absence of normality tests for error terms is a methodological limitation that future research should take into account to strengthen the analytical framework and ensure the robustness of findings.

5. Conclusions and implications

5.1. Conclusions

Using a mixed approach, this study contributes to increased knowledge regarding maize marketing in the district of Kandi, mainly—(1) in analyzing the current conditions of maize market. It sheds light on the current conditions, categories of market, and price variations over the year. (2) In identifying the key players on the market. The study reveals that Kandi maize producers predominantly engage with multiple market players, including brokers, collectors, cooperatives, and wholesalers. This understanding is crucial for mapping out the maize marketing landscape in the region. (3) In determining factors that influence the choice of maize producers in marketing channels. Age, distance to market, cooperative membership, informal credit, timing of sale, and the average price of maize bags influence the producers' choice. Further, informal credit was identified as a determinant factor since it leads to a high preference for collectors and wholesalers, the market's leading actors, based on their highest volume of transactions with farmers. Adversely, results indicated that although cooperatives offers the highest price, it have the lowest volume of transactions with farmers due to their strict requirements, such as the high standard for grain quality and the delayed payment. Lastly, further research on power relations and historical dynamics influencing choices would enrich understanding of the maize market.

5.2. Implications

Arising from this study's conclusions, key policy and implications at both the national and regional levels can be delineated to strengthen smallholder farmer livelihoods, foster inclusive agricultural transformation through cooperatives, and promote effective and sustainable maize marketing.

This study shows the need for holistic policies and targeted investments to empower smallholder farmers, strengthen farmer organizations, and build resilient local food systems. The research highlights that robust cooperative models can increase smallholders’ resilience by improving market access, reducing information asymmetries, and enhancing collective bargaining power. Therefore, it is imperative for the national government to enact policies and programs that provide financial, technical, and institutional support for the development of inclusive, farmer-owned cooperatives. Investing in rural infrastructure and agricultural extension services tailored to cooperative needs can further optimize their competitiveness and services to members. Regulations could be explored to increase transparency in value chain transactions and prevent unfair trading practices.

At the regional level, local authorities should invest in extension services providing farmers with real-time market price information to reduce information asymmetries. Facilitating contract farming arrangements between farmers’ groups and agribusiness could offer stable incomes, insulating farmers from price volatility. Further, cooperatives should improve services to members, specifically the provision of credit to maize producers. This strategic shift will enhance resource allocation, reduce financial pressure, and enhance marketing security amongst smallholders. Therefore, Cooperatives can guarantee patronage and marketing of maize, securing the market and enabling extended storage during the lean season. Finally, it is crucial to consider the geographical location when designing joint-selling strategies, particularly by providing transportation services or establishing collection centers for remote farmers. This will enable them to connect with more profitable market outlets such as cooperatives.

Ethical approval

Ethical approval was obtained from the University of Kagoshima.

6. Consents

Each of the participants gave consent to take part in this study.

Funding

This research was financially supported by the United Graduate School of Agricultural Sciences of Kagoshima University for data collection.

Data availability

The data sets for the current study are available from the lead author upon reasonable request.

CRediT authorship contribution statement

Godfrid Erasme Ibikoule: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Jaehyeon Lee: Validation, Supervision, Project administration, Conceptualization. Lise Audrey Godonou: Writing – original draft, Visualization, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Abbreviations

ATDA

Agence Territorial de Dévelopment Agricole

CSR

Catholic Relief Services

EMICoV

Enquête Modulaire Intégrée sur les Conditions de Vie des Ménages

FAO

Food and Agricultural Organization

INSAE

Institut National de la Statistique et de l’Analyse Economique

MAEP

Ministère de l’Agriculture, de l’Elevage et de la Pêche

XOF

West African CFA Franc

WFP

Word Food Program

Appendix.

Appendix 1. Maize producer's motivation for sale in relation to the season

Sale Motivations	Harvest (%) N = 140	Post- harvest (%) N = 187	Lean season (%) N = 142
Price	2.86	36.36	77.69
Oral contract	35.71	1.60	7.04
Financial needs	62.85	61.50	29.57
Personnal relashionship	1.43	1.07	3.52

Source: Authors' own calculation.

Appendix 2. Kandi's maize production areas and outputs between 2010 and 2020

	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020
Areas	41194	45961	29723	34100	41908	41535	61384	46670	N/A	41372	53294
Output	60847	81728	51719	45830	70576	69497	90289	77919	N/A	50639	81374

Source: Data from ATDA, 2023.