Assessment of sustainability in cocoa farms in Ecuador: application of a multidimensional indicator-based framework

ABSTRACT

Cocoa is one of Ecuador’s main export products. Over the last decades, cocoa production in the country has increased considerably, making it the largest producer of cocoa in Latin America. However, this increase has been accompanied by environmental degradation, endangering the sustainability of cocoa cultivation. This study utilizes an innovative indicator-based framework to evaluate the sustainability performance of fine flavour and CCN-51 cocoa farms in Ecuador, across economic, social, environmental, nutrition and health, and governance dimensions. Data were collected from 169 farms in Manabí Province and multiple statistical tests were employed to discern sustainability scores. The results highlighted significant differences in sustainability between the two cocoa production systems, influenced by factors such as ethnicity, age, and level of education. A 53% variation in sustainability scores between the fine flavour and CCN-51 cocoa farms was identified. The association of lower sustainability scores with limitations of physical and financial capital provides insights for enhancing sustainable cocoa farming in Ecuador.

KEYWORDS:

• Sustainability
• Indicators
• Statistical analysis
• Farms
• Cocoa
• Ecuador

Introduction

The cocoa industry holds a significant place in Ecuador’s economy, contributing over 700 million USD (8%) to the country’s agricultural GDP and employing 15% of the rural population (MAG, 2020). In 2022, cocoa cultivation spanned 591,557 hectares, equivalent to 41.3% of the country’s total agricultural land use, making it the most substantial area committed to a single perennial crop (INEC-ESPAC, 2022). Ecuador exports most of its cocoa production, with only 2% consumed domestically. Most of the cocoa is exported as cocoa beans, and only 10–12% of total production is exported as semi-finished cocoa products. Due to its economic relevance in the country, recent policies aim to boost cocoa yields and productivity and enhance industrial processes in small and medium enterprises, including the government’s Improvement Plan for Competitiveness in the Agro-industrial Development of the Cocoa-Chocolate Chain (‘Plan de Mejora Competitiva para Desarrollo Agroindustrial de la Cadena de Cacao-Chocolate’) (Arvelo et al., 2016). As a result, cocoa production in Ecuador is expanding, and the country’s cocoa industry has experienced robust growth, positioning Ecuador as a major player in the global cocoa market, where prices have risen by as much as 47% to $10,455 per ton in March 2024 (ICCO, 2024)

The cocoa industry in Ecuador faces a number of challenges and opportunities in terms of sustainability, one of the main challenges being the threat posed by climate change and deforestation, linked to the expansion of cocoa plantations. This also raises concerns about biodiversity loss and the impact on local ecosystems, which is directly related to SDG 15 (Life on Land) (Bleischwitz et al., 2018). In addition, social aspects, such as rural poverty and working conditions in cocoa plantations, underscore the importance of addressing SDG 1 (No Poverty) and SDG 8 (Decent Work and Economic Growth) (Luna & Barcellos-Paula, 2024). To ensure the long-term sustainability of the cocoa industry in Ecuador, it is crucial to adopt sustainable agricultural practices, promote the conservation of natural ecosystems, improve working conditions, and foster the social and economic inclusion of local communities (De Marco-Larrauri et al., 2016; Díaz-Montenegro et al., 2018).

Cocoa farming is a multifaceted industry that involves social, environmental, economic, nutrition, and governance dimensions, presenting significant challenges that ripple through farmers, communities, the environment, and the sector as a whole (Tennhardt et al., 2022). Socioeconomic issues are prominent aspects of these challenges. In Ghana, for instance, challenges include child labour as a driver and the effect of poverty linked to cocoa production systems (Luckstead et al., 2019). More broadly, in African smallholder cocoa systems and their Latin American counterparts, gender inequality lags behind other sustainability aspects and is evidenced as a persistent issue (Kuhn et al., 2023). In addition, most cocoa farmers (the majority of whom are small-scale farmers) grapple with scanty resources and substandard living conditions (Boysen et al., 2023).

Although global demand for chocolate is high, farmers receive only a tiny fraction of the final price of the product (Zambrano-Argandoña, 2020). Intermediation by multiple actors in the supply chain, a lack of price transparency, and unfair trade practices contribute to exploitation and a lack of economic benefits for farmers (Dontenville et al., 2022). Unpredictable prices in the global cocoa market and farmers’ limited bargaining power often culminate in inadequate income levels. These factors leave farmers unable to provide for their families or improve their living conditions (Kumar et al., 2022).

While the socioeconomic issues present significant hurdles for cocoa farmers, environmental challenges further complicate their efforts to sustain production. Among these environmental issues in cocoa production, deforestation associated with cocoa cultivation releases large amounts of carbon stored in forests, contributing to increased greenhouse gas emissions (Vervuurt et al., 2022). Furthermore, the intensive use of pesticides and chemical fertilizers in cocoa production contaminates soils and water sources, negatively affecting the health of ecosystems and local communities (Zambrano-Argandoña, 2020).

In addition to the environmental impacts, the nutrition and health of cocoa farmers and consumers are also at risk. For consumers, there is growing concern regarding food quality and safety in cocoa products. The presence of pesticide residues and contaminants in cocoa beans can affect the quality and safety of finished chocolate products (Villanueva et al., 2023). As for cocoa producers, micronutrient deficiencies in the diets of farmers and local communities can be a problem in cocoa-producing regions, exacerbating malnutrition and health challenges in those areas (Guest et al., 2023). In addition, the European Union’s regulation on cadmium in cocoa products also impacts and concerns various stakeholders in the country’s value chain. Based on the cadmium map in Ecuador’s cocoa by Argüello et al. (2019), areas like the northern Amazon and the coastal provinces of Manabí, El Oro, and Guayas exceed the EU’s 0.6 mg/kg standard. EU countries’ importance as cocoa buyers means addressing this to maintain market access is vital.

Effective governance in the cocoa sector involves establishing and enforcing regulations that ensure sustainable and ethical practices throughout the supply chain. The lack of effective regulations and control mechanisms in cocoa production hinders the maintenance of sustainable and ethical practices (Lee & Park, 2023). The cocoa supply chain is complex and often opaque, making traceability and accountability difficult in terms of fair labour practices, environmental conservation, and compliance with food quality and safety standards (Ton et al., 2023).

Innovation in the cocoa value chains in Ecuador involves various stakeholders who, through cooperation, seek to overcome technical and market challenges to improve the competitiveness and sustainability of the sector (Mora et al., 2024). The concept of agricultural sustainability has evolved significantly in recent decades, from focusing exclusively on production to a more holistic and balanced approach that considers environmental, economic, social, and governance aspects (Mottet et al., 2020; Perrin et al., 2023). It is increasingly recognized that sustainability involves not only environmentally responsible practices but also the economic viability of farms and the achievement of social criteria, such as equity, labour rights, and access to land and resources for farmers and rural communities (Hayati et al., 2010). Related efforts also promote the active participation of farmers in decision-making and the adoption of agricultural practices that contribute to the social and cultural well-being of local communities (Ton et al., 2023). In fact, international demand is beginning to favour countries that promote the holistic sustainability of their cocoa production chains (Bermúdez et al., 2022), a key aspect in conducting this sustainability-focused analysis.

However, existing research on cocoa sustainability in Ecuador is fragmented and often focuses on single dimensions (primarily environmental) (Anzules-Toala et al., 2021; Barrezueta-Unda & Paz-González, 2018; Tennhardt et al., 2022). This paper addresses this research gap by adopting a holistic approach that includes the five dimensions of sustainability (environmental, economic, social, nutrition and health, and governance). The aim of the study was to further scientific understanding of cocoa farm sustainability in Ecuador by providing a more comprehensive perspective of farm sustainability performance. To this end, we used a large dataset of 169 cocoa farms and a novel indicator-based sustainability assessment framework (FARMTOOLS) that includes 151 indicators grouped into different sustainability dimensions.

Specifically, the study assesses and compares the sustainability performance of cocoa farms producing the two main varieties of cocoa in Ecuador: fine flavour cocoa, a high-quality cocoa often used in gourmet products because of its exceptional taste; and CCN-51 cocoa, a high-yielding, but less aromatic bulk cocoa used in mass chocolate products. This comparison is relevant because CCN-51 cocoa is rapidly replacing fine flavour cocoa, which is considered a symbol of Ecuadorian identity and is key to Ecuador’s positioning in international cocoa markets (Díaz-Montenegro et al., 2018). In an effort to reverse this situation, Ecuadorian authorities have developed an economic development strategy, the National Fine Flavour Cocoa Reactivation Plan (Plan de Reactivación del Cacao Nacional Fino de Aroma or PRCN), which aims to alleviate poverty and improve the livelihoods of fine flavour cocoa farmers by promoting the production of this cocoa variety, improving productivity, reinforcing producer organizations, and strengthening the whole value chain (MAGAP, 2012).

The results of this study offer insights for scholars, policymakers, and practitioners regarding various performance indicators of cocoa sustainability in Ecuador, thus facilitating a better understanding of ways to achieve more sustainable cocoa farming systems. Furthermore, this paper provides reflections on the methodological framework applied and its potential implications for rural livelihoods and the sustainability of cocoa production systems.

2. Methodology

2.1. The indicator-based framework

Various frameworks have been established to measure sustainability in a meaningful way, and scholars hold diverse opinions regarding the optimal approach for this. In Ecuador, predominant frameworks for assessing farm-level sustainability include: the FAO’s Sustainability Assessment of Food and Agriculture Systems (SAFA) (Bonisoli, Galdeano-Gómez, Piedra-Muñoz, & Pérez-Mesa, 2019; Cayambe et al., 2023; Heredia-R et al., 2022); the Response-Inducing Sustainability Evaluation (RISE) (Heredia-R et al., 2020, 2021); and the Framework for Evaluating Natural Resource Management Systems by Incorporating Sustainability Indicators, or MESMIS for its Spanish acronym (Arnés & Astier, 2018).

All these frameworks address sustainability assessment as a process that must consider multiple criteria, address many conflicting dimensions, and involve an understanding of and adaptation to local contexts. However, the existing frameworks remain focused on the classic dimensions of sustainability (economic, social, and environmental), ignoring other important aspects (nutrition, health, and governance), which are gaining relevance in sustainable development (Oñederra-Aramendi et al., 2023).

Addressing this limitation, this study introduces a novel multidimensional indicator-based framework – FARMTOOLS – which encompasses 151 indicators across five dimensions of sustainability: economic (32), social (42), environmental (56), nutrition and health (7), and governance (14) (refer to Supplementary material 1) (Blanco-Gutiérrez et al., forthcoming). FARMTOOLS seeks to extend the boundaries of existing frameworks, offering a more nuanced and comprehensive perspective on farm sustainability performance. Positioned as a flexible, indicator-based tool, FARMTOOLS is geared towards informing policy and investment decisions through a four-phase approach: defining scope, selecting indicators, collecting data, and assessing sustainability.

2.2. Definition of the scope

This step includes defining the system boundaries and characterizing the rural livelihoods (farming context).

Defining the system boundaries includes selecting the agricultural product to be assessed and defining the geographic setting. These boundaries establish the limits within which sustainability indicators are identified. In the present study, the system boundaries were defined in collaboration with FAO and the Ecuadorian Ministry of Agriculture (MAG). Cocoa was chosen as the agricultural product to be assessed because it is an important cash crop in Ecuador and its sustainability is currently threatened. The geographic boundary was set as the province of Manabí in the western part of the country (the coastal region) (Figure 1). As illustrated in Figure 1, FAO’s Hand-in-Hand Initiative categorizes parts of the study area as ‘high priority’ and other parts as ‘medium priority with high agricultural opportunities’. This indicates that the study area is a region with medium to high poverty levels and high agricultural potential but with low agricultural efficiency due to poor infrastructure and services, market failure, and marketing issues.

Figure 1. Study area. Source: Author’s own creation.

The province of Manabí is the third largest producer of cocoa in Ecuador. In 2021, 48,661 metric tons of dry cocoa were produced in the province out of a national total of 102,207 ha (MAG, 2022). The humid climate and fertile soils of Manabí favour cocoa production. In fact, this province accounts for 12% of Ecuador’s total area of cocoa cultivation and has the highest level of suitability for cocoa cultivation in the country (MAG, 2020). The cocoa produced in Manabí is considered to be of high quality due to the unusually high presence of old-growth cocoa trees exhibiting the fine flavour characteristics. This situation is changing, however. The high-yielding cocoa variety CCN-51, known for its mediocre flavour, is rapidly replacing the fine flavour cocoa variety. Our study focuses on the key areas of Manabí for the cultivation of the two cocoa varieties: the canton of El Carmen for the CCN-51 variety, and Bolívar, Chone, Junín, and Tosagua cantons for fine flavour cocoa.

Rural livelihoods were characterized using the capital theory (DFID, 2000). This theory is based on the economic concept of wealth creation or ‘capital’ that considers five types of sustainable capital: natural (resources used to produce goods and services); human (people’s knowledge and skills); social (institutions and partnerships); physical (material goods or fixed assets that contribute to the production process); and financial (economic resources). Following Torres et al. (2022), different variables were used to represent farming households’ capital and characterize rural livelihoods in the study area. These variables are important because they may condition the sustainability of cocoa farms. According to Torres et al. (2022), the variables can be classified into three categories: human/social, natural, and physical/financial (Table 1).

Table 1. Types of capital and variables used to characterize the rural livelihoods of cocoa farmers.

Capital	Variables
Human and social	Ethnicity; gender; age; household members; education of household head; training; youth and child labor; membership in an association; trust in neighbors; COVID-19 infection.
Natural	Farm area (ha); cocoa area (ha); forest area (ha); pastureland (ha); area of other crops (ha); existence of natural water sources; diversity of crops.
Physical and financial	Internet access; land property rights; cell phone; motorcycle or car; chainsaw; grass strimmer; distance from the farm to a city (km); type of access road; access to credit; receives government subsidies; off-farm income; indebtedness; farm income covers debt payments.

Source: Author’s own creation.

2.3. Selection of indicators

The indicator selection phase focused on choosing the most suitable sustainability indicators. Initial selection was informed by an extensive review of literature related to cocoa farm sustainability assessment. This review resulted in a shortlist derived from the general list of indicators available in the FARMTOOLS framework (Appendix 1). This shortlist was subsequently reviewed by a team of FAO experts to ensure its relevance and suitability to the local context, culminating in a validated and finalized list of indicators.

In total, 35 indicators were evaluated (Table 2). These indicators were combined into 15 subdimensions, which were then combined into 5 sustainability dimensions (environmental, economic, social, nutrition and health, and governance).

Table 2. Indicators applied in the assessment of cocoa farm sustainability in the study area.

Dimension	Subdimension	Indicator	Description
Environmental	Biodiversity	Crops	Number of crops
		Animals	Number of animals
		Connectivity	Degree of connectivity between the agro-ecosystem and the landscape
	Risk of erosion	Row orientation	Cocoa row routing in relation to slope
	Soil conservation	Soil management	Soil alteration and % of soil covered by residues and crops
		Recycling	Level of recycling of crop residues and by-products
	Environmental efficiency	Water saving	Number of water-saving technologies applied
		Renewable energy	Production and use of renewable energy
	Management practices	Fertilization	Fertilizers applied
		Pest and disease control	Use of chemical pesticides
Economic	Economic risk	Productive activities	Number of sales (diversification)
		Commercialization	Number of commercialization channels
		Income stability	Income variability and ability to recover from shocks
		Vulnerability	Mechanisms to reduce vulnerability (e.g. insurance and loans)
	Value chain	Producer cooperation	Cooperation with producer organizations
		Negotiation power	Bargaining power in the market and relationship with consumers
	Economic viability	Indebtedness	Indebtedness and repayment capacity
		Economic profitability	Profits/Total assets
		Cost-benefit analysis	Benefits/Production costs
		Poverty	Monthly income per capita (USD)
Social	Satisfaction of basic needs	Basic services	Access to basic services (electricity and drinking water)
		Education	Level of education
	Working conditions and generational change	Working conditions	Level of satisfaction with labour conditions (salary and risks)
		Generational change	Abandonment of agricultural activity and emigration
	Ecological awareness	Knowledge and ecological awareness	Farmers’ environmental awareness
Nutrition and health	Adequate nutrition	Coverage of food needs	Provision of the household’s food needs
		Diversity and quality of nutrition	Quality of food
	Health	Hygiene and safety	Safety of facilities and hygiene on farms
		Health access and coverage of medical centres	Proximity and coverage of medical centres
Governance	Rule of law	Land tenure	Land rights and conflict resolution
		Law enforcement	Law enforcement effectiveness
		Transparency and corruption	Transparency and corruption in the municipal government
	Participation	Governance of land and natural resources	Participation of producers in land and natural resource governance
		Participation in networks and organizations	Participation of producers in local organizations
		Women’s empowerment	Women’s participation in decision-making and access to resources

Source: Authors’ own creation.

The quantification of these indicators followed a modified Likert-type scale from 0 to 4 for the least and most sustainable practices. Detailed information on the indicators and their measurement scale can be found in Supplementary material 2, Part II.

2.4. Data collection

Due to the lack of statistical information at the farm level, the data used for this study were collected through surveys conducted with cocoa producers. A total of 169 surveys were conducted: 100 with fine flavour cocoa producers in the cantons of Bolívar (42), Tosagua (28), Junín (17), and Chone (13), and 69 with CCN-51 cocoa producers located in the El Carmen canton (Figure 1).

The sampling scheme employed two techniques based on territorial and cultural complexity: (1) Groups of fine flavour and CCN-51 cocoa producers were defined and a cluster sampling approach was used (probabilistic) with the criteria that the producers belong to a producers’ association and identify as Montubio, Mestizo, or Indigenous (Flynn, 2001). (2) Producers within each group were selected and a snowball sampling technique was used (non-probabilistic), beginning with key informants (Kirchherr & Charles, 2018).

The survey consists of two parts: Part I has 39 open questions on rural livelihoods while Part II has 35 closed questions regarding aspects of sustainability. The scoring scale in Part II follows the TAPE methodology (FAO, 2019). For each question, respondents chose from five possible answers (ranked from least to most sustainable) with scores from 0 to 4 (a modified Likert-type scale). (See Appendix B for the full survey.)

The surveys were conducted face-to-face with household heads between July and September 2022. Preliminary tests were carried out on 10 heads of households to determine the validity of the questions, identify technical problems, evaluate the duration, and identify coherence and logic. The average duration of the survey was one hour and twenty minutes and each survey was conducted in an environment of trust in which the producers were assured of strict confidentiality. The process of obtaining free and informed consent was carried out with the support of the leaders of the producers’ associations, through whom all approaches to the households were made.

The surveys were conducted by a team of 12 research assistants, following a 3-day training course. Data were collected on mobile devices, using the digital data collection software KoBoToolbox, developed by the Harvard Humanitarian Initiative (https://www.kobotoolbox.org/).

2.5. Sustainability assessment

In this study, the assessment was performed at the indicator level (Table 2). The final value of each indicator was obtained by calculating the average of all the household scores for each indicator. Furthermore, a traffic light rating system was applied to facilitate the interpretation of the results (FAO, 2019). Each indicator was classified into four performance categories defined as: good performance (green, scores between 3 and 4); medium performance (yellow, scores of 2 to <3); fair performance (orange, scores of 1 to <2); and bad performance (red, scores of 0 to <1). When needed, aggregated values at the level of subdimension or dimension were obtained using the same procedure. The subdimension scores are the average values of the indicators within each subdimension, while the dimension scores are the average values of the subdimensions within each dimension.

The differences between the sustainability indicator scores of fine flavour and CCN-51 cocoa farms and capital variables were analysed using univariate analysis, namely the Mann–Whitney U test for continuous variables, Fisher’s exact test for binary variables, and the Chi-squared test for multinomial responses. A confidence level of 99% was used.

At the subdimension level, the differences in sustainability between fine flavour and CCN-51 cocoa farms were approached through multivariate analysis using discriminant analysis. To represent the variation of the total set of cocoa producers and to detect possible patterns in the subdimensions of sustainability simultaneously, a principal component analysis (PCA) was performed. The resulting graph (biplot) was used to explore the main sources of variation that explain the differences in sustainability between farmers and between fine flavour and CCN-51 cocoa production systems.

In addition, to determine the relationship between rural livelihoods (capital theory variables) and sustainability assessment, measures of association were calculated between the first principal component of the PCA (PC1) and the capital variables. The association was studied only for the first component since the second was linked to differences between cocoa producer groups (fine flavour and CCN-51), which were addressed using discriminant analysis. For quantitative capital variables, Spearman’s correlation coefficient was calculated. For the binary or multinomial equity variables, the Mann–Whitney U test was performed. Significant correlation coefficients (positive or negative) or significant differences in the U test indicated an association between capitals and the principal component, which produced the differences in the sustainability of the cocoa producers. Statistical analyses were carried out using the InfoStat package (http://www.infostat.com.ar).

3. Results

3.1. Characterization of rural livelihoods

3.1.1. Human and social capital

The variables of ethnicity, age, education, training, and membership in a producers’ association showed significant differences between fine flavour and CCN-51 cocoa producers (Table 3). (See Supplementary material 3 for the variables that did not present significant differences for each type of capital: human and social, natural, physical, and financial.)

Table 3. Variables with significant differences in human and social capital between fine flavour and CCN-51 cocoa farmers.

Variable		Fine flavour cocoa	CCN-51 cocoa	p-value
Ethnicity (%)	Indigenous	0.00	1.00	0.0004
	Montubio	59.00	29.00
	Mestizo	41.00	70.00
Age (years)		57.7	48.54	0.0001
Education (years)		7.67	9.54	0.0017
Membership of an association (%)	Yes	99.00	39.00	<0.0001
	No	1.00	61.00

Source: Authors’ own creation.

Montubio farmers are more engaged in fine flavour cocoa cultivation than in CCN-51 cocoa cultivation (59% of Montubio farmers compared to 29%). Moreover, farmers cultivating fine flavour cocoa tend to be older (by 9.29 years) than those growing CCN-51 cocoa (Table 3). By comparison, CCN-51 cocoa farmers are mainly Mestizo (70%) and have completed more years of formal education than fine flavour cocoa farmers (a gap of 1.87 years). In terms of membership in a producers’ association, more fine flavour cocoa farmers belong to an association than CCN-51 cocoa farmers (60% more).

3.1.2. Natural capital

The variables of total farm area, cocoa cultivation, area used for other crops, and existence of natural water sources showed significant differences between fine flavour and CCN-51 cocoa farms (Table 4).

Table 4. Variables with significant differences in natural capital between fine flavour and CCN-51 cocoa farms.

Variable		Fine flavour cocoa	CCN-51 cocoa	p-value
Total farm area (ha)		4.60	9.89	<0.0001
Cocoa cultivation (ha)		1.82	2.62	0.0009
Other crops (ha)		1.25	4.28	<0.0001
Existence of natural water sources on the farm (%)	Yes	40.00	96.00	<0.0001
	No	60.00	4.00

Source: Authors’ own creation.

CCN-51 cocoa farms are generally larger than fine flavour cocoa farms (by 5.29 ha). They also have a larger area for cocoa production (0.8 ha) and for other crops (3.03 ha) and more natural water sources on the farm (56%) (Table 4).

3.1.3. Physical and financial capital

The variables of ownership of a motorcycle or car, ownership of a grass strimmer, distance from the farm to a city, type of access road, indebtedness, and ‘farm income covers debt payments’ showed significant differences between farmers producing fine flavour and CCN-51 cocoa (Table 5).

Table 5. Variables with significant differences in physical and financial capital between fine flavour and CCN-51 cocoa farmers.

Variable		Fine flavour cocoa	CCN-51 cocoa	p-value
Has motorcycle or car (%)	Yes	47.00	84.00	<0.0001
	No	53.00	16.00
Has grass strimmer (%)	Yes	29.00	75.00	<0.0001
	No	71.00	25.00
Distance from the farm to a city (km)		6,28	12.83	0.0003
Type of access road	Gravel	28.00	41.00	0.0002
	Stone	24.00	11.00
	Asphalt	48.00	17.00
Has debts (%)	Yes	32.00	61.00	0.0003
	No	68.00	39.00
Farm income covers debt payments (%)	Yes	17.00	57.00	<0.0001
	No	83.00	43.00

Source: Authors’ own creation.

More CCN-51 cocoa farmers own motorcycles or cars (37%) and grass strimmers (46%) than fine flavour cocoa farmers. The CCN-51 cocoa farms are farther away from a city (by 6.55 km), have more debts (by 29%), and more heads of households consider that the farm income covers debt payments (by 40%) compared to the fine flavour cocoa farms.

3.2. Sustainability performance

3.2.1. At indicator level

Table 6 shows the sustainability scores per indicator. Fifteen of the 35 indicators evaluated (43%) show significant differences between fine flavour and CCN-51 cocoa production systems.

Table 6. Sustainability performance of farms growing fine flavour and CCN-51 cocoa.

Note: Red signifies bad performance; orange signifies fair performance; yellow signifies medium performance; and green signifies good performance. Indicators showing significant differences between fine flavour and CCN-51 cocoa are indicated in bold (* indicates statistically significant difference p<0.01, 99% confidence).

Source: Authors’ own creation.

In the environmental dimension, the average scores were 1.56 for fine flavour cocoa farms and 1.31 for CCN-51 cocoa farms (see Supplementary material 4). Of the ten indicators evaluated, six showed significant differences between the two production systems, namely pest and disease control, fertilization, connectivity, animals, row orientation, and renewable energy (Table 6). Of these, the only indicator with good performance was pest and disease control in fine flavour cocoa farms. The indicators with medium performance were connectivity in CCN-51 cocoa farms and fertilization in fine flavour cocoa farms. The indicators with fair performance were animals in both fine flavour and CCN-51 cocoa farms, and connectivity in fine flavour cocoa farms. Finally, the following indicators had bad performance: renewable energy and row orientation in both fine flavour and CCN-51 cocoa farms, and fertilization and pest and disease control in CCN-51 farms.

In the economic dimension, the average scores were 1.44 for fine flavour cocoa farming and 1.34 for CCN-51 cocoa farming (see Supplementary material 4). Of the ten indicators evaluated, five showed significant differences between the two production systems: income stability, economic profitability, negotiation power, producer cooperation, and indebtedness (Table 6). Of these, the only indicator with good performance was indebtedness for fine flavour cocoa farms. The indicators with medium performance were producer cooperation for fine flavour cocoa farms and indebtedness and economic profitability for CCN-51 cocoa farms. The indicators with fair performance were economic profitability and negotiation power for fine flavour cocoa farms, and producer cooperation for CCN-51 cocoa farms. The indicators with bad performance were negotiation power for CCN-51 cocoa farms, and income stability for fine flavour and CCN-51 cocoa farms.

In the social dimension, the average scores were 2.42 for fine flavour cocoa farms and 1.93 for CCN-51 cocoa farms (see Supplementary material 4). Of the five indicators evaluated, two showed significant differences between production systems: basic services and education (Table 6). Of these, the only indicator with good performance was basic services for fine flavour cocoa farms. The indicators with medium performance were education for fine flavour and CCN-51 cocoa farms, and basic services for CCN-51 cocoa farms.

In the nutrition and health dimension, the average scores were 2.32 for fine flavour cocoa farms and 2.16 for CCN-51 cocoa farms (see Supplementary material 4). Of the four indicators assessed, only the indicator hygiene and safety showed significant differences between the two production systems (Table 6): fine flavour cocoa farms showed good performance, while CCN-51 cocoa farms showed medium performance.

In the governance dimension, the average scores were 2.54 for fine flavour cocoa farms and 2.04 for CCN-51 cocoa farms (see Supplementary material 4). Of the six indicators evaluated, only law enforcement showed significant differences between the two production systems (Table 6), showing a medium performance for fine flavour cocoa farms and fair performance for CCN-51 cocoa farms.

3.2.2. At subdimension level

The PCA explained 53% of the variability among cocoa farmers in sustainability performance scores at the subdimension level, which was expressed as a 2-component biplot (PC1: 38.6%; PC2: 14.3%) (Figure 2).

Figure 2. Ranking of CCN-51 and fine flavor cocoa production systems according to sustainability performance variables at the subdimension level. Note: Biodiversity (ES1), Risk of erosion (ES2), Soil conservation (ES3), Environmental efficiency (ES4), Management practices (ES5), Economic risk (ECS1), Value chain (ECS2), Economic viability (ECS3), Satisfaction of basic needs (SS1), Working conditions and generational change (SS2), Ecological awareness (SS3), Adequate nutrition (NHS1), Health (NHS2), Rule of law (GS1), and Participation (GS2). Source: Authors’ own creation.

Most of the eigenvectors associated with PC1 are positive (a perpendicular projection of the biplot vectors on the horizontal axis). This indicates that the main differences between producers of the two types of cocoa are seen when ordering them from less sustainable (left) to more sustainable (right). There is some discontinuity in the ordering, differentiating a small group of producers (extreme cases on the left) with very low levels of sustainability for most of the subdimensions studied. These producers stand out for having low performance values in the subdimensions governance and participation (GS2), ecological awareness (SS3), adequate nutrition (NHS1), and management practices (ES5). These subdimensions are the most relevant in the explanation of the first component.

PC2, which accounts for 14.3% of the variability and is independent of PC1, separates cocoa producers according to fine flavour (top) and CCN-51 (bottom) production systems. This analysis addresses the differences between fine flavour and CCN-51 cocoa groups. The differences are further explained by discriminant analysis, which allowed us to fit a linear discriminant function based on the sustainability values at the subdimension level and separate CNN-51 and fine flavour cocoa farms with a low cross-classification error rate (5.92%).

Figure 3 shows the separation of fine flavour and CCN-51 cocoa farms on the discriminant axis. The frequency distributions of both types of farms on the discriminant axis presented a very low percentage of overlap and were widely different in their average values. This result indicates that, based on the combination of adjusted sustainability subdimensions (discriminant function, Table 7), the CCN-51 and fine flavour cocoa farm types were distinct.

Figure 3. Frequency distribution of CCN-51 and fine flavour cocoa farms on the discriminant axis adjusted based on the sustainability subdimensions. Source: Authors’ own creation.

Table 7. Weights of the subdimensions of sustainability in the discriminant function between fine flavour and CCN-51 cocoa farms.

Subdimension of sustainability	Code	Discriminant function
Risk of erosion	ES2	-0.34
Environmental efficiency	ES4	-0.31
Adequate nutrition	NHS1	−0.31
Soil conservation	ES3	−0.19
Working conditions and generational change	SS2	−0.19
Biodiversity	ES1	−0.18
Economic viability	ECS3	−0.16
Governance and participation	GS2	−0.12
Value chain	ECS2	−0.05
Ecological awareness	SS3	−0.02
Rule of law	GS1	0.11
Economic risk	ECS1	0.23
Satisfaction of basic needs	SS1	0.28
Health	NHS2	0.32
Management practices	ES5	0.96

Source: Authors’ own creation.

This indicates that the CCN-51 and fine flavour cocoa producer groups differ mainly in the subdimensions of sustainability shown in Table 7. Thus, fine flavour cocoa farms have higher values for sustainability than CCH-51 cocoa farms in management practices (ES5), health (NHS2), satisfaction of basic needs (SS1), and economic risk (ECS1), while having lower values for sustainability in risk of erosion (ES2), environmental efficiency (ES4), and adequate nutrition (NHS1).

3.3. Link between sustainability and rural livelihoods

The analysis shows that there are nine capital variables with significant correlations or mean differences in PC1 (Table 8). This information is useful in identifying possible capital variables that are related to the ranking of producers from highest to lowest sustainability in PC1. In general, producers with lower sustainability values (on the left in Figure 2) are located further away from cities and do not have access to the Internet or credit (physical and financial capital). In addition, producers with low sustainability performance are characterized by being small (smaller total farm area, pastureland, and forest area) and having no wild animals (natural capital). At the social level, it is mostly women who do not belong to an association. The opposite is true for these variables for the most sustainable producers (on the right in Figure 2).

Table 8. Links between capital theory variables and sustainability performance (PC1).

Capital theory variable		Measure of association		p-value
		Measure	Spearman’s correlation coefficient
Human and social
Gender	F		−0.74	0.0072
	M		0.44
Member of an association	Yes		0.43	0.0003
	No		−1.25
Natural
Total farm area (ha)			0.21	0.0074
Area of pastureland (ha)			0.21	0.0059
Area of forest (ha)			0.22	0.0035
Existence of wild animals	Yes	0.23		0.0003
	No	−2.07
Physical and financial
Distance from farm to city (km)			−0.19	0.0100
Internet access	Yes	0.33		0.0018
	No	−0.78
Access to credit	Yes	0.73		<0.0001
	No	−0.9

Note: Only significant correlations. P-values are <0.01, showing 99% confidence.

Source: Authors’ own creation.

3. Discussion

We applied a novel indicator-based framework to assess farm-level sustainability, encompassing five dimensions of sustainable development: environmental, economic, social, nutrition and health, and governance, offering a comprehensive perspective of sustainability performance. The proposed framework can be used as a self-assessment tool by farmers to monitor and improve their sustainability at the farm level and by policymakers to support responsible and sustainable investments and design actions to improve sustainability. As with other sustainability assessment techniques, the most important limitation in implementing the framework was data collection. In this study, and others like it, conducting surveys has been critical (FAO, 2019). In this regard, surveys were conducted in those areas within Manabí province where CCN-51 and fine flavour cocoa were most representative to ensure a significant number of responses for each cocoa variety. These areas, however, are different and there could be an effect of location that could be analysed in this study.

Overall, the lowest sustainability scores were obtained in the economic and environmental dimensions (fair performance), and the highest scores were in the governance, nutrition and health, and social dimensions (medium performance). This underlines the importance of governance and nutrition and health aspects in optimizing sustainability performance at the farm level and, in line with other studies, supports the need to go beyond the classic dimensions of sustainability (economic, social, and environmental) (Mottet et al., 2020; Oñederra-Aramendi et al., 2023; Tennhardt et al., 2022). Trade-offs and synergies between sustainability dimensions have not been analysed in the present study but can have important policy implications and should be considered in future research (Schader et al., 2016).

In addition, the analysis shows significant differences at the indicator level and allows the identification of hotspots (points with low sustainability performance), where policy interventions may have a significant impact. In the environmental dimension, the low scores obtained for CCN-51 cocoa farms are related to the indicators of fertilization and pest and disease control, which is corroborated by the fact that CCN-51 cocoa (a clone) requires greater use of fertilizers and pesticides (Zambrano-Argandoña, 2020). In fine flavour cocoa, the row orientation indicator is a serious problem, as it reveals soil erosion that often results in the degradation of ecosystem services and functions (Zambon et al., 2021).

Regarding the economic dimension, the worst-performing indicators for both types of cocoa farms are income stability, vulnerability, and negotiation power. Similar results were obtained in Indonesia, Peru, and Cameroon, where producers share concerns about price volatility and weak bargaining power (Mithöfer et al., 2017). Along these lines, recent studies (Ramírez-Argueta et al., 2022; Van Vliet et al., 2021) highlight the need to develop strategies aimed at improving market structures and diversifying cocoa systems (e.g. through agroforestry arrangements) to increase farmer incomes and promote economic stability.

In the social dimension, generational change is the lowest-scoring indicator for both cocoa types. This is consistent with cocoa production findings by Torres et al. (2022) in high mountain areas in Ecuador and by Bandanaa et al. (2021) in Ghana. These studies highlight the aging of cocoa farms and the abandonment of farming as major challenges and suggest implementing measures to support young farmers, including women. In contrast, two cocoa-sector concerns commonly reported in the literature (Luckstead et al., 2019; Tennhardt et al., 2022) are not significant problems in this study area, namely low levels of education and child and forced labour.

Regarding the nutrition and health dimension, the low scores were reflected in the indicators of health access and coverage of medical centres and coverage of food needs, for both fine flavour and CCN-51 cocoa-producing households. Studies in Indonesia and Uganda (Guest et al., 2023; Tennhardt et al., 2022) also suggest that poor health and nutrition are important constraints to improving cocoa production and producers’ livelihoods. In this regard, One Health approaches can be more effective in increasing productivity and improving the livelihoods of smallholder cocoa farmers through education, health, and income diversification strategies than conventional farmer training, which is based on agronomic and mechanical methods (Guest et al., 2023).

In the governance dimension, the lowest scores were in transparency and corruption for fine flavour cocoa farms, and in law enforcement, conflict resolution, and participation in networks and organizations for CCN-51 farms. Similar results were found in smallholder farms in Indonesia, where programmes were implemented to strengthen institutional management and successful collaborative networks were formed (Arsyad et al., 2020). Although the analysis did not identify a significant relationship between ethnicity (human capital) and sustainability performance (as seen in other studies, such as Heredia-R et al., 2022), ethnicity does appear to be related to governance outcomes. For instance, in this study, 60% more Montubio smallholder producers belong to associations and networks than Mestizo smallholder producers. This may explain why fine flavour cocoa farms, with a higher presence of Montubio producers, achieve better sustainability results in governance than CCN-51 cocoa farms (predominantly Mestizos).

In aggregated terms, the most sustainable production system is that of fine flavour cocoa, although it is categorized as fair, and performance across indicators and subdimensions is uneven. The greater sustainability of fine flavour cocoa farms may be partly explained by the greater level of membership in producers’ associations and the existing certifications of fine flavour cocoa. In this regard, Amfo et al. (2021) consider that membership in producers’ associations and certifications boost productivity, increase income, and lead to increased training in farm management, more access to agricultural information, and more adequate soil fertility management. This is reflected, in our case, in greater performance (as compared to CCN-51 cocoa farms) in the indicators related to fertilization and pest management, income stability, better producer cooperation, better access to basic services and education, and better knowledge and ecological awareness. However, in our study, lower economic profitability hampers overall economic sustainability for fine flavour cocoa farms, suggesting potential trade-offs between profitability and other key sustainability aspects. In this regard, other studies (Useche & Blare, 2013) show that even if producers are able to obtain a price premium for high-quality cocoa, this may not be enough to compensate for lower production and may be one of the reasons that motivate the expansion of CCN-51 cocoa production. On the other hand, farmers opting for fine flavour cocoa may look for other non-market benefits (Tennhardt et al., 2022; Useche & Blare, 2013) and, despite lower profitability, may benefit from other social and governance-related features that improve sustainability, as promoted by the PRCN.

Finally, our study reveals significant links between capital theory variables and sustainability performance, which are often overlooked (Díaz-Montenegro et al., 2018). The study shows that sustainability performance is lower when the head of household is female. This can be explained by the fact that the role of women in cocoa production has long been undervalued and limited to casual, often unpaid work (Barrientos, 2013). Kuhn et al. (2023) stress that sustainability efforts do not sufficiently address gender discrimination and call for gender-specific training and for increasing women’s access to credit in order to reduce the gender gap. Our study is consistent with these findings and underlines the importance of access to credit and technology, such as the Internet (that is, physical and financial capital), for all cocoa farmers to mitigate structural inequalities and promote sustainability (Barrezueta-Unda & Paz-González, 2018; Salazar et al., 2023).

Our findings also suggest that remoteness (understood as the distance from the farm to the nearest city) has a negative impact on sustainability. However, this relationship is not clear in the literature. Piette (2022) shows that some socioeconomic sustainability indicators (education and farmers’ income) can be negatively affected by remoteness, but other environmental indicators (reduced use of pesticides and agroforestry share) can be positively affected. A broader consensus is found on the relationship between sustainability and being a member of a farmers’ association (social capital) and between sustainability and the presence of pastureland, forests, or wild animals on the farm (natural capital). As already noted, membership in a farmers’ association and diversification of cocoa systems positively influence sustainability outcomes (Amfo et al., 2021; Vogel et al., 2020).

5. Conclusions

This study demonstrates that the sustainability profiles of fine flavour and CCN-51 cocoa farms exhibit pronounced distinctions, particularly within the governance and nutrition and health dimensions, while the profiles of the two types of farms are more similar in the environmental, economic, and social aspects. A closer alignment in sustainability practices has been identified among fine flavourcocoa producers. Furthermore, discernible spatial disparities in sustainability have been observed between fine flavour and CCN-51 cocoa farms. These disparities correlate with several factors, such as: proximity to urban areas; access to tools, the Internet, and credit; farm area dimensions; presence of wildlife; and female participation in associations, which aligns with capital theory.

Producers of both fine flavour and CCN-51 cocoa grapple with a myriad of challenges spanning social, environmental, economic, nutritional, health, and governance dimensions. Addressing these challenges necessitates a holistic, collaborative approach involving concerted efforts from governments, chocolate companies, international organizations, farmers, and local communities. Implementing policies and practices that foster economic equity, multidimensional sustainability, food security, and human rights is imperative for advancing ethical, fair, and sustainable cocoa production.

Our research underscores the pressing need for policy interventions and practices that aim to enhance governance and address health-related aspects in cocoa production. The observed variability between fine flavour and CCN-51 cocoa farms highlights opportunities for targeted improvement. Nevertheless, the applicability of our findings might be constrained by the extent of data collection, signaling a need for expansive studies in varied agricultural settings.

Given the notable disparities in the sustainability profiles of fine flavour and CCN-51 cocoa farms, future research endeavours should probe deeper into the complexities underlying these variations. Further exploration of the repercussions of these disparities on the global cocoa supply chain could provide pivotal insights, aiding the development of more sustainable agricultural practices.

Acknowledgments

The authors thank: the Ministry of Agriculture of Ecuador and the FAO Country Office in Ecuador for their guidance in the specification of the case study; the trained enumerators, who diligently conducted surveys in Manabí; and all the farmers who responded to the surveys.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data used in this research are available upon reasonable request from the corresponding author.

Additional information

Funding

This work was supported by the Food and Agricultural Organization of the United Nations (FAO), under the project FARMTOOLS: ‘Design of farm business optimization tools in the context of economic and environmental crises’ [project number 351488].