Determinants Influencing Cocoa Farmers’ Satisfaction with Input Credit in the Nawa Region of Côte d’Ivoire

Abstract

Assessing farmers’ satisfaction with Input Credit (IC) is essential for agricultural modernization in Sub-Saharan African countries. Therefore, based on farmers’ socio-economics data, this study aimed to determine the potential factors influencing farmers’ satisfaction with IC. The data were collected through a questionnaire from a random sample of 311 farmers in the Nawa region of southwestern Côte d’Ivoire in November 2022. Farmers rated constraints in the use of IC on three main indicators: (1) availability of inputs, (2) accessibility, and (3) credit repayment conditions. In addition to the descriptive statistics, a logistic regression model was constructed to compute the results using Stata 17.0 software. The main findings of descriptive statistics showed that 61.97% of the farmers were dissatisfied with using IC, while 38.03% were satisfied. The results also indicated that about 61.15% of farmers were constrained by credit services for inputs, compared to 38.85%. The logistic regression results revealed that the factors that significantly influence farmers’ satisfaction were annual production required and the number of years (at 1%) of IC use, training programs for farmers, farm size, input price (at 10%), and age (at 5%). Accordingly, cacao production in Côte d’Ivoire still faces multiple and complex factors. So, the results provide practical implications for policymakers and innovators to support smallholder farmers in providing high-quality technology innovation adoption programs.

1. Introduction

Recently, the issue of improving agricultural production and farmers’ livelihoods has been the focus of scientific research. However, the agricultural sector still faces several challenges, such as input and access to investment [1]. Cocoa (Theobroma cacao) is an economically important agricultural product for many people worldwide. Existing studies demonstrate that in 2018 about 40 million people depended on cocoa production, ensuring their livelihoods, with nearly 6 million farmers [2]. Around 80–90% of the world’s cocoa crop is produced by smallholder farmers [3]. Cacao is a tropical crop that grows in equatorial weather conditions. The ideal conditions for growing cacao include a warm, humid climate with temperatures ranging from 18 to 32 °C (64 to 90 °F) and annual rainfall between 1000 and 2500 mm. Also, cacao trees require well-drained soils with good fertility and a pH between 6 and 7.5 [4]. Some of the significant cacao-growing regions in the world include West Africa, South America, and Southeast Asia. Within these regions, cacao is typically grown in areas with suitable climatic and soil conditions and where there is a tradition of cacao cultivation. However, some regions’ climate change and soil degradation remain uncontrolled [5]. Thus, several solutions have been proposed to remedy this issue; however, they have failed in some regions.

The International Cocoa Organization (ICCO) predicted worldwide production of cocoa beans to have exceeded 4.0 million metric tons from 2015 to 2016. According to a French Agricultural Research Centre for International Development (CIRAD) report published in 2022 [6], the sown land area dedicated to cocoa cultivation has increased significantly over the last 60 years, rising from around 4.4 million hectares in the 1960s to almost 12 million hectares in 2021, for a production of over 5 million tons of cocoa. However, West Africa, notably Côte d’Ivoire, Ghana, Cameroon, and Nigeria, is set to remain the world’s largest cocoa producer, with output currently estimated at 74.8% of the world’s production in 2023. The International Cocoa Organization (ICCO) forecasts that production will increase by 3.7% for Africa, compared to 2.5% in America from 2022 to 2023. At the same time, a decrease of 0.37% is noted in Asia and Oceania (ICCO 2022). In a nutshell, Côte d’Ivoire, Ghana, Cameroon, and Nigeria were the top three cocoa-producing nations between 2021 to 2022, contributing 2.12, 0.68, 0.29, and 0.28 million metric tons to world production, respectively (ICCO 2022, https://www.icco.org/statistics/ accessed on 28 June 2023).

The agricultural industry continues to be crucial to Côte d’Ivoire’s economic development, creating jobs and reducing food insecurity. Notably, this sector employed approximately 42% of the working population, contributing 22% of the country’s GDP [7]. Along the same line, about 70% of growing cacao households are small farmers, and the Nawa region in the southwest represented the most critical Côte d’Ivoire cacao production. According to the International Cocoa Organization, in 2021, nearly 2.25 million tons of cocoa were exported from Cote d’Ivoire, compared to 1.05 million tons from Ghana. Consequently, it is worth noting that cacao production is significant in Soubre and Meagui jurisdiction, Nawa Region, with 30% of the total. In addition, Regional Directorate of the Coffee-Cocoa Council (RDCCC) data indicates that, in 2020, the Nawa region supplied more than 500,000 tons of cocoa, or 40% of the country’s total output. Also, nearly 90% of the region’s cultivated land is occupied by over 80 agricultural cooperatives working across 480,000 hectares. About 93% of the population depends on growing cocoa, occupying approximately 215,791 hectares [8]. Therefore, regarding the above background, several challenges still weaken the potential cocoa production in Cote d’Ivoire and need long-term solutions.

Meramveliotakis and Manioudis (2022) [9] pinpointed that in recent decades there has been an apparent steady trend in development studies for far more middle-range theories that gradually and systematically shift the attention to the contextual or situational aspects of development. However, Klarin (2018) [10] has pointed out that many countries have not achieved the sustainable development goal, and the gap between developed and underdeveloped countries has widened. The same authors consider how the fundamental constraints to implementing sustainable development are the degree of socioeconomic development that many countries have not yet achieved, combined with a lack of financial resources and technology, as well as the diversity of political and economic objectives on a global scale. In other words, adopting the Sustainable Development Goal for 2030 is an opportunity for the agricultural modernization of African countries, but the achievement is still tiny. That is why the fundamental goal of the third edition of the book “African Aspirations for 2063” was for the African continent to become prosperous through inclusive economic growth and development [11] based on the promotion of suitable agriculture.

Under this situation, cacao production is the backbone of economic sectors in many countries of the African Continent, Côte d’Ivoire in particular. However, smallholder farmers encounter many challenges in boosting productivity and implementing sustainable agriculture practices. They engage in traditional agriculture with heavy reliance on inputs [12,13]. In other words, cacao production is still conducted in the traditional method in Côte d’Ivoire. So, improving the production level requires solid input and knowledge of agriculture. But in recent years, among the different challenges, improving agricultural output has emerged as the most urgent problem facing the global development community [14], in developing countries in particular. In addition, the COVID-19 pandemic has posed enormous obstacles to the agri-food system, including a lack of inputs and technical support [15]. According to a report published by Ayanlade A. et al. (2020) [16], pandemic-related restrictions caused delays in the delivery of agricultural products and seeds and price increases for inputs such as fertilizers and pesticides. Supply chain disruptions have also resulted in a lack of farm equipment and spare parts, hampering agricultural production. Therefore, various policies and strategies have been implemented in most countries, including Cote d’Ivoire, to encourage farmers to adopt innovations and sustain agricultural growth [14,17].

What does existing research reveal about the factors influencing farmers’ satisfaction levels and the circumstances in which agricultural policies and services act to mitigate the non-adoption of technology? What are the best methodologies for conducting a holistic analysis to assess the relationship between farmers’ satisfaction levels and the factors influencing them? What challenges can survey data reveal to better understand the main factors determining farmers’ satisfaction levels after using input credit? However, given this context and these questions, it should be noted that an in-depth analysis of agricultural innovation technology programs is needed for Côte d’Ivoire. In other words, Côte d’Ivoire lacks relevant studies at the national level concerning the implementation and monitoring of agricultural technology innovation programs, and therefore, requires a holistic investigation. Accordingly, previous studies have often analyzed the issues of farmer satisfaction and adoption of agricultural innovation technologies separately. Government regulation is essential to smallholders’ behavior in many developing countries [18]. Or, Agricultural policies play a crucial role in agricultural economic growth [19], and the state’s role in agricultural development seems insignificant.

An integrated analysis of farmer satisfaction after using new agricultural technology with survey data is still limited. Thus, Côte d’Ivoire has not sufficiently understood and measured farmer reactions after implementing an agricultural innovation technology program. In other words, analysis of the monitoring and evaluation of agricultural technologies remains problematic. Therefore, from the social survey point of view, determining the fundamental factors influencing the level of satisfaction of farmers using input credit in the Nawa region may provide crucial information on the sustainable development of agriculture in Côte d’Ivoire. So, the “ADVANS-Côte d’Ivoire” (Advans-ci) financial institution’s “Input Credit” project, a pioneer in agricultural financing in Côte d’Ivoire, has been noted among these actions. IC is agricultural financing provided to farmers via cooperative groups. These cooperative organizations choose the farmers who will receive agricultural inputs based on a few criteria, such as farm size and the land guarantee. The cooperation mechanism facilitated input distribution, gathering the producers, and repaying the loan to Advans-ci. IC is “a package of inputs” (fertilizers, pesticides, equipment, sprayers, etc.) that make it easier for farmers to address issues, establish sustainable farms, and increase their output. Even though the success of adopting an innovative technology depends on farmers’ contentment with that agricultural technology, there has not been much published scientific research evaluating farmers’ satisfaction with agricultural innovation technologies in the Ivorian context.

However, the Nawa region significantly contributes to Côte d’Ivoire’s cacao production, and many issues hide farmers’ satisfaction with input to meet high productivity. In other words, the innovations described above as presented are limited, such as delayed distribution and limited access to inputs due to high prices. From then on, the overarching objective of this current study aims to determine farmer satisfaction levels with the agricultural innovation technology of the IC provided by Advans-ci. Precisely, to determine the socioeconomic factors impacting farmers’ satisfaction levels with IC in the Nawa region of Côte d’Ivoire, this study aims to: (1) describe the socioeconomic characteristics of farmers, (2) determine the constraints related to the use of the IC and the level of satisfaction of the beneficiaries, and (3) identify factors that affect farmers’ satisfaction with IC. This study’s main finding can fill information gaps and improve the cacao production system in Côte d’Ivoire.

2. Theoretical Framework

2.1. Quality of Input from Input Credit Technology

Two factors may influence the performance of inputs from IC innovation: first, the user’s subjective assessment of the effectiveness or ineffectiveness of the inputs, and second, the user’s perception and attitude toward their annual production and farm status. Successful agricultural innovation technology meets the requirements of farmers in different agroecological contexts. In addition, it also requires a solid management and planning system for farmers. This situation is possible through regular interaction between the annual production and the state of the farmer’s operation. Undertaking a systematic evaluation to address accountability and learning concerns is essential for effective interventions to ensure the effectiveness and efficiency of inputs from the credit-input innovation. Various criteria are used to assess the performance of an agricultural technology objectively.

The criteria most frequently used to evaluate agricultural innovation technologies are relevance, effectiveness, efficiency, and impact. Some authors have identified three key factors to consider when assessing the quality of a farming innovation technology: (1) feasibility criteria (costs, maturity, and potential); (2) impact criteria (economic, social, and environmental impacts); and (3) the sensitivity of the technology to the consequences of climate change, including temperature increases, droughts, and floods [20,21].

2.2. Satisfaction with the Input Credit’s Performance

Farmer satisfaction with agricultural innovation technologies is crucial in determining their widespread adoption. Perception of the quality and effectiveness of these technologies is a critical factor for farmer satisfaction [22]. Thus, the mental or emotional response to the efficacy of input credit can be used to measure farmer satisfaction. In agricultural technology promotion, farmer satisfaction is defined as the fulfillment of expectations regarding the quality and effectiveness of input credit technology. Measuring farmer satisfaction reflects the extent of the mental image created and the influence of input credit technology. This satisfaction has an impact on future interactions between farmers and technology developers. Satisfied farmers are more inclined to participate in the promotion of input credit technology and to share their experiences with other farmers, which can lead to wider adoption of the technology [23]. Therefore, it is essential to ensure that farmers clearly understand the benefits and limitations of input credit technology to meet their expectations. The quality and effectiveness of the technology must be guaranteed to ensure farmer satisfaction and encourage their participation in promoting the technology. By keeping farmers satisfied, technology developers can reinforce their relationships with farmers and improve their reputation within the farming community.

2.3. Factors Affecting Farmers’ Satisfaction with the Effectiveness of Input Credits

Focusing on influential aspects is critical to obtaining an overall picture of farmer satisfaction with IC performance. Previous research has examined variables affecting satisfaction in this environment, including socioeconomic characteristics, interaction with innovators and engagement in dissemination efforts, use of multiple communication channels, and perceived quality of information [12]. It has been reported that certain variables, including age, gender, education, annual income, and annual productivity, are likely to impact farmer satisfaction [24]. In this study, satisfaction is conceptualized as a farmer’s actual response to using inputs from the input credit. Specifically, we used the same concept as [24], defining satisfaction as fulfilling particular prior product or service expectations. Understanding and considering these factors when analyzing and interpreting farmers’ satisfaction with input credit thus became theoretically and empirically important. The conceptual framework of this study was developed based on the assumption that several factors, including personal and demographic, economic, institutional factors, and more, influence farmers’ satisfaction with input credit. The following variables were conceptualized for this study and affect farmers’ satisfaction with input credit: personal and demographic characteristics (age, gender, experience, education, and region), economic factors such as (annual production and farm size), institutional factors such as (timing of input distribution, input price, and training), other factors such as diversities of activities and method of credit repayment, and swollen shoot disease.

3. Presentation of the Study Area

This study was conducted in the Nawa region of southwestern Côte d’Ivoire. Nawa is located between 5°57′57″–7°7′58″ west longitude and 5°17′40″–6°44′29″ north latitude. It covers an area of 9193 km², with an estimated population of 1,165,472 inhabitants, with 127 inhabitants/km² (RGPH, 2021). The climate of Nawa is sub-equatorial and is characterized by a dry season (December–March) and two rainy seasons (April–June and September–November). Average temperatures range between 26 and 28 °C and can reach 30 °C during the dry season. The average rainfall is between 1300 and 1600 mm/year for 115 days of rain. The relief of Nawa is made up of vast plateaus surmounted in places by some elevations, such as Mount Trokoi (OKROUYO) and Mount Zatro (GRAND ZATTRY) [25]. Developed on ancient eruptive rocks, the soils of Nawa are predominantly ferrallitic, relatively weathered, and vary in texture from clayey silt to silty sand. Deep and porous, these soils are generally well suited to cash crops such as cocoa [26]. The Nawa region comprises four departments: SOUBRE, MEAGUI, GUEYO, and BUYO (Figure 1). Agriculture is the primary source of income, accounting for 75% of economic activities and contributing 94% of the population’s income [27]. Cocoa farming covers 215,791 ha in this locality, and 93% of the population lives from this crop (Conseil Café-cacao 2020).

4. Material and Methods

4.1. Sampling Strategy and Data Collection

The data used in this study were obtained from farm households through a survey conducted in the communes Soubre and Meagui, Nawa region, during November and December 2022. Concerning the sampling strategy, the study used a multilevel random sampling method to gather primary data. In the first stage, the Nawa region was purposely selected because it represents the first region in which input credit was introduced and one of the largest cocoa-producing regions in Côte d’Ivoire. Then, the two communes were randomly selected from among the region’s communes according to their importance in terms of farmers and cooperatives. The random sampling technique used the input credit to select 311 farmers’ households. The data accuracy process showed that the responses of six farmers were discarded due to their lack of logic. As a result, 305 questionnaires were used for the final analysis. The farmer respondents were identified from a list indicated by cooperative delegates. In addition, interviews and group discussions were used to supplement the data obtained from the field survey. Most (99%) of those interviewed were users of the input credit.

The use of questionnaires in research allows researchers to collect massive amounts of data in a short time. It can be beneficial for addressing many questions in a standardized manner [28]. In addition, it can effectively collect information from farmers’ perceptions of a specific problem. The questionnaire is composed of three parts. The first part focused on the socioeconomic characteristics of the interviewees. The second part focused on the difficulties encountered by farmers in acquiring inputs, as measured by (delays in distribution and the high price of inputs). The third part focused on the level of satisfaction of farmers with the quality of inputs from the input credit technology. Farmers were asked to indicate their satisfaction on a Likert-type scale of 0 = dissatisfied and 1 = satisfied according to five quality attributes of the input credit technology: (1) availability, (2) accessibility, (3) diversity, (4) relevance, and (5) effectiveness. CommCare HQ 2.53.0 software was used as a tool for data collection.

4.2. Determining Relevant Fundamental Factors

Annual farm production: Gross farm income represents the product’s value from selling farm products. An increase in annual agricultural production is expected to result in a positive perception of the price, while low production should produce the opposite effect. Thus, yearly production is expected to correlate with satisfaction positively [29]. A study on farmers’ satisfaction with an agricultural input voucher (AIV) system in China reported that income variable has a significant negative relationship with satisfaction [30]. These two factors were essential in risk management. The rational farmer seeks to maximize his production and hence his farm income. The expected theoretical sign is positive.

Age: It has been hypothesized that as a farmer’s age rises, new innovation-related tasks no longer satisfy them as much. This was because these farmers might not be as risk-averse as younger farmers and were less focused on expansion. On the other hand, older farmers have accumulated knowledge and experience over the years and are better qualified to assess and profit from the performance of practices than younger farmers [31]. Consequently, there was uncertainty over how age affects farmers’ satisfaction.

Gender: Previous research into the connection between gender and satisfaction with the effectiveness of agricultural technologies has produced mixed findings. Women were generally less likely to acquire agricultural technologies than men [32]. Women were less dedicated than men because fieldwork was challenging. As a result, it is possible to speculate that male farmers are more content with new, inventive activities than female farmers. As a result, farmers’ satisfaction was positively influenced by the gender variable.

Education: Education improves a person’s resources and capacity to carry out tasks, but it also broadens their understanding of their options and the benefits anticipated from their actions. Farmer education has a favorable impact on satisfaction [33]. The rationale for this claim was the ability of educated people to become aware of and adopt better farming practices. Consistent with this logic, we suggest that their level of education positively influences farmers’ satisfaction with IC.

Farm size: Since the effects of inputs on small and large farms vary, land ownership could impact farmers’ attitudes toward agricultural innovation approaches. We hypothesize that farmers with large landholdings were less satisfied with the effectiveness of IC than those with small landholdings. Farmers with large areas may not have enough time and financial resources to purchase and apply inputs more intensively than farmers with small areas because inputs are expensive and require enough time for application. It has been shown that the likelihood of adopting microdosing technology is negatively related to the size of the cultivated land [34]. We, therefore, suggest that farm size harms farmer satisfaction.

Agricultural Training Programs: The training variable means that the farmer has received training on the conditions for using the technology. In theory, training lets farmers know about the technology and its use. A well-trained farmer undergoing an innovation activity had a greater propensity to adopt the technology than a farmer without training. Thus, we have the hypothesis that training positively influences farmer satisfaction.

Farmer experience: Farmers’ experience represents the time the producer has used the IC. Long-term users will probably profit more from the technology since they are more likely to be familiar with the technology usage guidelines and climatic needs. As a result of this study, it is anticipated that the length of use will be directly tied to how satisfied the farmers are with the effectiveness of the technology.

Activity variety: It was predicted that farmers engaged in other activities (business, trade, livestock) would be less focused on crop production and less satisfied with the IC strategy. Off-farm activities were also reported to positively impact farmers’ satisfaction with agricultural technologies. This situation was because income from off-farm activities is expected to provide farmers with an alternative source of liquid capital to purchase productivity-enhancing inputs [35]. Therefore, there was uncertainty about how off-farm activities affect farmers’ satisfaction.

Input distribution times: Fertilizer application and farm treatment were made according to a specific schedule. However, the more the inputs were distributed according to the calendar, the more the farmers respected the treatment and application periods. A well-treated farm that follows the recommended agricultural calendar gives discounted results, increasing the farmer’s satisfaction and leading to higher agricultural production. Therefore, we hypothesized that farmers who received the inputs according to the agricultural calendar could produce more than those who received the inputs outside the agricultural calendar. Thus, the time of distribution negatively influences farmers’ satisfaction.

WATPRICE: The cost of using a new technology is one of the critical factors in determining its acceptance. Adopting new technologies in agriculture is hindered by their high cost [36]. The high cost of technologies has been identified as a barrier to adoption in previous studies on factors influencing technology adoption. Therefore, farmers were less satisfied when the price of acquiring the technology was high. Thus, we hypothesized that the cost of IC could have a negative impact on farmer satisfaction.

Regions: The departments of Meagui and Soubre were characterized as two major cocoa-producing areas (measured as a dummy variable). Farmers in the Meagui area were more exposed to swollen shoot disease (SSD) than those in the Soubre area. As a result, the IC performance was reduced in the Meagui area, resulting in sudden orchard mortality and lower production. We expected farmers in the Soubre area to be more satisfied with the performance of the IC than farmers in the Meagui area.

The following explanatory factors were selected and used to develop a theoretical association between the level of satisfaction and household characteristics: annual production, year of use, age, gender, level of education, training, farm size, input prices, delivery time, activity diversity, and regions (

Table 1. Description of the explanatory variables and their expected signs supposed to affect the level of satisfaction of farmers with the performance of the input credit.

Variables	Explanation	Expected Sign
LEVEL OF SATISFACTION (dependent variable)	Input credit usage is rated as 0 if the farmer is not satisfied, 1 if the farmer is satisfied
PRODUC (kg)	the average net production of the farmer with the use of the input credit per year	+
AGE	Age of the farmer (1 = 18–30 years, 2 = 31–40 years, 3 = 41–50 years, 4 = 51–60 years, 5 = 60 years more)	±
GENDER	The sex of the household head (0 if female, 1 if male)	+
EDUC	Level of education (0 = no school, 1 = primary, 2 = secondary, and 3 = university)	+
FARMSIZE (ha)	the size of the farmer’s farm (1 = 1–5, 2 = 6–12 3 = 13–18, 4 = 18 ha and more)	±
TRAINING	1 if the farmer has received input credit training, 0 otherwise	+
FARMEXP	Number of years the farmer has been input credit (in years)	±
DIVERACT	Diversity activities: 1 if the farmer practices other activities and 0 if not	±
WATDIST	The reaction of the beneficiary in the distribution of the input credit: 0 if there is a delay in the distribution, 1 if there is not.	-
WATPRICE	The reaction of the farmer to the price of the input credit: 0 if a low price, 1 if a high price	-
REGION	the location of the farmer’s farm: 1 if Meagui, 2 if Soubre	+

). The variables were selected primarily because they cut across social and economic domains, allowing a comprehensive understanding of farmers’ satisfaction with the IC. Each explanatory variable and the expected theoretical relationship with satisfaction are briefly described here.

4.3. Construction of the Logistic Regression Method

Various statistical analysis tools were available to separate the variables influencing farmers’ satisfaction with the effectiveness of agricultural innovation techniques. It is possible to employ probabilistic models like the linear probability model (LPM), logistics model (LM), and probit model (PM). However, the ideal method to include farmers’ opinions in performance evaluation studies needs to be carefully chosen. For the farmer satisfaction survey, it is recommended to use the LM [37], as it offers the following advantages over the other models:

√

The logistic distribution is calculated to ensure that the rate of estimated probability is always between 0 and 1.

√

The heteroscedasticity issue is consequently resolved because the probability does not rise linearly with a unit change in the value of the explanatory variables, as is the case in the LPM.

√

Compared to the PM, it is simpler to calculate and explain. In other investigations, the dichotomous LM has been applied to examine farmer satisfaction [37].

A logit regression analysis predicts a binary or multinomial dependent outcome from a set of independent variables, which calculates the likelihood of an event happening or not. For instance, a farmer’s satisfaction state regarding IC use can have values of 1, 2, or 3, which denote dissatisfaction, neutrality, and satisfaction, with the state depending on numerous independent elements relating to the farmer. The LM can therefore be used to calculate the level of satisfaction of a randomly chosen farmer with an agricultural innovation technology [27,37]. The LM comprises three categories for assessing the farmer’s satisfaction level odds. Suppose Pr (Y = 3), the probability that a farmer is satisfied using IC, is the reference group. In that case, Equations (1) and (2) are Logit functions for the categories “unsatisfied farmer” and “neutral farmer”, respectively.

$$Z_1\left(X\right)=ln\left(\frac{\mathrm{Pr}\left(Y=1|x\right)}{\mathrm{Pr}\left(Y=3|x\right)}\right)=ln\left(\frac{P_1}{P_3}\right)={\beta }_{10}+{\beta }_{11}{X}_1+{\beta }_{12}{X}_2+\cdots +{\beta }_{1k}{X}_k$$

(1)

and

$$Z_2\left(X\right)=ln\left(\frac{\mathrm{Pr}\left(Y=2|x\right)}{\mathrm{Pr}\left(Y=3|x\right)}\right)=ln\left(\frac{P_2}{P_3}\right)={\beta }_{20}+{\beta }_{21}{X}_1+{\beta }_{22}{X}_2+\cdots +{\beta }_{2k}{X}_k$$

(2)

where:

$X_1,X_2\dots ,X_k$ denote the set of explanatory factors assumed to affect $Y$.

$Y$ is the dependent variable (“satisfied”, “neutral”, or “not satisfied”) ${\beta }_{10}$ et ${\beta }_{20}$ represent the intercepts ${\beta }_{11},\dots ,{\beta }_{1k}$ et ${\beta }_{21},\dots ,{\beta }_{2k}$ represent the slopes of the Logit regression functions $Z_1\left(X\right)$ et $Z_2\left(X\right)$, respectively [31,37,38].

Following the dichotomous LM [22,27,38] where a farmer is considered satisfied ($Y$ = 1) or not ($Y$ = 0), and considering the 3-category multinomial Logit function as 2 dichotomous logistic functions, it can be shown that the probabilities of the 3 categories can be shown as:

$$p_3=\mathrm{Pr}\left(Y=3|x\right)=\frac{1}{1+e^{z_1}+e^{z_2}}$$

(3)

$$p_2=\mathrm{Pr}\left(Y=2|x\right)=\frac{e^{z_2}}{1+e^{z_1}+e^{z_2}}$$

(4)

$$p_1=\mathrm{Pr}\left(Y=1|x\right)=\frac{e^{z_1}}{1+e^{z_1}+e^{z_2}}$$

(5)

with

$$p_3+p_2+p_1=1$$

(6)

The maximum likelihood method estimates the model’s parameters [39]. The Logit model was employed in this study to identify the variables that significantly impact farmers’ satisfaction with IC.

A structured questionnaire was given to 305 use-credit-input farmers in the Soubre and Meagui regions who had been chosen by stratified random selection during the harvest season in November 2022. To administer the questionnaire, five local enumerators who spoke the native language at the study location underwent training. Annual production, year of use, age, gender, training, farm size, input prices, delivery time, activity diversity, and regions were among the data gathered. This case’s Logit model was described as follows:

$$Y=f\left(x_1,x_2,x_3,\dots \dots \dots ,x_k\right)$$

(7)

where:

$Y$ is the dependent variable, and in this example, satisfaction with the IC is the dependent variable. It is anticipated that independent variables $x_1,x_2,x_3,\dots \dots \dots ,x_k$ will have an impact on $Y$.

Using Stata 17 software, the characteristics examined through the questionnaire were evaluated to identify those significantly affecting farmers’ satisfaction.

5. Results and Discussion

5.1. Socioeconomic Profile of Farmers

The analysis of the socioeconomic profile of farmers is presented in

Table 2. Distribution of farmers according to socioeconomic characteristics (N = 305).

Variable	Items	Frequency/ Moy	Percentage (%)
Production Net	-	2229.68	-
Years of using	-	5.20	-
Farm size	1–5 ha	202	66.2
	6–12 ha	74	24.2
	13–18 ha	13	4.2
	>18 ha	16	5.2
Agricultural training	No	47	15.4
	Yes	258	84.5
Age	18–30 years	42	13.7
	31–40 years	89	29.1
	41–50 years	57	18.6
	51–60 years	65	21.3
	>60 years	52	17.0
Gender	Male	283	92.7
	Female	22	7.2
Education	Not school	186	60.9
	Primary school	65	21.3
	Secondary school	51	16.7
	University	3	0.9
Distribution constraint	No delay in the distribution	85	27.8
	delay in the distribution	220	72.1
Input price constraint	Low price	89	29.1
	High price	216	70.8
Diversity Activity	Not practice diversity activity	225	73.7
	Practice diversity activity	80	26.2
Regions	Meagui	72	23.6
	Soubre	233	76.3

. The data showed that 29.1% of farmers are 31–40 years, 21.3% are 51–60 years, 18.6% are 41–50 years, 17.0% are more than 60 years, and 13.7% are 18–300 years. This means that about half (61.6%) of the farmers were in the 18–50 age group and were still agile and active. The age distribution among cocoa farmers in this study confirmed the contention that farmers in Côte d’Ivoire and the same results were observed in Ghana with the 25–55 years group [40]. Gender indicated that most (92.7%) of the farmers were male, compared to 7.2% of females. In other words, many cocoa farmers are male; therefore, the diffusion of credit-input agricultural technology would be unevenly distributed among male and female cocoa farmers. This result could confirm the mixed factor in women’s participation in cocoa farming as indicated by some sector organizations, such as the World Cocoa Foundation and international NGOs, and as Oxfam’s Behind the Brands Campaign [36,40]. In addition, 71.5% of participants were married, compared to 28.4% living in a union. This meant that married people were more active in cocoa farming, which could be due to their increased responsibilities as married people responsible for meeting the needs of their families.

The analysis showed that 60.9% had no schooling, 21.3% had primary education, 16.7% attended secondary school, and 0.9% had a university education. This result confirms that nearly 60.0% of the farmers had no schooling [41]. The remaining 40.0% of the total farmers, partially most, had only attended elementary school (partially or fully), while only 10.0% of all cocoa farmers surveyed had gone beyond elementary school. This suggested that most cocoa farmers in the study area were illiterate. This could serve as a disincentive in adopting IC technologies, thus influencing farmer satisfaction. Education is an essential factor influencing the adoption of agricultural innovations [42]. The majority (66.2%) of the farmers had a farm size of 1.0 to 5.0 ha, while 24.2% and 9.5% had a farm size of 6.0 to 15.0 ha and more than 16.0 ha, respectively. The size of a farm is a crucial determinant of expected yield. Training is a critical element of the diffusion of technology. Most of the 84.5% of the producers interviewed have received training on the conditions for using the products in the input package, and (15.4%) have not received training. The results of the descriptive analysis on the diversity of activities showed that out of all the cocoa farmers interviewed, only 73.7% cultivate cocoa, while 26.2% practice other activities. This was because cocoa farming is complicated; cocoa farmers rarely engage in other activities.

5.2. Distribution of Farmers Based on Level of Satisfaction Regarding Input Credit

Table 3. Distribution of respondents according to their satisfaction level with the input credit.

Level of Adoption	Whole Sample N = 305	Regions
		Meagui (N = 72)	Soubre (N = 233)
Low	62.0	17.4	44.6
High	38.0	6.2	31.8

shows farmers’ satisfaction level with the IC in the study area. The results showed that the respondents could be classified into two categories according to their level of satisfaction (high satisfaction level and low satisfaction level). Thus, in the Meagui region, 17.4% of the cocoa producers surveyed had a low level of satisfaction, compared to 6.2% who had a high level of satisfaction. In contrast, in the Soubre region, 31.8% of the producers surveyed had a high level of satisfaction with the IC, compared to 44.6%, who had a low level of satisfaction. Consequently, the level of satisfaction was low in the study area, with a rate of 62.0%, against 38.0% of producers with a high level of satisfaction. The plausible reason for these results could be that most of the low-level adopters recorded felt that the cost of inputs would be increased, the terms of repayment of credit would be inappropriate, and their despair related to SSD. The results of this study are consistent with the observation that several factors, including household socioeconomic characteristics, production operation and management, market practices, processing characteristics, degree of awareness by producer organizations and social networks, influence the level of satisfaction with new technologies in agriculture in Sub-Saharan Africa [24].

5.3. Satisfaction of the Farmers with Input Credit Services

Fertilizer distribution, as in the agricultural sector, is subject to many constraints influencing user satisfaction. The result observed in

Table 4. Distribution of respondents according to constraints encountered in using the innovative agricultural technology of input credit.

Constraints Influencing Adoption	Frequency/Percentage
	Unsatisfied	Satisfied
Delay in the distribution of input	220 (72.1%)	85 (27.8%)
Disease (swollen shoot)	107 (35.1%)	198 (64.9%)
The high price of the input package	217 (71.1%)	88 (28.8%)
Constraints related to the repayment condition	202 (66.2%)	103 (33.7%)
Overall Satisfaction	186 (61.1%)	119 (38.8%)

showed the constraints affecting farmers’ satisfaction levels with the input credit technology. Delay in distribution (72.1%), the high price of input package (71.1%), and repayment conditions of credit (66.2%) were the main constraints of the farmers. In comparison, 35.10% of the farmers identified the disastrous effect of SSD on the farms as another type of constraint. On the restrictions related to farmers’ satisfaction with the credit input, Table 4 indicated that 61.1% of the respondents were dissatisfied compared to 38.8%. This finding corroborated the result by (Diran Olawale 2015) [43], according to which the delay in fertilizer distribution was an essential factor in adopting some new agricultural technologies. It has been confirmed in some studies that inefficient information and communication management limits the spread of innovative agricultural technologies [44].

The lack of product pricing, flexibility in accessing IC and the lack of organization, and visibility between innovators and their partners in distribution and repayment have limited the spread. However, the results of this study observed that the use of IC is still unknown and that the low level of satisfaction of farmers with agricultural innovation technologies resulted from the lack of communication and information on its various factors.

5.4. Determinants of Satisfaction Level

The binary logistic regression model was used to investigate the prediction potential of the specified demographic and socioeconomic characteristics on farmer satisfaction. In this study, 11 independent variables (annual production, year of use, age, gender, level of education, training, farm size, input prices, delivery time, activity diversity, and regions) were included in the model. To evaluate for multicollinearity, we conducted preliminary hypothesis tests. The tolerance values for each explanatory variable were more than or equal to the customary tolerance level (not less than 0.10), demonstrating that the multicollinearity presumption was not violated. This was also confirmed by the variance inflation factor (VIF) values for each independent variable, which were well below the threshold of 10 (not more than 10).

Table 5. Determinants of farmers’ satisfaction with input credit technology (Logit regression).

Level of Satisfaction	Coeff.	Std. Err.	z	O.R	p > |z|	[95% Conf. Interval C.I]
Production (kg)	0.000 ***	0.000	4.94	1.001	0.000	0.000	0.001
Number of years of use of the input credit	−0.186 ***	0.048	−3.85	0.830	0.000	−0.282	−0.091
Farm Size (ha)
6–12 ha	−0.455	0.362	−1.26	0.634	0.209	−1.165	0.255
13–18 ha	−1.457 *	0.867	−1.68	0.233	0.093	−3.158	0.242
>18 ha	0.073	0.932	0.08	1.08	0.937	−1.755	1.901
Training
Yes	0.696 *	0.393	1.77	2.007	0.077	−0.074	1.468
Age
31–40 ans	−0.726	0.457	−1.59	0.484	0.113	−1.623	0.170
41–50 ans	−0.341	0.514	−0.66	0.711	0.507	−1.350	0.666
51–60 ans	−1.027 **	0.501	−2.05	0.358	0.040	−2.010	−0.045
>60 ans	−1.120 **	0.538	−2.08	0.326	0.038	−2.176	−0.063
Gender
female	−0.684	0.522	−1.31	0.504	0.191	−1.709	0.340
Education
Primary school	−0.058	0.356	−0.16	0.943	0.870	−0.756	0.640
Secondary school	0.143	0.392	0.37	1.154	0.715	−0.625	0.912
University	−1.781	1.509	−1.18	0.168	0.238	−4.739	1.177
Distribution constraint
delay in Distribution	0.255	0.331	0.77	1.291	0.442	−0.395	0.905
Input price constraint
high price	−0.612 *	0.371	−1.65	0.542	0.100	−1.340	0.116
Diversity of activities
Practice other activity	−0.202	0.329	−0.61	0.817	0.539	−0.847	0.443
Regions
Soubre	−0.404	0.351	−1.15	0.667	0.250	−1.093	0.283
Constant	1.307	0.694	1.88	3.697	0.060	−0.052	2.667

Coef: estimation coefficient; Std. err: Standard error; z: t-test; OR: Odds ratio; *: p < 0.1; **: p < 0.05; ***: p < 0.01; C.I.: Confidence Interval.

shows the logistic regression model results. The likelihood ratio results in the logistic regression test showed that the regression model containing all predictor variables was statistically significant, with a Chi-square statistic of 2.28. This result indicated that the model’s explanatory variables were significantly related to satisfaction level. The explanatory variables were well chosen and can be used to predict the dependent variable. The model explained the variance in satisfaction level between 25.2% (Cox and Snell R-squared) and 34.3% (Nagelkerke R-squared), and correctly classified 88.7% of the cases. According to the results presented in Table 5, farmers’ annual production (p < 0.001), number of years of use (p < 0.001), farm size (p < 0.1), training (p < 0.1), age (p < 0.05), and input price (p < 0.1) all demonstrated predicted results when predicting farmers’ satisfaction level. This suggests that these six independent variables have been statistically significant. The other explanatory variables (gender, education level, distribution time, diversity of activities, and regions) did not show significant results when explaining the level of satisfaction.

Annual agricultural production had a positive and significant relationship with farmers’ satisfaction levels. It increased the probability that a farmer was highly satisfied with IC. A unit increase in agricultural production increased the probability that a farmer had a high level of satisfaction with IC by 1, according to OR. This result is expected because an increase in agricultural output motivated most farmers and could be interpreted by many farmers as meaning that the agricultural activity was doing well. However, farmers with low production could attribute their low income to inefficient IC, leading to lower satisfaction with IC. Since most cocoa-producing households derived a large part of their livelihood from the crop, farmers satisfied with IC will likely be those with high production. This situation indicates that satisfaction with IC was related to the farmer’s economic status. Production levels were essential to most smallholder farmers and influenced their perceptions and satisfaction with development programs and policies. The result of this research agreed with Anang B. (2016) [29], who found that gross income positively influenced farmers’ satisfaction with cocoa pricing in Ghana.

The number of years the farmer has used IC technology decreased the likelihood that they would be more satisfied. This result is more or less expected because when farmers have a long year of experience, they tend to lose interest in this technology and move towards new technologies. The findings indicated that a one-unit increase in the number of years of IC use decreased the probability that a farmer had a high level of satisfaction with IC by 0.83. Our findings were in conformity with Bradley D. E. et al. (2004) [45]. However, some authors such as Achigan-Dako G. et al. (2014) [44] have shown that the long year of use of improved maize seed had a positive and significant effect on farmers’ satisfaction level in Benin.

The farm size negatively and significantly affected farmers’ satisfaction with IC. The results showed that increasing the size of a farm by one hectare made it less likely that a farmer would be very satisfied with IC by 0.23. This result is more or less expected, as it might have been explained by the fact that farmers with large farms did not have the time and quantities of inputs needed to process their farms compared to farmers with small farms. As a result, they might be dissatisfied when expected results are not achieved as often as they would like. Our results were in line with some, including Djokoto et al. (2016) [46] and Onyeneke (2017) [47], who showed that increasing farm size by one unit reduced the likelihood of adopting the appropriate planting depth for rice production by 0.017. However, Elias et al. (2016) [31] and Rao et al. (2022) [48] reported that farmers with large farms were more satisfied with extension services in the Organization of Eastern Caribbean States (OECS) than farmers with small farms.

Training on using IC was positively and significantly related to farmers’ level of satisfaction with IC. This variable determined whether receiving training on IC technology made farmers more likely to have a high level of satisfaction. They were more likely to be satisfied with IC because they knew the technical possibilities of applying inputs in a smallholder context. The result demonstrated that farmers who actively participate in training programs initiated by their cooperative have an increased probability of having a high level of satisfaction with an IC of 2.01. This finding was more predictable because trained farmers were more likely to be mindful of IC’s possible advantages and better equipped to evaluate the technology’s quality gap than untrained farmers. Our results agreed with some authors, such as Gomo et al. (2014) [37], who investigated the functioning of the Mooi River Irrigation Project in South Africa and found that farmer training was statistically and significantly related to satisfaction.

Age showed a significant and negative relationship with farmers’ satisfaction with IC, implying that satisfaction with IC decreased with increasing age. This result showed that farmers aged over 51 years were less satisfied with IC than younger farmers. A unit increase in farmers’ age decreased the probability of a farmer having a high level of satisfaction with IC by 0.32. This result was consistent with our expectation that, due to their age, older farmers might have higher expectations than younger farmers, which might influence their level of satisfaction with IC. Our results were consistent with some studies, including Achigan-Dako G. et al. (2022), Onyeneke et al. (2017), and Bradley et al. (2004) [44,45,47]. However, some research showed a positive relationship between age and the satisfaction of cocoa farmers in a “new cocoa pricing system” in Ghana [29].

The logistic regression results demonstrated that the price of acquiring IC significantly and negatively influenced farmers’ degree of satisfaction with the IC. The observed negative sign estimate showed that farmers with access to IC were less likely to be satisfied than farmers who did not have access to IC. Farmers who used IC were 0.54% less likely to be satisfied with it than those who did not. The likely reason for this result could be that the high price of acquiring credit did not help farmers alleviate capital constraints, and thus, prevented them from purchasing the inputs they wanted promptly with their limited resources. The result of this study disagreed with Damisa et al. (2010) [49].

Accordingly, the current study highlighted several insights that may participate in enhanced agricultural modernization policies and Côte d’Ivoire. Many studies have been published in different areas with different research directions focusing on improved cacao production. From theoretical and practical contributions, this study extends the understanding of the cacao production challenge, namely the role—and simultaneously, highlights the limits faced by financial institutions’ input credit—in agricultural modernization in developing countries such as Côte d’Ivoire. Ultimately, the contribution of this research was the insight provided by the analysis in identifying the main political practices for implementing sustainable and suitable cacao production in Côte d’Ivoire.

6. Conclusions

In this study, the limitations on the utilization of IC were investigated, as well as the variables influencing the degree of satisfaction of cocoa farmers in Côte d’Ivoire. Regarding availability, accessibility, and efficacy, respondents expressed satisfaction with the restrictions relating to the usage of IC. We concluded that 61.9% of farmers were dissatisfied with using IC, while 38.03% were satisfied. It was also concluded that IC services constrained 61.15% of farmers, while 38.85% were not. However, regarding the qualitative feature of diversity, farmers had varying perspectives on these services. Because they were dissatisfied with the timing of distribution, it was clear that these services needed to be enhanced in terms of crop chronology respect, costs of input, and the delivery of services for diverse agricultural activities to many stakeholders utilizing a variety of communication channels.

The logistic regression model’s findings revealed the areas that innovators and cooperative management need to focus on to raise farmers’ satisfaction levels. The empirical results demonstrated that farmers’ satisfaction was statistically and significantly influenced by annual production, number of years of use, farm size, education, age, and input price. The likelihood of farmer satisfaction would be reduced due to the large and detrimental effects of the number of years of IC use, farmer age, and farm size. The impact of Swollen Shoot Disease (SSD) on farms in the research area also caught the responders’ attention. In order to combat this plague, this study advises using participatory strategies, such as providing support and ongoing training for cooperative agricultural inventors and leaders, in addition to the extension service.

Accordingly, it is necessary to provide recommendations to meet farmers’ expectations. So, from this investigation, cacao production and modernization need a long-term solution. These solutions include the distribution and access to agricultural technologies and the establishment of accountability frameworks. These measures should encourage monitoring, training, and evaluation of agricultural innovation technologies. In addition, future studies should focus on monitoring agroecological conditions likely to increase input efficiency and the impact of agricultural technologies on production. In a nutshell, strengthening human capital knowledge and encouraging land consolidation can optimize cocoa production and enhance farmers’ livelihood.