Agroecology (AE) is now recognised as a transformative approach to sustainable agriculture, although it has its origins in Indigenous and local practices. This section synthesises the features of AET experiences in the Latin American region. It illustrates historical developments, driving and limiting factors and policy implications to contextualise our study.

Historically, in Latin America, AE is rooted in indigenous farming practices, including varied cropping systems, soil health, and balanced relationships with nature (Altieri and Nicholls, 2017). Nevertheless, the region has been subject to, among others, environmental deterioration and land concentration that have promoted the gradual increase of export-oriented monocultures (Altieri, 2004).

In this paper, AETs are understood holistically to include the multilayered changes needed, more or less simultaneously, for small-scale farmers (Tittonell, 2019) to accomplish shifts in perceptions, attitudes and agricultural production practices. AETs involve transformations at different scales, from individual farmers to their farms, the wider community (including consumers), the surrounding landscape and the institutions associated with the agricultural system (Tittonell, 2019). AE is a viable alternative to conventional agriculture, which is rooted in mono-cropping, the use of chemical fertilisers and pesticides, and oriented towards maximising production and profitability (Dale, 2020). While higher yields and profits are desirable goals for small-scale farmers, conventional agriculture causes much damage, such as the negative effects of chemical fertilisers and pesticides on soil, local fauna and human health (Altieri and Toledo, 2011; Holt-Gimenez et al., 2008).

Socioculturally, conventional agriculture approaches often label local indigenous knowledge ‘backwards’, thus endangering its preservation by aggressively replacing it with technological innovations stemming from the Green Revolution (Altieri, 2004). Moreover, mono-culture (mono-production, mono-knowledge) – aka homogenization – practices make farmers more vulnerable to socio-environmental disruptions and disasters.

According to Tittonell (2019) AET model, AETs occur gradually, but non-linearly, through different stages. Namely, by i) optimization of management practices to increase productive efficiency, ii) inputs substitution, iii) the redesign of the system. A driving force to transition to stages one and two is consumer demand for healthier products that promote the preservation of local knowledge and cultural traditions. Consumers’ cultural beliefs, norms and values, and appreciation for environmental sustainability, play an important role in determining their demand (Tandon et al., 2021). Studies have shown that motivations for consumers’ demand for organic¹ products also include the health benefits attached to their consumption (Pishbahar et al., 2020). Therefore, consumer demand combined with changes in regulations – such as the requirement of specific certifications like the Participatory Guarantee Systems (PGSs)² or the organic labels – can support achieving input substitution³.

An increased demand for organic and AE products in a context where trustworthy relationships between producers and consumers are highly valued has been another incentive for farmers to adopt AE approaches (Kaufmann et al., 2023). Specifically, Kaufmann et al (2023) study describes five examples of Participatory Guarantee Systems (PGS). In Mexico, there are three: Tianguis Orgánico Chapingo, Tianguis Alternativo Tlaxcala, and Tianguis Alternativo El Pochote Xochimilco; in Chile, there is the Ecoferia de la Reina; and in Bolivia, there is the ECO Feria. All are specialised markets that offer food under the PGS system.

Latin American countries count various examples of AET experiences. In Brazil, AE movements succeeded through the emergence and growth of social rural movements – such as the Landless People’s Movement (MST, Movimento dos Trabalhadores Rurais Sem Terra) – promoting more equitable land distribution and sustainable agriculture to counter neoliberal and conventional approaches and to promote food security and sovereignty (Cohn et al., 2006; Petersen et al., 2013). In Argentina, AE practices have been adopted by smallholder farmers and promoted by local movements like the Union of Land Workers (UTT, Unión Trabajadores de la Tierra) in response to economic pressures and environmental concerns, leading to improved soil health and increased biodiversity (Parodi, 2018). In Ecuador, like in other Andean countries, the element of trust has been central to its social economy, which relies on the concept of social capital – as a practice where the relationship between consumers and producers and their transactions mutually influence and reinforce each other (Deaconu et al., 2021). In Guatemala first and in the rest of Central America after, Holt-Giménez (in Cohn et al., 2006) explains, the Campesino a Campesino movement represents a farmer network where knowledge exchange and solidarity against the ills of oppressive systems of production promote the preservation of traditional and sustainable practices. In Mexico, CEDICAM (Centre for Integral Campesino Development of the Mixteca) has centred their work around the role of native seeds and local traditional knowledge to promote sustainable agricultural practices for food sovereignty locally and abroad (Dahl-Bredine, 2006). Similarly, in the Yucatan state of Mexico, the Slow Food movement responded to the damage caused by chemical fertilisers and hybrid seeds. Slow Food supports locally produced food and local knowledge and traditions, yet it lacks government support (Blanco-Gregory et al., 2020). Differently, in Nicaragua, AETs have been supported by state policies, although in the context of a hybrid approach to agricultural production. External funders and local civil society actors have strengthened their networks, within which the National Movement of Agroecological and Organic Producers (MAONIC) has played a central role in the development of AE at the national level and in influencing policy-making. Furthermore, the role of women at the forefront of the movement has made the aspect of gender equity particularly prominent in the region (Christina et al., 2021; Dorrego Carlón and Bottazzi, 2025; Mestmacher and Braun, 2021; Schiller et al., 2020). In Bolivia, where our study was conducted, AE is described as having a ‘long’ and a ‘short’ trajectory (Catacora-Vargas et al., 2017). Within the former, indigenous agricultural methods connected local communities’ livelihoods with their natural surroundings, where ecological conservation and the transmission of social and cultural traditions coexisted with a productive agricultural system (Catacora-Vargas et al., 2017). The latter is akin to the processes that occurred in other countries in the region in response to the environmental and sociocultural degradation resulting from the Green Revolution. The Suka Kollus Program (PROSUKO) – which promotes the preservation of indigenous knowledge – and the AGRECOL Andes Foundation – which supports the creation of local ecological markets (see the section on ‘Context’ for more details about AE in Bolivia) are among the institutions that contributed to the spread of AE in Bolivia.

AE implementation and scaling up in Latin American countries still face various challenges, despite gradual steps over time. Limited access to land, resources, and technical support (Rosset, 2008) are among the main factors that hinder the spread of AE practices among small-scale farmers.

AE’s scalability and its wider spread are hampered by state policies promoting the interests of agribusiness and agro-exports and restricting funding for AE-related research and extension services (Giraldo and McCune, 2019). AETs require a policy framework and institutional arrangements that facilitate them (Al-Kaisi and Lal, 2020). Agrarian reforms to redistribute land and resource ownership can contribute to empowering smallholder farmers and supporting them in implementing AE practices (Altieri et al., 2012). Additionally, investing in participatory and farmer-led research and extension services, together with incentives to promote AE products in markets, can strengthen the growth of AE farming approaches (Altieri and Toledo, 2011).

Important developments have taken place to advance AE across the Latin American region.

However, issues related to socio-economic inequities, institutional shortcomings and environmental degradation associated with dominant conventional approaches remain to be addressed (Zimerman, 2024). To further food sovereignty and more sustainable food systems, Latin American countries require joint efforts across multiple stakeholders, including researchers, grassroots and activist organisations and policy-makers.

The project and study that this paper draws on were conducted in Cochabamba, Bolivia (Figure 1). Cochabamba is a medium-sized city with 2 million inhabitants⁴. It is located in an inter-Andean valley, primarily focused on dairy farming and agriculture. The most commonly produced crops are potatoes and other tubers in the highlands, and corn and vegetables in the valley. Summers are short, hot, and mostly cloudy; winters are short, cool, and mostly clear. Throughout the year, temperatures generally range from 4 °C to 26 °C. The climate is fairly dry year-round, with an annual rainfall of 300 mm, concentrated between November and March (Navarro and Maldonado, 2002). The valley’s production is sold at farmers’ markets and popular fairs, from where it is also distributed to the rest of the country through transporters and wholesalers. Popular fairs are peri-urban spaces where producers arrive directly to sell to wholesalers in the early morning and to the final consumer throughout the day. This is why prices are affordable and they are popular with middle- and low-income families. AE food prices are sold in 80% of cases at the same price or lower than conventional food (Alem et al., 2018).

After an exploration of several fairs in the region based on specific criteria for selection – namely, existing permanent AE offers and sales stalls, and farmers’ willingness to be involved in the project – seven popular fairs were selected. We focused our work on local fairs with the aim of strengthening the democratisation of access to AE foods for low- and middle-income sections of the population, countering the notion that such foods are only accessible to people or families with high incomes. The criteria for selecting these fairs were: that they be established fairs within a specific geographic area (Kanata Metropolitan Region), that they host AE producers with a permanent supply, and that the consumers correspond to a low- to middle-income level. These included: Quillacollo, El Playón, and Apote in the Municipality of Quillacollo; Loa in the Municipality of Tiquipaya; Villa Obrajes in the Municipality of Sacaba; and Las Rieles and Primero de Mayo in the Municipality of Cercado. However, the project worked in five of the seven popular fairs (Quillacollo, El Playón, Apote, Loa and Villa Obrajes). In Las Rieles fair, no AE farmers with a permanent offer at the fair were found. In the fair Primero de Mayo, the AE farmers’ group already had other institutional support so they did not work with us (Figure 1). The objective was to strengthen the AE offer in five popular fairs through the exchange of experiences among farmers and AE training events.

Training events’ topics included AE pest management, bio-inputs’ preparation and use, sales tips, food’s nutritional components and effects of the use and consumption of pesticides. The study also tracked changing agricultural practices, motivations and AETs during the intervention.

In Bolivia, particularly in the Interandean Valleys and the Altiplano, AE production was promoted through training provided by institutions and municipal governments, with two fundamental objectives: to (i) generate commercial alternatives for agricultural products, and (ii) guarantee the food sovereignty of farmer families in a context of climate change. AE production can contribute to accessing and consuming a diversity of healthy foods and promoting better nutrition, as income generation is directly related to market access (Deaconu et al., 2019).

The National Agricultural Census (2013) indicates that 95% of the agricultural production units(APU) correspond to family farming and that AE farming is carried out mainly by family farming andhunter-gatherers (Instituto Nacional de Estadı́stica (INE) Bolivia, 2016; MDRyT and Fundación AGRECOL Andes, 2018). Therefore, AE farming is implemented mainly by small farmers and, within these, it is mostly in the hands of women (Llanque et al., 2018; Lindemann et al., 2024). The majority of them are unpaid workers, which puts them at a great disadvantage compared to men, although there is a slight increase in women classified as self-employed (Tito-Velarde, 2021). Rural women invest more effort in generating their economic income, which they complement by selling their labour force as agricultural labourers, so often their main obstacle is a lack of time (Llanque et al., 2018).

Although much progress has been made in policies and regulations on paper since 2006, in reality, deforestation has increased, and the relative value of small products has fallen compared to agro-industrial ones. Likewise, successful AE-to-organic production cases are fractional, reflecting the difficulties of scaling AE in Bolivia (Terán, 2020). Finally, a study in Cochabamba reported 6,200 farmers with organic production, who contribute only to 2,1% of the food production there (26.050 tn), and only 0.02% of their production had the accreditation of the PGS (Alem et al., 2018; Fundación Agrecol Andes, 2018). This shows that the regulatory progress since 2006 has not induced an increase in AE production with PGS in Cochabamba.

This study used a qualitative approach, including semi-structured interviews and observations in combination with a classification rubric used to identify categories of farmers that apply pesticides (coined “Conventional”; 100% reliance on these inputs), partially apply pesticides (coined “Mixed”; less than 100% of reliance on these inputs), and not apply pesticides (coined “Agroecological”, 0% reliance on these inputs). We aggregated farmers into these three groups to better understand their motivations, AE practices and AET processes.

Notably, this study did not employ the Tool for Agroecology Performance Evaluation (TAPE) because it focuses on the performance of various criteria of agricultural and food systems (Mottet et al., 2020). Differently, this investigation explored farmers’ perceptions and experiences in their AET process to understand how and what local factors motivate or hinder them. In doing so, we connected the experiences shared by participants with Tittonell’s steps to AETs (Tittonell, 2019) outlined in section 1. The focus was on comprehending the nature of the AET processes taking place without rigorously assessing their AE performance in the way that the TAPE tool does.

This approach was in line with our research interest in learning from local experiences how to improve AET interventions in the context.

We present the results in 3 main sections: (i) Typology of farmers in the popular fairs, (ii) Typologies’ motivations and practices and (iii) Factors that promote or hinder AETs.

This study combines the experiences of conducting AE training and evaluations with farmers in five fairs in Cochabamba, Bolivia, over four years. These activities and interactions unravelled compelling learnings about the farmer typologies we encountered and the factors that influence their perceptions and practices. These learnings bear implications for approaches to AETs in Cochabamba and similar contexts.

The limited behavioural change despite knowledge of health risks is consistent with loss aversion and probability weighting. Farmers may overweight the small but salient probability of a severe pest shock, which keeps them anchored to a pesticide-based input bundle that feels safer than experimenting with AE practices (Kahneman and Tversky, 1979; Kemeze et al., 2020).

As presented in the results, the three main farmer typologies identified were i) conventional, ii) mixed and iii) AE. We found that their motivations for and factors affecting their production practices play a significant role in determining their typology and position within AE transitions.

Conventional farmers are primarily motivated by risk-aversion towards production and income generation. Several studies have shown that risk-averse farmers are more likely to use pesticides as a strategy to stabilise yields and secure income (Carpentier, 2017; Chèze et al., 2020; Pan et al., 2020; Su et al., 2022; Cai et al., 2024). Curiously, AE knowledge and awareness of the risks associated with the use of agrochemical pesticides and fertilisers for human health and soil quality do not appear to strongly influence their practices and, in some cases, their perceptions (Finger et al., 2022). At the same time, the adoption of certain AE practices is frequently rooted in inherited family traditions rather than in formal training to preserve soil fertility. This pattern is not limited to conventional producers, but is also evident among producers in the three categories, indicating that the transmission of agricultural knowledge through family networks plays a significant role across different production systems.

For conventional producers, however, family tradition is combined with the fact that their plot size is usually larger due to the prevalence of mono- and market-oriented crops, which determines the kind of AE practices implemented. Conventional farmers do implement AE management practices as suggested by Tittonell (2019) and other practices that need to be adjusted to facilitate AE transitions. This can have interesting implications for how AE innovations are designed and co-created, where acknowledging and taking advantage of the important role played by local traditions may positively influence the adoption of AE practices even among conventional farmers.

For mixed farmers, their risk-aversion and economic motivations are expressed through the use of chemicals when there is a higher danger of pests and diseases, and to ensure high production of commercial crops. At the same time, their plots being smaller than those of conventional farmers and including a priority for family consumption motivates them to implement more – but also different – AE practices that are inherited from family tradition and learnt from AE training. Different from conventional farmers who used AE practices due solely to family traditions and for better production, mixed farmers implement AE practices also for soil and human health reasons. About the AE transition model outlined above, the mixed farmer typology appears to be progressing towards input substitution. However, to further transform behaviours towards AE approaches, perceptions of risk-aversion remain to be addressed, as noted by Kishi (2002) in Indonesia, by demonstrating to farmers that reduced use of chemical pesticides does not equal reduced income (Kishi, 2002, p. 175). A study conducted in China is critical to our results. It concluded that factors related to perceptions of risks to profit maximisation contributed to decreasing farmers’ pesticide use by demonstrating both the health and financial damages of excess use of pesticides (Pan et al., 2020).

As noted in the results, overall, AE farmers implemented all the AE practices included in the study. Their main motivation lies in the prioritisation of family health, while the factors that influence choices related to their production methods are that they are not concerned about the risks associated with commercial production, as agricultural production is not their main source of income and they have small-size plots, which are easily manageable using AE methods. Additionally, within this group, the role of women emerged as central for the implementation of AE practices, largely due to their prioritisation of family health. AE practices provided a means to enhance women’s empowerment within the household by offering healthier food, time with children, and expendable income. At the same time, this central role could increase their workload and limit opportunities outside the household, illustrating the ambivalent effects of empowerment in the context of entrenched gendered responsibilities. The importance of the role of women in relation, for instance, to reduced pesticide use was found among Iranian female farmers, too, as compared to male farmers (Abdollahzadeh et al., 2016). Daza et al. (2020) state that in the development of AE in Ecuador, the role of women has been fundamental because they are in charge of caring for the production plots, as well as for the responsibilities of the home, tasks which primarily benefit the family environment since they reinvest the profits from their sales at fairs in improving the production system, as well as in the health and education of their children. Perez and Izurieta (2022) add that organisations of AE women farmers in Ecuador and Mexico have an important role in caring not only for the family and its individuals but also for the space they inhabit, the soil, water and air, as hopeful spaces that also have the strength to expand to society as a whole.

Our results showed that different percentages of women dedicated their time to agricultural activities depending on the category of farmers. Fewer women were involved in production activities in conventional farmers when compared to AE farmers. This trend also occurred when we compared the mixed farmer category and the AE category. This shows that when farming systems are more agroecological, at least from the farmers engaged in this study, the role of women is more prominent. Our study also revealed that when referring to commercialisation, the role of women is also more prominent regardless of the group of farmers. Future studies should focus on designing studies of gender to understand power asymmetries within the different systems.

Scientists and practitioners should pay more attention to addressing the gender imbalances identified in production and commercialisation. It is essential to push for gender-sensitive policies in AE that guarantee equitable access to subsidies, training opportunities, and research initiatives that fully recognise and value women’s contributions. Such policies should not only improve women’s access to productive resources but also strengthen their leadership and decision-making capacity within AE systems. At the same time, promoting the integration of care economy perspectives is key to acknowledging and alleviating women’s unpaid care roles. Initiatives such as time banks, childcare services at markets, and community-based support systems can help redistribute care responsibilities, reduce women’s workload, and enable greater participation in both production and commercialisation activities.

AE farmers encompass those with PGS accreditation and self-declared ones, and the practices they have in common are those related to family tradition, which plays an important role and influences their productive practices. This indicates that, in this group too, family traditions play an influential role. Moreover, this group also mentioned the importance of consumer demand in driving their production choices. Hence, this farmer group with their characteristics (i.e., PGS accreditation for some, self-declared for others, and the role of consumer demand) presents a position closer to the stage of input substitution in AE transitions (Tittonell, 2019). However, without institutional and policy support, this is only possible when AE production is not their main source of income, and it is unlikely to drive AE transitions at larger scales. This behaviour also aligns with livelihood vulnerability frameworks: where formal insurance, credit, or price support are weak, households rationally prioritise avoiding losses. Institutional gaps therefore, amplify loss aversion and contribute to the observed ‘pesticide lock-in’ (Scoones, 1998).

Blanco-Gregory et al (2020) found that greater assistance to producers was given by NGOs compared to the support of public authorities in the Yucatán region of Mexico for the promotion of AE ventures, such as facilitating distribution and sales channels. Similarly, farmers in our study feel abandoned by the state, and some of them self-organise to buy sales spaces with their funds without any state support to promote their products, sales spaces, or their sales themselves, apart from the assistance they receive from NGOs and the state university. This provision is not significant and, therefore, not enough. Additionally, NGOs’ support is mostly directed at farmers with a PGS certification, but not at self-declared AE farmers, who are even more forgotten. In this regard, a study in China highlighted that using market incentives could help reduce the use of chemical pesticides by small farmers better than government regulations, such as higher prices for food produced without the use of pesticides, more stable marketing channels and markets in the medium and long term, and better information on consumer demand and the prices paid for safer products (Zhao et al., 2017). Another study in Ecuador concluded that strengthening the participation of small AE farmers in sales markets, as well as strengthening their organisations, are important factors in promoting the AETs (Borja et al., 2024).

Regarding the motivations for applying synthetic pesticides, Robinson et al. (2007) classify this practice into those who do not use pesticides, those who use them only before the pest attack, those who use them only after the pest attack and those who use them before and after. This classification goes hand in hand with the typologies in our study. Those who do not use pesticides are the AE ones, those who use them sometimes before or after are the mixed ones, and those who always use them before and after are the conventional farmers. Additionally, the factors that predispose them to use or not use pesticides are risk aversion, trying to minimise losses with their application, and the availability of cash. If a farmer has their cash or credit, it is expected that they would apply pesticides without hesitation. For Robinson et al. (2007), another factor is training. The more pest management training they receive, the more receptive they are to AE production practices. Different from our study, the typologies proposed by Robinson et al. (2007) highlight training as a fundamental factor for the farmers’ receptivity to AE. In our study, instead, AE knowledge increase did not always result in the adoption of more AE practices. In self-declared AE farmers, this may be because they are dedicated to the production of diverse vegetables from different, generally short, cycles. For them, the main barriers are a lack of time, permanent technical support and availability of inputs or materials required to make preparations. As Levins (2006) remarked, agricultural producers’ decisions depend on the purpose of production and, in the case of production for consumption, diversity and quality are prioritised. Concurrently, though, land productivity and workforce availability determine the size of the cultivated plot (Levins et al., 2006).

Another lesson in our findings is how chemical pesticide removal can be an entry point for AE transitions, by engaging in co-learning processes with farmers that include local traditions and context-specific characteristics (Belmain et al., 2022). Particularly, our results show that farmers with lower levels of risk-aversion and whose motivations (i.e., what they value) are focused on health are more likely to undertake AE practices.

The discussion above reveals that the farmer typologies in our study tend to focus on aspects of the AE transition process related to reducing the effects of chemicals in their production and implementing input substitution practices. A combination of their economic motivations, risk aversion, and external factors related to a lack of institutional support hinders their AETs from including aspects of systems redesign.

The results of this study suggest that it is imperative to consider context-specific and complex aspects for successful AE interventions. As shown in the study by Muleme et al. (2017) in Uganda, providing knowledge training can help influence perceptions and attitudes. However, that is not sufficient to change practices related, for instance, to pesticide use. Akin to our participants, Ugandan farmers’ risk-aversion towards economic losses was often stronger than their knowledge of the hazards of pesticide use and their family’s natural traditional methods. Similarly, the lack of contextually appropriate practices and consideration for farmers’ priorities and socio-economic situation caused the failure of IPM initiatives in Cambodia (Dunn et al., 2023). Furthermore, Dunn et al. (2023) identified a phenomenon of ‘pesticide lock-in’ as a form of pesticide dependency that Cambodian producers had developed, whereby their risk-aversion played such a significant role in determining their behaviour that they would not reduce pesticide use despite possessing adequate knowledge of their risks as well as of potential alternative natural pest control practices. An analogous occurrence was found in smallholder farmers in Kenya, who held views that biopesticides were not as effective as chemical ones (Constantine et al., 2020).

It is also crucial to acknowledge that all farmers have practices, and we need to understand how to integrate new ones into their management. Equally crucial is to find ways to communicate our results to non-converted farmers (clearly shown in Figure 3) and what the limitations would be around them.

Additionally, the role of women can be central in facilitating AE practices’ adoption and contributing to shifting values within households otherwise more oriented towards conventional production (Mestmacher and Braun, 2021). It is important for women to not merely play an instrumental role, but to become empowered in the process, such as the examples in our results where women had agency over their income (Borja et al., 2024; FAO, 2018; Zaremba et al., 2021).

Finally, our investigation calls attention to the importance of understanding how the convergence of all the factors above, together with the necessary enabling conditions for farmers, yields dissonant responses in farmers’ actions in AE transition processes. These aspects will need to be further explored through approaches that combine interventions and research, where the learnings can be used to support AETs.

Overall, the persistence of chemical use can be seen as the result of loss-averse decision heuristics operating under conditions of institutional vulnerability, reinforcing the need for policies that buffer downside risk if AE adoption is to expand (Scoones, 1998; Pan et al., 2020).